Content uploaded by Kevin McDonough
Author content
All content in this area was uploaded by Kevin McDonough on Jan 30, 2019
Content may be subject to copyright.
Can the Liberal State Support Cultural Identity Schools?
Author(s): Kevin McDonough
Source:
American Journal of Education,
Vol. 106, No. 4 (Aug., 1998), pp. 463-499
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/1085625
Accessed: 11/01/2010 11:19
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to
American Journal of Education.
http://www.jstor.org
Can the Liberal State Support
Cultural
Identity Schools?
KEVIN MCDONOUGH
McGill University
Recently, several groups in the United States and Canada-for example,
American Indian/First Nations people, African-American males, and the
deaf-have claimed the right to receive state support for cultural identity
schools-that is, separate schools whose educational aims and practices
are designed to reinforce a particular cultural identity. It is widely assumed
that liberalism must be committed to a principle of cultural neutrality that
prevents a liberal state from assigning legitimacy to such demands. This
article provides a close examination of the ethical principles a liberal state
may adduce in making political judgments about such matters. First, two
dominant perspectives that have emerged recently within liberal politi-
cal and educational theory are developed and critically evaluated and
their educational implications examined. Specifically, a distinction be-
tween "strong" and "moderate" cultural identity schools is identified, and
it is argued that a liberal state may legitimately support the latter but not
the former. I conclude by considering several contextual factors a liberal
state may have to consider in determining the legitimacy of specific de-
mands for moderate cultural identity schools, especially demands made by
disadvantaged minority cultural groups.
Introduction
Recently, several groups in the United States and Canada (e.g., American
Indian/First Nations people, African-American males, and the deaf)
have claimed the right to receive state support for separate schools
whose educational aims and practices (curricula, admission and hiring
policies, teaching methods, and so on) are designed to reinforce and
affirm a particular cultural identity.' For children from such groups,
the task of forging an identity must frequently be undertaken within a
AmericanJournal
of
Education 106 (August
1998)
? 1998
by
The University
of Chicago.
All
rights
reserved.
0195-6744/98/0604-0001$02.00
August 1998 463
__
Can the Liberal State Support Cultural Identity Schools
?
context where the value, and the very survival, of ancestral traditions is
seriously threatened if not altogether eclipsed by the hegemony of ways
of life developed by others. Furthermore, a considerable body of socio-
logical and anthropological evidence suggests that existing public com-
mon schools, whose aims and practices must (in principle, at least) ac-
commodate diverse cultural backgrounds and traditions, are places in
which the task of forging an identity is likely to be a seriously debilitating
one for many children from minority cultural groups.2 The demands of
parents for separate schools can be understood as a partial response to
this depressing, and often desperate, situation.
These separate schools, which I call "cultural identity schools," have
two salient characteristics. First, they seek to restrict membership to in-
dividuals from a particular cultural, racial, ethnic, national, or linguistic
group. Second, they seek to advance a separate education affirming
and reinforcing the identity of the group in question. Cultural identity
schools may also take two distinct forms depending on the scope and
duration of the separate education they seek to advance. I call these
"strong" and "moderate" cultural identity schools, respectively. Strong
cultural identity schools characteristically seek to foster a separate edu-
cation of extensive scope and duration that is meant to ensure that chil-
dren adhere to a distinct ancestral identity throughout their lives. At a
more advanced level, this education would enable individuals to evaluate
the group identity and defend it against competing cultural identities.
Moderate cultural identity schools characteristically
seek to foster a sepa-
rate education of more limited scope and duration that is meant to offer
young children an initial, stable conception of the good. But, in contrast
with strong cultural identity schools, this initial conception of the good
serves as a starting point for autonomous ethical reflection, which en-
ables children to revise their inherited cultural identity and even reject
it in favor of alternative identities. In this article, I consider what politi-
coethical principles a liberal democratic state may invoke in order to
evaluate group claims to state support for cultural identity schools. Ulti-
mately, I argue that these principles prohibit state support for strong cul-
tural identity schools, while generating a justification for moderate cul-
tural identity schools under certain circumstances.
KEVIN MCDONOUGH is assistant professor of philosophy of education
in the Department of Culture and Values in Education at McGill Univer-
sity. He has published articles on moral education in the Journal of Phi-
losophy of Education and Studies in Philosophy and Education.
464 American Journal of Education
McDonough
The remainder of the article is divided into three main sections. In the
first section, I outline how the central issues in this article converge with
issues of pressing interest in recent nonphilosophical, scholarly discus-
sions of multicultural education. The latter discussions frequently argue
that liberalism faces formidable obstacles in providing an adequate jus-
tification for multicultural education policies and programs in general
and for group claims to cultural identity schools in particular. However,
the version of liberalism against which such charges are leveled is philo-
sophically obscure. Thus, my discussion in this section also seeks to illu-
minate a philosophically coherent and influential version of liberal ide-
ology that proponents of multicultural education have targeted. In this
first section, I also briefly outline a powerful liberal response to these
criticisms that suggests that the obstacles liberalism is supposed to face
in respecting and recognizing particular cultural identities are less for-
midable than is often assumed. This points the way to the second section
of the article, which discusses and evaluates recent philosophical at-
tempts to remedy liberalism's past failings with respect to cultural, and
hence educational, diversity. In this section, I first outline and criticize
an influential set of arguments, which I refer to as the cultural recogni-
tion thesis, that purport to provide a liberal foundation for justifying
group rights, including a right to state support for cultural identity
schools. Second, I examine a powerful alternative to this view, which I
label the cosmopolitan perspective, which rejects the legitimacy of any
forms of state support for cultural identity schooling in a liberal state.
Through examining these two competing perspectives of multicultural
education in the liberal state, I develop and defend a middle ground
between the two that generates ajustification for moderate cultural iden-
tity schools.
It is important at this point to distinguish between the task of clarifying
and evaluating a philosophical justification for separate schools, which is
the task I focus on in the first two sections of the article, and the task of
applying this
justification in light of the real or probable effects of actual
cultural identity schools, which I discuss in the third section. Both tasks
are important for determining the legitimacy of claims to state support
for cultural identity schools. Thus, I take seriously T. H. McClaughlin's
point that
philosophical considerations alone cannot settle whether, and in
what way, [separate schools] should be established. Many other is-
sues (for example, of a practical, pedagogic, psychological, socio-
logical, political and demographic kind) need to be considered in a
full assessment of practical educational policies. Ethical and philo-
sophical reflection must be conducted in relation to this fuller
August 1998 465
Can the Liberal State
Support
Cultural
Identity
Schools
?
range of complex considerations and not in an abstract way inde-
pendent of them. It is rash, for example, to condone or condemn
certain kinds of separate school on grounds of philosophical prin-
ciple alone. Much depends on how the institutions actually operate,
and what their effects actually are on students and the broader com-
munity. (McClaughlin 1992, p. 115)
Nevertheless, it is important to add a proviso to McClaughlin's point. A
clear understanding of the philosophical principles that are supposed to
underlie any justification for cultural identity schools must still be logi-
cally prior to an assessment of empirical effects and operations. Other-
wise, there would be no nonarbitrary and noncapricious way to assess
and evaluate the relative ethical weight and significance we ought to
assign to the multiple, complex, and conflicting effects that separate
schools will have on students and the broader community. So, it is to an
examination of these principles that I now turn.
Liberal Neutrality and Multicultural Education
Michael Olneck's (1993) interesting analysis of recent scholarly litera-
ture on multicultural education argues that the aspirations of a wide
range of recent multicultural education programs-including bilingual
education, multicultural curricula, and community control movements
designed to assign greater educational autonomy to local cultural, racial,
linguistic and ethnic communities-have become severely distorted, wa-
tered down, and misshapen in practice. Olneck's criticisms are leveled
not merely at the inadequacies of a hodgepodge of multicultural educa-
tion programs. Ultimately, they are directed at the hegemonic principles
of liberal equality that underwrite the political legitimacy of these pro-
grams and that frame the discourse of public policy within which they
become educational reality (Olneck 1993, pp. 234-36). Nevertheless,
the conception of liberal equality against which Olneck directs his criti-
cisms is philosophically obscure. This makes it difficult to determine pre-
cisely what obstacles are supposed to stand between liberalism and a
more adequate response to the aspirations of multicultural education.
Fortunately, these obstacles can be illuminated by outlining a philosophi-
cally sharper interpretation of liberal equality than the one Olneck him-
self provides.
A number of political philosophers have recently criticized contem-
porary liberal theory on the grounds that it fails to adequately account
for, or recognize, the value of cultural identity (Kymlicka 1989a, 1989b,
1995; Tamir 1995; Taylor 1994). Furthermore, according to Will Kym-
466 American Journal of Education
McDonough
licka, liberalism's failure in this respect can be attributed to a misguided
commitment on the part of its most prominent sponsors to a doctrine of
cultural neutrality. According to this doctrine, the state must treat spe-
cific minority cultural groups with nothing more than "benign neglect."
As Kymlicka puts it, cultural neutrality is the view the state must "not
oppose the freedom of people to express their particular cultural attach-
ments, but nor does it [or is it permitted to] nurture such expression"
(Kymlicka 1995, p. 3). Something very much like this notion of cultural
neutrality is the fundamental target of Olneck's (1993) criticisms of U.S.
multicultural education policy and practice. For example, he points out
that state support for bilingual education programs, Afrocentric schools,
or all-male schools for African-American children, may be justified or not
"on the basis of [their] capacity to raise test scores and grades, and to
reduce dropout rates" (Olneck 1993, pp. 249-50). Similarly, Olneck
suggests that "liberal ideology's" commitment to cultural neutrality also
explains the federal government's unwillingness to endorse or support
bilingual education programs that seek to reinforce and maintain non-
English-speaking children's native language and thus go beyond merely
serving as a "temporary and transitional" bridge to English-only class-
rooms (Olneck 1993, pp. 237-39). Although Olneck does not discuss
them, a liberal conception of "culture-neutral" equality would forbid the
state from supporting cultural identity schools. For to do so would be to
rely on criteria that assign privileged, nonneutral status to a group's lin-
guistic, racial, or cultural identity.
The force of Olneck's criticisms depends on a liberalism that is com-
mitted to a doctrine of cultural neutrality. However, philosophical com-
mentators such as Kymlicka, Yael Tamir, and Charles Taylor argue that
liberalism need not entail a commitment to the doctrine of cultural neu-
trality
even though many leading contemporary exponents such asJohn
Rawls (1971) and Ronald Dworkin (1978) have failed to recognize this.
To the extent that liberalism can be divorced from a doctrine of cultural
neutrality, then, at least one commonly assumed obstacle to a liberal jus-
tification of cultural identity schools can be removed. But the task of
disengaging liberalism from cultural neutrality requires a slightly fuller
outline of what values constitute the core of liberal ideology.
The difficulties illuminating such a core are notorious since, asJeremy
Waldron writes: "If we examine the range of doctrines or principles that
are classified [as liberal] ... we are unlikely to find any set of doctrines
or principles that are held in common by all of them, any single cluster
of theoretical or practical propositions that might be regarded as the core
or the essence
of the ideology in question" (1993, p. 36). Nevertheless, a
useful starting point can be found in one especially prominent interpre-
tation of the liberal tradition-an interpretation whose advocates
August 1998 467
Can the Liberal State
Support
Cultural
Identity
Schools?
include Immanuel Kant,
John Stuart Mill, andJohn Dewey.3
This strand
of liberal political thought privileges a moral ideal of personal autonomy
as the source of political legitimacy. This ideal of autonomy requires that
individuals develop and exercise the kinds of capacities required for
rigorous self-examination that enables them to evaluate, revise, and
perhaps even reject inherited cultural roles, commitments, loyalties,
traditions, values, and beliefs in the light of meaningful alternatives.
As Eamonn Callan points out, this conception of liberal autonomy re-
quires substantial limits on diversity
in actual liberal societies since it "ac-
commodates diversity only so far as diversity results from the exercise
of ideals of autonomy or individuality regarded as constitutive of the
good life and politically privileged in the institutions of a free society"
(1995, p. 260). Thus, an autonomy-based liberalism allows considerable,
though not unlimited, room for educational diversity. There is nothing
in it to preclude, for example, state support for separate schools so long
as those institutions can be shown to support the kind of rigorous self-
examination that personal autonomy requires. Similarly, there is no
reason that an autonomy-based liberalism must forbid the state from
endorsing educational policies intended to enable separate schools to
flourish-for example, judiciously designed school choice or voucher
programs and charter schools. However, according to autonomy-based
liberalism a liberal state must advance the ideal of autonomy through
its political and educational institutions, even if doing so means that
some existing cultural communities-for example, those whose cultural
and ethical traditions marginalize or preclude personal autonomy alto-
gether-must revise their own educational practices to accommodate
the liberal ideal of personal autonomy. Is a thoroughgoing commitment
to cultural neutrality necessary in order to protect liberalism's prior com-
mitment to ensuring the political and educational conditions whereby
all citizens can develop and exercise the capacity for personal autonomy?
Kymlicka argues that the answer to this question must be "no" since
the principle of cultural neutrality undermines rather than supports the
liberal commitment to individual autonomy (1989b, p. 903). According
to Kymlicka, individuals require a secure and rich cultural structure in
order to develop and exercise their capacity for autonomy-a cultural
structure that most of us inherit, not as members of a common political
community, but as members of particular cultural subcommunities. As
Kymlicka says, "We decide how to live our lives by situating ourselves
in ... cultural narratives, by adopting roles that have struck us as worth-
while ones, as ones worth living" (1989a, p. 165). But the range of cul-
tural narratives, roles, and options from which we choose is given to us
by the cultural structure within which we are raised and educated. Fur-
thermore, there is a multiplicity of such cultural structures in any liberal
468 American Journal of Education
McDonough
society-some of whose survival and flourishing may be threatened.
When one's inherited cultural structure is threatened, so is the value of
individual autonomy for members of the threatened group. According
to Kymlicka, it is this strong connection between culture and personal
autonomy that requires liberalism to sever its ties with the doctrine of
cultural neutrality.
As Kymlicka says, "Once we recognize the importance
of the cultural structure [something that liberals such as Rawls
and Dwor-
kin have failed to do] and accept that there is a positive duty on the state
to protect the cultural conditions that allow for autonomous choice,
then cultural membership does have political salience. Respect for the
autonomy of the members of minority cultures requires respect for their
cultural structure, and that in turn may require special linguistic, educa-
tional, and even political rights for minority cultures" (Kymlicka 1989b,
p. 903).
The main conclusion at this point is that liberal attempts to justify
group rights need not be blocked by a thoroughgoing commitment to
cultural neutrality. The positive liberal arguments that can be adduced
in support of special group "cultural rights" will be elucidated below.
However, it is already obvious why a liberal state cannot afford cultural
rights, including rights to cultural identity schools, for all cultural groups
that demand them. It can only assign such rights to groups whose cul-
tural identity is hospitable to the rigorous self-reflection that personal
autonomy demands. But not all cultures support autonomous reflection
by individual members. It follows that any liberal attempt to justify cul-
tural rights must be capable of distinguishing autonomy-supportive from
autonomy-corrosive cultures. In the next section, I argue that it is pre-
cisely the difficulty of making this distinction that undermines recent
liberal attempts to justify cultural rights. Furthermore, I argue that while
this problem precludes a liberal justification for strong cultural identity
schools, it does not preclude ajustification for moderate ones.
Cultural Recognition versus Cosmopolitanism
The Cultural Recognition Thesis
In the past few years, some of the most influential and sophisticated con-
temporary political philosophers have challenged the common assump-
tion that a liberal state committed to individual freedom cannot provide
special recognition or support to particular group cultural identities.
Indeed, they have argued, liberalism's commitment to individual au-
tonomy requires
a commitment to special group rights for minority cul-
tures (Kymlicka 1989a, 1995; Tamir 1995; Taylor 1994). In this section, I
August 1998 469
Can the
Liberal State
Support
Cultural
Identity
Schools ?
reconstruct and evaluate the most interesting and sophisticated version
of this argument. In doing so, I identify and evaluate a common justifi-
cation for cultural rights that underlies the various specific theories de-
veloped by particular theorists. For rhetorical convenience, I call this
common justification the cultural recognition thesis. My discussion will
focus on a close analysis of Kymlicka's position, while, for the most part,
confining points of comparison and contrast between Kymlicka, Taylor,
and Tamir to several discursive footnotes.4 This way of proceeding has
two advantages. First, Kymlicka's
is the most systematically and fully de-
veloped position, and therefore the one in which the cultural recogni-
tion thesis is most easily discerned and most rigorously defended. Sec-
ond, concentrating on the views of a single author makes it easier to
elucidate the justificatory structure that must underlie all attempts tojus-
tify cultural rights from a liberal perspective.5
The cultural recognition thesis consists of two main threads. First,
membership in a distinct cultural structure is necessary to enable indi-
viduals to develop and exercise their capacity to make meaningful and
autonomous choices about how to lead their lives. According to the cul-
tural recognition thesis, it is one's own cultural identity that is fundamen-
tal to one's sense of self and the capacity to make autonomous choices;
and so it is this cultural identity, and not just any cultural identity that
happens to be available, that requires recognition. Second, since there
is a multiplicity of cultural structures in liberal democratic societies,
members of each of these cultures have at least a prima facie claim to
cultural recognition in the form of cultural rights.
Kymlicka develops the first strand when he claims that cultural struc-
tures are valuable only to the extent that they provide the necessary foun-
dation for securing the conditions of autonomy and self-respect: "Liber-
als should be concerned with the fate of cultural structures, not because
they have some moral status of their own, but because it's only through
having a rich and secure cultural structure that people can become
aware, in a vivid way, of the options available to them, and intelligently
examine their value" (1989a, p. 165).6 Furthermore, according to Kym-
licka, there is a close relationship between the capacity for personal au-
tonomy and the value of self-respect.7
Following Rawls (1971), Kymlicka
argues that individuals have a highest-order interest in being able to de-
velop the capacity for revising and altering their cultural affiliations and
commitments because their existing commitments may turn out to be
worthless, or because commitments that were previously worthwhile may
come to lose their worth. Thus, in order to secure the conditions of self-
respect, individuals require the capacity to revise their existing commit-
ments when these come to lose their worth. As Kymlicka
points out, we
can only judge the worth of our various commitments by examining the
470 American Journal of Education
McDonough
range of options available to us within the cultural structure we happen
to inhabit (1989a, p. 165). In short, the cultural recognition thesis sup-
ports claims to recognition by minority cultural groups on the grounds
that such recognition is necessary in order to support the values of self-
respect and personal autonomy.8
Obviously, proponents of the cultural recognition thesis must regard
the educational aim of promoting autonomous development as essential
to any morally adequate education. Furthermore, the cultural structures
that Kymlicka,
Tamir, and Taylor have in mind must themselves be mul-
ticultural since, otherwise, the range of cultural options would be too
meager to provide the conditions for individuals to develop and under-
take the rigorous sort of self-examination that underpins a genuine ca-
pacity for personal autonomy. It follows that the cultural recognition
thesis can, at most, support only a narrowly circumscribed right to state-
supported cultural identity schools. Specifically, such a right must be
confined to cultural groups whose educational aims seek to foster initia-
tion into cultural traditions that are themselves already extensively lib-
eralized and therefore capable of supporting autonomous reflection.
Proponents of the cultural recognition thesis might not deny this
conclusion, acknowledging that cultural rights, including the right to
cultural identity schools, cannot justify educational practices designed
to foster a monocultural identity that would undermine autonomy.
Throughout his work, Kymlicka
stresses that maintaining a culture does
not mean maintaining it as static. In addition, he stresses that supporting
cultural survival does not and cannot mean ensuring that such cultures
are "isolated and impervious to external influences." Indeed, Kymlicka
explicitly endorses the process of cultural interaction and interchange as
"an opportunity for enrichment" (1995, pp. 101-2).9 Nevertheless, such
an endorsement raises a deep problem for proponents of the cultural
recognition thesis.
As Taylor points out, demands for state-supported measures designed
to ensure survival are usually demands for measures that will enable
members of such groups to "maintain and cherish distinctness, not just
now but forever," so that the group identity "may never be lost" (1994,
p. 40).10
Kymlicka, too, believes that individual identity is so strongly at-
tached to cultural membership that to lose it is to threaten the very basis
of self-respect for individual members, except in rare and extreme cases.
Thus, Kymlicka, like Taylor, views cultural rights as designed to ensure
the permanent survival of liberalized cultures (Kymlicka 1995, pp. 84-
90). However, the problem arises precisely at this point since it is far
from clear how ajustification of cultural identity schools on the grounds
that they are needed to foster the liberal ideal of personal autonomy is
compatible with a justification on the grounds that they are needed to
August 1998 471
Can the
Liberal State
Support
Cultural
Identity
Schools?
ensure "permanent" cultural survival.
Liberalized cultures are always
in
a process of transition and transformation. And any liberal education
must seek to foster individual choices that contribute to such cultural
changes. Kymlicka
often stresses this point himself-for example, when
he says, "It is right and proper that the character of a culture change as
the result of the choices of its members" (1995, p. 104). But what pro-
ponents of the cultural recognition thesis, including Kymlicka, often
seem to deny is that the process of cultural transition that is driven (in
part) by the autonomous choices of individual members is a process
whose outcome may always carry the potential of cultural assimilation,
fragmentation, or even decay and outright extinction.1 If individuals are
educated in ways that enable them to make meaningful choices, includ-
ing choices to change and reject their inherited cultural traditions, then
the future survival of a culture seems to be thrown into jeopardy. Of
course, cultural rights might be justified on the grounds that they are
needed to ensure cultural survival. But it is hard to see how such rights
could do so without undermining the conditions whereby individuals are
able to develop the capacity to reflect on and revise existing cultural
traditions, norms, and institutions. As such, it is unclear how the cultural
recognition thesis can generate a justification for separate schools de-
signed to foster the dual educational aims of promoting cultural survival
"forever," on the one hand, while also boosting individual autonomy, on
the other. Schools that are justified on grounds of cultural survival can-
not also guarantee they will respect the value of personal autonomy,
while schools that are justified according to the value of personal au-
tonomy cannot ensure that they will respect the value of cultural survival.
Kymlicka's
distinction between the cultural structure as a context of
choice and the character of a culture might seem to provide a way out of
this dilemma. First, Kymlicka refers to the "cultural structure as a con-
text of choice," which he describes as a "good in its capacity of provid-
ing meaningful options for us, and aiding our ability to judge for our-
selves the value of our life-plans." The cultural context of choice must
be distinguished from the cultural character,
which refers to the existing
"norms, values and attendant institutions" that currently characterize a
particular culture (Kymlicka 1989a, p. 166). According to this distinc-
tion, it is really a cultural context of choice that individuals require in
order to maintain the conditions for personal autonomy and not the
particular norms, values, and institutions that currently constitute the
cultural character. Indeed, a cultural structure whose existing character
was altogether impervious to change would necessarily be one whose
members lacked the capacity for cultural reflection and revision that is
required for personal autonomy. So, on this view, liberal cultural identity
schools cannot be justified on the grounds that the self-respect of indi-
472 American Journal of Education
McDonough
vidual members requires firm adherence to a predetermined set of an-
cestral cultural norms, values, and institutions (i.e., a particular cultural
character). Thus, cultural identity schools must be justified by their as-
piration to preserve ("forever") the culture as a context of choice if they
are to be justified by the cultural recognition thesis. Nevertheless, this
attempt to defend the cultural recognition thesis is problematic for rea-
sons I alluded to earlier.
Specifically, in order to generate this sort ofjustification, there needs
to be some principled way of distinguishing between two different kinds
of cultural contexts of choice: first, cultural contexts of choice within
which changes in the cultural character are neutral with respect to (or
supportive of) the autonomy of individuals members; and, second, cul-
tural contexts of choice within which changes in the cultural character
undermine or corrode the autonomy of individual members. Kymlicka
does not offer such a principled distinction and it is difficult to see how
one could be offered. According to one interpretation of Kymlicka's
po-
sition, the notion of a cultural context of choice is so thin and ghostly
that almost all cultural structures qualify for protection by cultural rights,
including cultural structures that are so unstable as to undermine the
personal autonomy of individual members. According to a second inter-
pretation, the cultural context of choice must be conflated with the cul-
tural character, and thus cultural rights designed to protect the cultural
context of choice must also make the cultural character impervious to
change. Either way, a conception of the cultural recognition thesis that
relies on this distinction must betray its commitment to the value of per-
sonal autonomy.
In order to develop this objection, we might begin by examining ex-
emplars of each of the two kinds of cultural structure identified in the
previous paragraph. First, the Quebecois culture provides Kymlicka's
paradigmatic example of a cultural character that has undergone radical
changes without affecting the cultural context of choice in ways that ad-
versely affect the conditions for personal autonomy. Kymlicka points out
that Quebec during the Quiet Revolution in the 1960s underwent a
gradual transition from a "Catholic, conservative and patriarchal" cul-
ture to a diverse, liberal, secular, but no less distinctive, Quebecois cul-
ture. In this case, Kymlicka claims, Quebecois culture underwent dras-
tic changes in the particular cultural norms, values, and institutions
(i.e., its cultural character) that had characterized it in the past; never-
theless, its survival
as a cultural "context of choice" was never threatened
(1995, pp. 87-88). Second, anthropologists Rosemary Henze and Lau-
ren Vanett's (1993) description of several remote, northern Yup'ik Inuit
cultural communities exemplify the second type of cultural context of
choice. Some of the Yup'ik cultural communities described by Henze
August 1998 473
Can the Liberal State
Support
Cultural
Identity
Schools?
and Vanett are so unstable that they provide infertile ground for devel-
oping capacities of personal autonomy and self-respect. These commu-
nities are, according to Henze and Vanett, a complex, incoherentjumble
of disparate cultural influences in which "an extremely complex cultural
shift was underway, making it difficult to see coherence in either Yup'ik
culture or Western culture" (1993, p. 119). Such communities appear to
exemplifyJohn Tomasi's point that "a person's own cultural context may
be so unstable that it is inadequate to help her affirm the worth of her
choices" (1995, p. 589). Nevertheless, Yup'ik people inhabit the cultural
structure they inhabit and not some other one. But this does not mean
that Yup'ik children are thereby deprived of their own distinctive cul-
tural context choice. The unstable, alienating cultural context that cur-
rently exists is, for better or worse, their own culture. Furthermore, it is
precisely this unstable cultural structure whose survival is not threatened
for Yup'ik children. However, the bare notion of a cultural structure of
choice provides too thin a basis for justifying cultural rights, including
cultural identity schools, because it fails to distinguish between stable
and unstable (autonomy-corrosive) cultures. Cultural rights, including
the alleged right to cultural identity schools, must be justified on the
grounds that they are necessary to protect the cultural structure that con-
stitutes the identity of a group of individuals and not some other cultural
structure. Unfortunately, for the Yup'ik, cultural identity schools de-
signed to initiate children into the existing cultural structure might well
be worse than useless since it is precisely this sort of initiation that makes
it so difficult for individuals to attain the capacity for genuine self-
determination in the first place. Thus, the attempt to base ajustification
for state sponsorship on the bare notion of a cultural context of choice
may, in some cases, be tantamount to state endorsement of cultural
degradation because the educational institutions that the cultural rec-
ognition supports must function to "protect" a cultural structure that
itself corrodes and undermines values like self-respect and personal au-
tonomy. But these were supposed to be the values that cultural rights,
and cultural identity schools, were supposed to protect. This, of course,
is a result cultural recognition proponents do not intend to support, but
it follows from their theoretical commitments nonetheless.
Proponents of the cultural recognition thesis may respond at this
point by arguing that it is not initiation into just one's existing cultural
structure they have in mind, but rather initiation into a stable cultural
structure that is required. Kymlicka
himself sometimes seems to suggest
this interpretation when he speaks of the need for individuals to have a
"stable" or "secure" cultural structure and not simply the cultural struc-
ture they happen to inhabit at present (e.g., see Kymlicka
1989a, pp. 167,
474 American Journal of Education
McDonough
169, 170). Kymlicka
does not clearly define what he means by a stable or
secure minority cultural structure, and it is difficult to see how he could
do so without conflating cultural context of choice and cultural charac-
ter. But once these two are conflated as the basis for a justification of
cultural rights, the value of autonomy must be sacrificed and thus the
justification becomes illiberal.
Tomasi usefully points out that Kymlicka sometimes seems to re-
gard instability in the cultural context of choice as involving only those
changes to the cultural character that "threaten ... the history, language
and culture of the group" (Tomasi 1995, pp. 589-90). But this account
of instability simply reintroduces the same dilemma in a new context. For
example, were the changes in the cultural character of Quebec during
the Quiet Revolution threats to the culture and history and language of
Quebec? So far, of course, Quebec has survived and flourished as a dis-
tinct culture with its own history, language, and culture. But how are we
to know whether the transformations undergone in Quebec are part of
an ongoing transformation of a stable and secure cultural structure or
simply stages in a longer process of cultural assimilation to some larger
(perhaps regional or even global) cultural structure? Clearly, some ap-
parently benign changes to the cultural character of a group may turn
out in the future to be profound threats to the group's history, language,
and culture. More importantly, it is impossible, a priori, to determine
which changes will turn out to constitute such threats and which ones
will not. Kymlicka's
notion of a stable cultural structure as a context of
choice cannot provide us with a principled way of distinguishing be-
tween different sorts of cultural transformations. Thus, changes in the
cultural character may always turn out, sooner or later, to be changes
that threaten the continued existence of a group's language, history, and
culture.
Are such threats also threats to the individual autonomy of cultural
members? Not always. For example, John Danley has pointed out that
many ethnic groups in the United States-Italian Americans, Irish Ameri-
cans, Polish Americans-have evolved in such a way that a distinctive
culture no longer exists as a "context of choice distinct from the domi-
nant culture into which they have become assimilated" (Danley 1991,
p. 214). Of course, it does not follow from this that members of assimi-
lated ethnic cultural groups ever lacked their own distinctive cultural
context of choice. The cultural context of choice simply changed (albeit
gradually) for them. However, the threat to autonomy seems more likely
in cases like that of the Yup'ik. Here, changes in the cultural structure
not only threaten extinction of the group's history, language, and cul-
ture, they do so in ways that undermine the autonomy of cultural
August 1998 475
Can the
Liberal State
Support
Cultural
Identity
Schools ?
members. Nevertheless, this interpretation of Kymlicka's
notion of cul-
tural stability in terms of threats to the "history, language, and cul-
ture" is so thin that it fails to distinguish in any principled way between
cultural structures that are autonomy supporting and those that are not.
To this extent, it also fails to distinguish between the legitimacy of state-
sponsored separate cultural identity schools designed to foster a cultural
context of choice within which children may develop a secure identity as
members of a distinct, inherited cultural identity, and separate schools
designed to foster cultural assimilation or separate schools that foster
degradation of children's personal autonomy.
The alternative for proponents of the cultural recognition thesis is to
argue that any change to a group's cultural character is a threat to its
language, history, and culture. And therefore, in order to have the con-
ditions for personal autonomy and self-respect, a group's cultural char-
acter must be protected from change. It follows that a justification for
cultural identity schools based on this interpretation of cultural stability
would have to endorse state support for educational measures that would
preserve the stability of the existing cultural character. However, this re-
sponse runs into problems that we have already discussed. Any successful
educational attempt to protect the cultural character from changes must
also undermine the conditions of personal autonomy for individual cul-
tural members. For these measures would have to ensure that individuals
are initiated into cultural traditions in ways that disable them from re-
flecting on and revising the existing norms, values, and attendant insti-
tutions of their culture. This does not mean that such practices must
indoctrinate children in ways that make them utterly unreflective peons
to cultural orthodoxy. For example, it might allow for children to en-
gage, at least at advanced levels of cultural reflection, in critically and
creatively evaluating the meaning and interpretation of existing cultural
norms, values, and institutions, and to defend them against adversaries.
Nevertheless, the sort of criticism and creativity allowed could not ex-
tend beyond a kind of "ingenious repetition" of accepted cultural and/
or linguistic traditions (Lizhi and Link 1996).12 It cannot allow for the
kind of critical questioning that might lead to rejection of the traditions
themselves because if it did, the process of self-examination might lead
to cultural changes that contribute to instability in the cultural charac-
ter. Educating for firm, unshakable commitment to ancestral values
and roles would therefore undermine the sort of rigorous, critical self-
examination that liberal proponents of the cultural recognition thesis
must endorse. Of course, separate cultural identity schools might be em-
ployed in attempts to support the ends of cultural survival and stability
in the cultural character (indeed, this is one plausible way of interpreting
demands by some minority cultural groups for separate cultural identity
476 American Journal of Education
McDonough
schools); nevertheless, such schools cannot be justified in the name of
the values of self-respect and personal autonomy, at least as these are
understood by liberal proponents of the cultural recognition thesis.
So far I have argued that the cultural recognition thesis cannotjustify
state support for cultural identity schools on the grounds that they are
required to ensure the survival of (a) an existing cultural context of
choice; (b) a stable cultural context of choice. Each of these notions is so
emaciated that no limits are imposed on educational or social con-
ditions that would foster cultural assimilation or cultural destruction.
Or, they turn out to conflate cultural context of choice and cultural
character, thereby justifying educational measures that would enforce a
highly traditionalist cultural orthodoxy hostile to the liberal value of per-
sonal autonomy.
"The Cosmopolitan
Alternative"
The cultural recognition thesis is founded on an important insight-
namely, that individuals cannot acquire or exercise the capacity to
choose from among a range of conceptions of the good life unless they
have the cultural conditions that enable children to learn what it is like
to have a conception of the good in the first place. Furthermore, the
cultural recognition thesis recognizes that these conditions may be very
fragile or altogether absent for children from some minority cultures. I
now show how this insight can be used to infer a liberal justification for
moderate cultural identity schools. The justification begins from a ver-
sion of liberalism whose most philosophically sophisticated defense has
been developed byJeremy Waldron (1995, 1996). Waldron regards his
version of "cosmopolitan" liberalism as incompatible with the cultural
recognition thesis-both as a conception of cultural identity and as
a conception of multicultural education. However, while I agree with
Waldron that cosmopolitanism offers an important corrective to the con-
ception of cultural identity offered by proponents of the cultural recog-
nition thesis, he also exaggerates the conflict between these two views.
In contrast to the proponents of cultural recognition, Waldron holds
that individuals do not require membership in a particular cultural struc-
ture in order to develop and sustain a sense of personal autonomy.
Waldron acknowledges that, for real people, different roles, narratives,
and commitments can only acquire meaning within a specific cultural
context. Nevertheless, "it does not follow that there must be one cultural
framework within which each available option is assigned a meaning.
Meaningful options come to us as items or fragments from a variety of
cultural sources" (Waldron 1995, p. 106). Thus, Waldron argues that
August 1998 477
Can the Liberal State
Support
Cultural
Identity
Schools?
individuals need not have a single cultural structure in order to make
meaningful choices about their lives; rather, individual identity should
be conceived as a kind of melange, or hodgepodge, of commitments,
affiliations, and roles that reflect disparate cultural influences (Waldron
1995). According to Waldron, if "we live the cosmopolitan life, we draw
our allegiances from here, there, and everywhere. Bits of culture come
into our lives from different sources, and there is no guarantee that they
will all fit together" (1995, p. 112).
According to Waldron, educational practices that attempt to initiate
children into distinct, separate cultural structures or frameworks are
morally invidious to the extent that such initiation would necessarily un-
dermine the values of personal autonomy and self-respect. Waldron ar-
gues that when we treat someone as though their identity is constituted
by a whole, single cultural structure "we run the risk of oppressing [an
individual's] own identity in favor of some version of it we have con-
structed" (1996, p. 113), and we risk treating them as "mere artifacts of
the culture of the one community to which we think they ought to be-
long" (Waldron 1996, p. 114). Thus, Waldron says, "To put it crudely, we
need culture but we do not need cultural integrity. Since none of us
needs a homogenous cultural framework or the integrity of a particular
set of meanings, none of us needs to be immersed in one of the small-
scale communities which, according to Kymlicka and others, are alone
capable of securing this integrity and homogeneity. Some, of course, still
may prefer such immersion, and welcome the social subsidization of
their preference. But it is not, as Kymlicka
maintained, a necessary pre-
supposition of rational and meaningful choice" (1995, p. 108). In a mo-
ment, I shall argue that it is seriously misleading to claim that "none of
us" needs what Waldron calls "cultural integrity." For now, I will simply
point out that Waldron's objection to cultural recognition seems to de-
pend on a conception of homogenous, small-scale, monocultural struc-
tures rather than diverse, heterogeneous, multicultural structures. I have
already argued that cultural recognition's liberal proponents do not and,
at any rate, cannot claim that individuals require an education that ini-
tiates them into a homogenous, small-scale cultural community in order
to have a sense of self-respect and personal autonomy. They must be ar-
guing that individuals require an education that initiates children into
extensively multicultural, liberalized, heterogeneous cultural structures.
Once we recognize this, Waldron's charge that educational practices
that initiate children into a single cultural structure "run the risk of op-
pressing [an individual's] own identity in favor of some version of it we
have constructed" (1996, p. 113) is not exactly false; but it is overblown.
In practice, any form of education can be misapplied in oppressive ways.
478 American Journal of Education
McDonough
But there is no special problem here for those who support initiation
into a single, multicultural cultural structure. Such an initiation can pro-
vide individuals with the capacity to choose, revise, and reject the cul-
tural commitments that constitute their inherited identity, including the
ability to reject ancestral traditions in favor of traditions whose cultural
source is alien to those of one's ancestors. Thus, according to the cultural
recognition thesis, an education that initiates a child into a single cul-
tural structure does not in itself present any particular risks with respect
to "imposing" some prior, orthodox version of cultural identity. For, to
do so would have to constitute a morally grievous form of miseducation
from the perspective of cultural recognition, just as it would from the
perspective of cosmopolitanism.
Cosmopolitanism's corrective to the cultural recognition thesis can be
discerned by examining their divergent educational implications. Spe-
cifically, while cosmopolitanism need not reject the possibility that au-
tonomous individuals can be members of distinct cultural structures, as
Waldron points out, these structures are neither necessary nor sufficient
for individuals to develop and sustain an autonomous identity (1996,
p. 113). As a result, a multicultural education that respects individual
identity must treat children as though they are members of more than a
single cultural structure and members of less than a single cultural struc-
ture. Waldron develops this point as follows. First, cosmopolitan multi-
cultural education requires that we treat individuals as members of less
than a single cultural structure because any particular child may find that
some of the traditions, norms, institutions, and values of her cultural
community are simply irrelevant to her identity. People who are mem-
bers of the same cultural community are "located" differently within
that community according to their own cultural choices and according
to the status they have been assigned within the community. As a result,
a child's identity may be constituted only partially by the traditions of
her cultural community-for example, those parts within which she
feels included and at home. She may view other aspects of her cultural
tradition as utterly remote and detached from her own identity. Indeed,
it is highly unlikely that in vastly diverse (post)modern pluralistic cul-
tures, any single individual will find her identity constituted by the whole
range of cultural options her cultural structure has to offer. Thus, ac-
cording to Waldron, in order to respect someone, we must understand
the particularity and partiality of their position within any single cultural
structure.
Second, cosmopolitanism requires that multicultural education treat
individuals as members of more than merely a single cultural structure
because children's identities are constructed and shaped through the
August 1998 479
Can the
Liberal
State
Support
Cultural
Identity
Schools?
interaction of cultural traditions. This is true of children who have
adopted fragments of alien cultural traditions as their own. It is also true
of children from isolated cultures, such as the Yup'ik, whose identities
are nonetheless powerfully shaped by outside forces. The words of a
poem by a Yup'ik girl, Etta Bavilla, capture the point of multicultural
education here: "My
people must learn from both worlds as they collide"
(Henze and Vanett 1993, p. 116). The image of a collision of cultural
worlds suggests that children's identities are not, or at least cannot re-
main, products of a single, distinct, cultural structure, however culturally
diverse the structure itself may be. Rather, our identities are constructed
out of cultural materials far more fragmented, partial, conflicting, open-
ended, and incomplete. Cosmopolitanism, unlike cultural recognition,
fully and clearly recognizes the corrosive effects that an education for
autonomy must have on the aspiration to cultural survival. David Hollin-
ger has recently said that "while cosmopolitanism is willing to put the
future of every culture at risk through the sympathetic but critical scru-
tiny of other cultures, pluralism is more concerned to protect and per-
petuate particular, existing cultures" (1995, p. 85). Hollinger's contrast
between cosmopolitanism and what he calls pluralism could equally il-
lustrate the contrast I am emphasizing here between cosmopolitanism
and what I have called cultural recognition. Cultural recognition need
not take the extreme conservative view that existing cultures must be
immunized from all influences that have a toxic effect on existing cul-
tural norms, roles, and traditions (cultural character). But it does re-
quire that such toxins be neutralized so as not to affect the survival
of the
culture itself-for example, through assimilation to the dominant cul-
ture. As we saw previously, cultural recognition faced a problem in
clearly defining what, if anything, was left to protect once cultural inter-
change and reflection infect the stability of the existing cultural charac-
ter. But cosmopolitanism does not face such a problem since it recog-
nizes that once a cosmopolitan ideal of personal autonomy becomes an
educational ideal, the cultural survival of existing groups is always
at risk.
I said earlier that Waldron was seriously misleading when he claims
that "none of us needs a homogenous cultural framework or the in-
tegrity of a particular set of meanings," and now we are in a position to
see more clearly why this is so. Cosmopolitanism's deep suspicion of
multicultural education justified on the grounds of cultural survival is
compatible with a recognition of the importance that cultural integrity
may have as the basis of education for young children newly embarking
on the journey of cosmopolitan identity formation. But Waldron ne-
glects this since he is insufficiently sensitive to the complex and diverse
processes by which a cosmopolitan, multicultural identity might be
formed. Specifically, he does not consider the possibility that, at least for
480 American Journal of Education
McDonough
some children, a cosmopolitan identity might best be formed precisely
through one's initial upbringing in a singular, homogenous, monocul-
tural tradition.
Waldron regards the novelist Salman Rushdie as exemplifying this cos-
mopolitan ideal of cultural identity, and it is worth considering just how
remote this ideal may be from the lives of children from threatened mi-
nority cultural structures who experience the kind of ambivalence I have
been describing (Waldron 1995, p. 93). In one passage that Waldron
does not cite, Rushdie describes an artist's
work as a process of forging
an identity out of various, disparate cultural traditions:
The Indian writer, looking back at India, does so through guilt-
tinted spectacles. (I am of course, once more talking about myself.)
I am speaking now of those of us who emigrated ... and I suspect
that there are times when the move seems wrong to us all, when we
seem, to ourselves, post-lapsarian men and women. We are Hindus
who have crossed the black water;
we are Muslims who eat pork. And
as a result-as my use of the Christian notion of the Fall indicates-
we are now partly of the West. Our identity is at once plural and
partial. Sometimes we feel that we straddle two cultures; at other
times, that we fall between two stools. But however ambiguous and
shifting this ground may be, it is not infertile territory
for a writer to
occupy. If literature is in part the business of finding new angles at
which to enter reality, then once again our distance, our long geo-
graphical perspective, may provide us with such angles. Or it may be
that it is simply what we must think in order to do our work. (Rush-
die 1991, p. 15)
Rushdie is a self-consciously imaginative artist, confident in his own pow-
ers, and sure of the value to be found in the disparate traditions from
which he forges his identity. However, the feasibility of constructing a
cosmopolitan identity a la Rushdie presumes that sensibilities and intel-
lectual capacities are established-for example, capacities of what the
novelist Anita Desai has called "the unconfined imagination and danger-
ous fantasy" (Desai 1993, p. 26). But the same sensibilities and capacities
might be extremely difficult to establish in the lives of children of disad-
vantaged communities whose ancestral ways of life are threatened by
outsiders.
For example, in the case of an artist like Rushdie, exercising an un-
confined imagination frequently entails challenging established cultural
traditions and orthodoxies and establishing new or revised, often con-
flicting, cultural meanings in their places. The "danger" here, then,
might refer to the threats that "fantasy"
may pose for those like the pro-
ponent of cultural recognition who wishes to preserve established
August 1998 481
Can the Liberal State
Support
Cultural
Identity
Schools ?
cultures and commitments. For the cosmopolitan artist, unlike for the
guardian of cultural survival, this sense of danger is coupled with the
confidence and excitement that new meaning can be found in the me-
lange of cultural fragments available for examination and reflection. But
for Yup'ik children, both the excitement and confidence that are sup-
posed to go with the exercise of imagination may be decidedly lacking
precisely because their ability to find value in ancestral cultural traditions
is so deeply tainted with ambivalence from the outset. For the child try-
ing to forge an identity in a highly volatile, unstable, disadvantaged, "fail-
ing" cultural community, the value of one's initial cultural beliefs are
going to be extremely difficult to establish in the first place. The sort of
deep ambivalence that such children may experience makes it extremely
unlikely that they will be able to examine and evaluate their inherited
cultural traditions with the confidence and sense of excitement that
Rushdie evinces. The survival of these ancestral traditions is severely
threatened by ways of life developed by others-namely, by the vast cul-
tural and technological hegemony of Western civilization. But for many
children like the Yup'ik, the collision of ancestral and Western ways of
life has not ushered in a cosmopolitan multicultural context in which
they can confidently choose among a rich smorgasbord of cultural op-
tions. Rather, it has wrought a disorienting sense of cultural dislocation
marked by a torrent of social and cultural problems-selfishness, greed,
and alienation, as well as rapidly increasing alcoholism, drug use, rising
suicide and school dropout rates, unemployment, and so on. For liber-
als-whether supporters of cosmopolitanism or of cultural recognition-
the point of multicultural education within this highly unstable and de-
graded cultural context is to somehow enable children to undertake the
difficult task of identity construction that requires them to evaluate and
choose among a wide range of conflicting ways of life-namely, those
offered by ancestral traditions and those offered by Western cultural
traditions.
Of course, in order for such choices to be meaningful ones, children
must be capable of appreciating the relative value of the options avail-
able. However, such a capacity may be extremely difficult to develop for
children whose ancestral ways of life are threatened by vastly more pow-
erful and hegemonic cultural and technological imperatives of Western
postmodernity. For example, these children may find it extremely diffi-
cult to find value in their ancestral traditions because those traditions
bear the stigma of a failing culture. But the same traditions are ways of
life developed and sustained by those the children love. Under these
conditions, any sense of pride or confidence that children might have
in their ancestral culture is likely to be silenced by the destructive roar
of Western cultural and technological dominance. Thus, children who
482 American Journal of Education
McDonough
grow up in a threatened culture may develop a strong attachment to
their ancestral culture-like it or not, it is a culture they have inherited
from those closest to them-while at the same time developing a strong,
competing sense of shame about their ancestral ways
of life, because they
bear the stigma of a failing or failed culture. Under these conditions, it
may be impossible for children to develop the capacities needed for en-
gaging in meaningful evaluation of their inherited traditions.
It is easy to see how this same sense of ambivalence on the part of
aboriginal children about their cultural inheritance might provide the
backdrop to the failure of personal autonomy and agency that propo-
nents of the cultural recognition thesis worry about in developing their
arguments for cultural rights. We can now see clearly how cosmopolitan
liberalism might justify moderate cultural identity schools. Both cosmo-
politanism and cultural recognition require that individuals be educated
in ways that foster the capacity for choosing among a wide range of con-
ceptions of the good life. This means that the liberal education children
receive must acknowledge and foster the capacity to evaluate different
ways of life as they come to us from disparate and conflicting cultural
influences. Cosmopolitanism, unlike cultural recognition, recognizes
the threat this poses for advocates of permanent cultural survival. Nev-
ertheless, in order to be able to choose among different conceptions of
the good life, children must be able to have some idea of what it is like
to have a conception of the good life in the first place. And thus the
cosmopolitan liberal must recognize the insight, generated by the cul-
tural recognition thesis, that for some children, initiation into a stable,
established cultural identity may be a necessary prerequisite for the later
development of our more mature cosmopolitan reflective capacities.
Liberalism and Cultural Identity Schools
Importantly, the difference I have emphasized so far between strong and
moderate cultural identity schools rests primarily on a philosophical dis-
tinction of their divergent educational aims and not does not say any-
thing about the actual effects of real cultural identity schools. But this
philosophical distinction does not, by itself, determine which actual cul-
tural identity schools can or must be supported by a liberal state and
which ones cannot. Nevertheless, in my critique of Waldron, I empha-
sized the need for sensitivity to the diverse ways in which children may
develop an autonomous identity in real cultural contexts. Communities,
parents, and educators may adhere to a common set of educational as-
pirations and ideals, but how those aspirations are realized may vary
greatly across different communities, schools, and classrooms. At the
August 1998 483
Can the
Liberal
State
Support
Cultural
Identity
Schools
?
same time, the actual effects of schools may diverge from, and even con-
flict with, the educational aims to which they aspire. For example, a com-
munity whose cultural identity schools explicitly aspire to permanent
cultural survival may in fact provide students with the cultural rooted-
ness that allows them to develop an autonomous identity. Alternatively,
aspiring moderate cultural identity schools may reinforce children's
cultural identities in autonomy-corrosive ways.l3
The effects schools have
on individuals, local communities, and the larger community will un-
doubtedly depend on the complex interactions among numerous
factors-school curricula, pedagogy, organization, home environment,
geography, and so on. As McClaughlin points out, in considering the
issue of state support for separate schools, "Many other issues (for ex-
ample, of a practical, sociological, political and demographic kind) need
to be considered in a full assessment of practical educational policies"
(1992, p. 115). I do not have the space to undertake anything like a full
assessment of these practical issues. Nevertheless, I would like to discuss
briefly some concrete, contextual issues to which a liberal state may need
to be sensitive in determining whether to offer support for cultural iden-
tity schools.14
One problem in evaluating demands for cultural identity schools in-
volves determining who is demanding them and what precisely are they
demanding when they seek to foster cultural survival. Molefi K. Asante's
notion of Afrocentric education is often thought to underwrite a de-
mand for permanent cultural survival,
and his work is shot through with
explicit and implicit assumptions (undefended) about the necessity of
African-American children receiving an education that permanently re-
inforces their ancestral identity.15 To this extent, Asante and his followers
seem to demand strong cultural identity schools. Yet there is much
amidst Asante's grandiloquent rhetoric that is compatible with a more
moderate version of cultural identity schools. For example, in what can
aptly be described as a cosmopolitan mode, he says that "multicultural-
ism in education is a non-hierarchical approach that respects and cele-
brates a variety of cultural perspectives on world phenomena" (Asante
1991, p. 172). And thus, separate African-American immersion schools,
or single-sex urban schools for black males, might cultivate an Afrocen-
tric identity merely as a stepping stone that can be revised and perhaps
even left behind once children are capable of developing a more cos-
mopolitan identity (Asante 1991, p. 172). Indeed, there may be no pure,
unadulterated demands for strong cultural identity schools. But the lack
of purity may result not from the fact that Asante and others do not as-
pire to cultural identity schools that foster cultural survival, rather, it may
result from a failure on the part of proponents of those schools to be
fully sensitive to the ethical and educational significance of the distinc-
484 American Journal of Education
McDonough
tion between educational practices that foster autonomous reflection
with respect to cultural traditions and practices that foster cultural iden-
tity in ways that impede autonomy.16
A less ambiguous demand for strong cultural identity schooling can be
found in Dianne Longboat's claim that "education can enhance the sur-
vival of First Nations people only if it contributes to identity development
through learning our languages, our cultural traditions and our spiritual
beliefs. The development of Indian human beings involves this type of
affective growth.... A First Nations person must first know himself, his
clan, his nation, and his responsibilities if he is to function as an Indian.
An Indian identity provides a framework of values upon which one views
life, the natural world and one's place in it" (1987, p. 40). Here, Long-
boat seems to equate education for Indian human beings with the devel-
opment of individuals who must adhere to ancestral traditions, roles,
and beliefs throughout their lives.17
In fact, however, First Nations chil-
dren in North America often find themselves torn at a very early age
between the ancestral cultural traditions of their parents and the much
more powerful cultural traditions that constitute hegemonic Western
technological and economic imperatives. The desire on the part of
Longboat and many other advocates of state-supported cultural identity
schools may be motivated by the fact that for many children this highly
unequal cultural tug-of-war provides an unstable cultural context that,
far from providing fertile soil for the development of personal autonomy
and self-respect, is more likely to result in alienation and despair. This
concern about the failure of personal agency is also the one that moti-
vates liberals who support cultural rights. But it is worth noting here that
the conditions under which personal autonomy may begin to flourish in
the lives of individuals can hardly be established or restored by initiating
children into the kind of ancestral tradition that Longboat advocates. An
aboriginal child who grows up in a cultural context marked by the uneasy
and highly unequal coexistence of ancestral traditions alongside those
of Western postmodernity is likely to experience frequent and acutely
conflicting cultural commitments. When this happens, as it frequently
does, children who are to develop a genuine sense of self-determination
must be able to intelligently examine the full range of options that give
rise to the conflicting obligations in the first place-in other words, to
examine meaningfully the cultural options offered by both Western and
ancestral cultural traditions.18
But it is not always easy to determine what sorts of educational
and other empirical conditions are needed in order to foster this aim.
But determining these conditions may be crucial for deciding when
actual claims to cultural identity schools translate into an obligation on
the part of a liberal state to support such schools. For example, it seems
August 1998 485
Can the Liberal
State
Support
Cultural
Identity
Schools?
reasonable, even preferable, for a liberal state to support moderate cul-
tural identity schools when these are most likely to provide for young
children a "continuity between the values of the family and the ethos of
the separate school" and thus help to provide "a more solid ground-
work" for developing a cosmopolitan identity than, say,
common schools
would be able to provide (Callan 1995, p. 267). But not all minority cul-
tures, and not all communities within a particular culture, will find that
moderate cultural identity schools provide the desired congruity be-
tween family values and school ethos. And some may find that the ethos
of such schools clash harshly with family environments.
In this light, consider one such set of factors that involves issues of
pedagogy in cultural identity schools. A number of empirical studies
have suggested that culturally sensitive teaching styles are linked to im-
proved academic effectiveness in North American Indian schools. Here,
according to D. Deyhle and K. Swisher, most available evidence is based
on studies of non-Indian teachers of Indian students (1997, p. 149).
But what counts as culturally sensitive pedagogy depends radically on a
number of contextual factors, including the interaction between the cul-
ture of the school and the background culture of the local community.
For example, Deyhle and Swisher describe the case of a teacher who
once tried to engage her Pomo Indian students with an ancestral story
about Slug Woman. The students, according to Deyhle and Swisher, "re-
sponded with either open hostility or stone silence." Furthermore, the
authors attribute this response to the fact that "the students saw the
event as one more attempt to set them up for failure by superficially 'ad-
mitting' a limited and static view of their culture into the classroom"
(1997, p. 165). How might we explain the children's stony and hostile
reaction to the Slug Woman story?
One interpretation explains the children's hostility and silence to the
Slug Woman story as a result of the fact that the story is not genuinely
part of their inherited identity at all. Thus, what the teacher takes to be
the children's ancestral culture may simply be irrelevant to their cultural
identity.19 According to this interpretation, what at first appears to be
culturally sensitive pedagogy might in fact constitute a form of cultural
misrecognition. In order to explain why this might be so, suppose the
background to this case were fleshed out further in the following way.
Not all Indian communities are like the highly unstable one described
earlier by Henze and Vanett. Suppose, for example, that the classroom
in the example above is situated in a cultural community much more
akin to the stable, modern, liberalized Quebecois culture described by
Kymlicka. Suppose also that students in the school come from homes
that tend to embrace the cosmopolitanism of the larger culture. In this
case, children may not require initiation into a stable ancestral tradi-
486 American Journal of Education
McDonough
tional culture as a basis for developing a capacity for autonomy, and they
may view attempts to do so as condescending rather than respectful. In
Waldron's terms, they may view such schools as insulting because the
schools treat children as mere artifacts of a single, homogenous culture
to which the school thinks they should belong, but to which in fact they
do not belong at all. The upshot of this is that a liberal state should not
assume a consistency that does not exist across communities within a
single cultural group (e.g., the Yup'ik) and certainly not across different
cultural groups. Some communities and groups may find moderate cul-
tural identity schools appropriate vehicles for helping students negotiate
the intricate and hazardous path between ancestral and Western cul-
tures, while others may find them useless at best and condescending at
worst. In cases like this, it may be appropriate for a liberal state to afford
wide latitude to parental and local community choice in determining
whether moderate cultural identity schools are appropriate or not.
A second interpretation of this example draws on the earlier psycho-
logical speculation about aboriginal children's potential ambivalence
about their inherited, ancestral identity. According to this interpreta-
tion, the children's hostility and silence in response to the Slug Woman
story may reflect the experience of deep ambivalence about their inher-
ited ancestral culture that we discussed earlier. On the one hand, the
children may have a strong attachment to the story of Slug Woman since
they understand it as part of the cultural heritage of the people they love.
On the other hand, they may be repelled and embarrassed by virtue of
the fact that it represents the stigma of a failed or failing culture. These
emotions may be intensified if the teacher in the case is a cultural out-
sider. Thus, their ability to find value in their cultural heritage in the
classroom is made difficult not because their heritage is irrelevant to
their identity, but because they find it difficult to see the value of ances-
tral traditions, which have been stigmatized by the hegemony of other,
more powerful ways
of life. Suppose further that the larger cultural back-
ground to the school in this case is more like the unstable Yup'ik culture
described earlier than the Quebecois culture described by Kymlicka.
As
Deyhle describes, public schools in some Indian communities fail to re-
dress, and even contribute to, social problems such as cultural alien-
ation, marginalization, high dropout (or "pushout") rates, alcoholism,
drug problems, and so on (Deyhle and Swisher 1997). In contexts like
this, moderate cultural identity schools might be necessary as a means of
establishing for kids an initial sense of value in their ancestral traditions
so that later they may be examined in light of alternative cultural op-
tions, rather than merely overwhelmed by the hegemony of other ways
of life. And here the state may have a strong obligation to support mod-
erate cultural identity schools and even encourage communities to
August 1998 487
Can the Liberal
State
Support
Cultural
Identity
Schools?
develop them. It might also need to support and encourage efforts to
train and recruit teachers who are cultural insiders.
The two accounts of the Slug Woman story I have outlined are not
meant to exhaust the possible range of interpretations of this example.
Furthermore, both interpretations are based on extensive, though plau-
sible, psychological and sociological speculation. The purpose of exam-
ining the Slug Woman example was to illustrate some of the complexity
that a liberal state must take account of in determining when moderate
cultural identity schools may require support and what kind of support
might be involved.
A second contextual factor relevant to determinations of legitimacy
for moderate cultural identity schools has to do with curriculum. It is also
important to keep in mind that moderate cultural identity schools, which
are committed to kindling a keen appreciation of the value of ancestral
traditions so that children can examine these traditions wisely and mean-
ingfully in the light of competing traditions, may be difficult to distin-
guish from strong cultural identity schools, which attempt to indoctri-
nate children into an exclusive permanent cultural identity. This can be
illustrated by a second example, which happens to concern curricular
issues in the context of demands for separate, public single-sex schools
for African-American males. Supporters of such schools sometimes give
a nod in the direction of multicultural curricula that include more than
merely ancestral traditions and values; nevertheless, it is not clear that
the nod is vigorous enough to satisfy the conditions for personal au-
tonomy development. For example, supporters of Afrocentric curricula
in separate schools for African-American males are sometimes said to
require examining "many cultures, but do so from the perspective of
African-American people" (cited in Karp [1991], p. 89). But personal
autonomy is not necessarily fostered simply by including a study of other
cultures in the curriculum, and indeed some forms of inclusion may
positively obstruct the development of personal autonomy. For example,
teaching about other cultures from the perspective of what is taken to be
a single, homogenous culture of upbringing may involve teaching that
the other cultures are rival, alien cultures and traditions to be rejected
and resisted in favor of the traditions and values privileged by the culture
of upbringing. In this way, children might be educated in ways that dis-
courage them from meaningfully evaluating and adopting cultural tra-
ditions and values that conflict with those of their ancestral culture. I do
not know whether the comment attributed to the speaker in the example
was intended to advocate such an illiberal view of education. Here the
point I want to stress is that merely including the study of alien cultures
within cultural identity schools does not necessarily do anything to
develop the kind of multicultural identity that nurtures personal au-
488 American Journal of Education
McDonough
tonomy; indeed, such a curriculum may work to erode autonomy, and
hence by itself does nothing to justify state sponsorship of cultural iden-
tity schools.
Suppose, however, that in a given cultural context, we are able to de-
termine with reasonable certainty the curricular and pedagogical condi-
tions under which moderate cultural identity schools are capable of fos-
tering a coherent link between the values of local community families
and the ethos of the school. Suppose also that such schools included a
curriculum (texts, films, materials) which, at advanced levels, sought to
seriously excite children's imaginative exploration of diverse cultural
traditions. In this context, some conditions would support ajustification
for state support of moderate cultural identity schools, but demographic
and other factors may still weigh against such a justification. For ex-
ample, capacities for sympathetic imagination of alien cultural traditions
may be difficult if not impossible to develop in geographically remote
areas where students' contact with actual members of alien cultural tra-
ditions are limited or nonexistent. Furthermore, even in geographically
less remote areas, relationships between local communities and the
larger community may impede the cosmopolitan aspirations of moder-
ate cultural identity schools. For example, Indian communities that de-
sire moderate cultural identity schools may be quite willing to open
their tribal schools to non-Indian children from surrounding regions
(Richardson 1993, p. 31). But such willingness may be, and often is, met
with indifference or hostility. At the same time, extraschool opportuni-
ties for productive, cooperative, or civil interaction between Indians and
non-Indians, or blacks and whites in urban U.S. communities may be few
and far between in a racist, "savagely
unequal" society (Kozol 1991). To
the extent that such conditions obtain, the aspirations of moderate cul-
tural identity schools will be increasingly difficult to realize in practice.
None of these conditions-taken individually or collectively-deter-
mine whether a liberal state is justified in supporting moderate cultural
identity schools. They may have to be evaluated in the light of still other
factors. For example, the justification for state support of separate
schools needs to be evaluated in the light of available and feasible alter-
natives. In other words, what are cultural identity schools seeking sepa-
ration from? Suppose in a particular community attendance at common
public schools commonly generates high dropout (or "pushout") rates
and low self-esteem. Suppose also that there was evidence that separate
cultural identity schools could significantly alleviate such problems even
though they were unwilling, say, to provide opportunities for meaning-
ful interaction with members of alien cultures. Suppose also that they
sought to foster children's strong adherence to ancestral traditions in
ways that endangered autonomous reflection. State support for such
August 1998 489
Can the Liberal
State
Support
Cultural
Identity
Schools ?
separate schools might, in some such cases, be preferable to coerced at-
tendance at common schools on the grounds that the common schools
pose a greater threat to children's autonomy than the separate schools.
On the other hand, where feasible, a liberal state might also respond to
such conditions by developing more sensitive and attractive
public school-
ing options. A policy of bussing children to or from remote Indian reser-
vations, or across large urban areas, may or may not be feasible and/or
desirable. But, for example, incentives have sometimes worked to lure
affluent white educational populations to inner-city, historically black,
and disadvantaged public schools (Foster 1993). And similar incentives
could help to overcome obstacles that currently hamper cooperative
educational endeavors between Indian and non-Indian populations.
Relationships and divisions within groups may also become relevant in
determining the likely or probable effects of separate schools. For ex-
ample, K. Alston worries that single-sex schools for African-American
males may foster "a univocal construction of black maleness into which
the participant will be inducted-without any reassurance that the con-
struction will allow for different approaches and embodiments of this
'new' cultural stance-toward maleness, manhood, and women" (1993,
p. 124). Such worries have been expressed by many African-American
women concerned about the potential for such schools to reinforce
sexist attitudes and gender inequalities within the African-American
community. Indeed, a school whose admission restrictions (gender or
race based), curricula, and other factors tended to reinforce a univocal
identity of any kind would conflict with the moderate cultural identity
school's aim of fostering personal autonomy. And the fact that curricula
for black male academies in Detroit and elsewhere have at least implicitly
blamed African-American women for the educational disadvantages ex-
perienced by black males, needs to be noted (Alston 1993, p. 8; Garden-
swartz 1993, p. 634). Once again, however, an open mind needs to be
maintained since such factors need not preclude ajustification for mod-
erate single-sex cultural identity schools. For example, moderate cultural
identity schools mightjustify separation on the basis of race according to
the need for (moderate) cultural rootedness; and separation on the basis
of gender might be justified on the grounds that such separation is aca-
demically beneficial for males. But this can be done without necessarily
demeaning whites or black women.
A further, very important, consideration concerns larger structural
economic factors. It seems likely that any attempts by a liberal state to
create and/or support moderate cultural identity schools will meet with
limited success at best in the absence of economic conditions (the pros-
pect of decent jobs, and so on) that enable children to envision mean-
ingful and productive lives. And such conditions are often all but inac-
490 American Journal of Education
McDonough
cessible to children from disadvantaged minority cultures. Once again,
however, this need not mean that moderate cultural identity schools are
unjustified from a liberal perspective. It does mean that the feasibility of
achieving their desired effects is likely to be severely limited unless they
are accompanied substantial reforms designed to decrease structural
economic inequalities.
The point of these reflections is that the business of determining how
liberal principles that establish the legitimacy of moderate cultural iden-
tity schools translate into concrete policy in particular cases is inevitably
an intricate, complex and highly context-dependent undertaking. Fur-
thermore, the process of making such judgments is likely to be a hazard-
ous one in the absence of a more fully established body of empirical
evidence about the real-world effects of cultural identity schools. The
discussion above has attempted to outline some of the relevant com-
plexities while also showing that a liberal commitment to the value of
autonomy requires substantial restrictions on parental and community
demands for cultural identity schools-restrictions involving curricula,
pedagogy, student admission policies, and teacher hiring.
Nevertheless, there are varieties of liberalism that would assign consid-
erably more weight to parental and community demands than the ver-
sion I have developed here. Robert Nozick (1974) is the most notable
advocate of libertarian liberalism. He argues that the state has few legiti-
mate reasons to impose on citizen liberty in order to foster larger social
aims. Also, Chandran Kukathas recently developed libertarian (and pur-
portedly liberal) arguments, based on the right to freedom of associa-
tion, which would assign a great deal of weight to parental and local
community demands of almost unlimited scope for cultural autonomy,
including cultural identity schools. For Kukathas, the scope of a liberal
right to freedom of association can only be limited by "norms forbid-
ding slavery and physical coercion" or "prohibitions on cruel, inhuman
and degrading treatment" (1992, p. 128). Such restrictions would af-
ford parents wide latitude in determining what sort of cultural identity
their children are brought up to endorse-including strongly tradition-
alist, autonomy-corrosive identities. In addition, they could easily be
made to support, for example, school choice policies that leave local
communities free to establish schools that are virtually entirely autono-
mous from state restrictions-including strong cultural identity schools
(Chubb and Moe 1990).
I do not have the space to respond to such views fully here, although I
have done so elsewhere (McDonough 1994, pp. 91-100). However, two
points need to be stressed. First, there are good reasons for believing that
the liberal right to freedom of association generates much more sub-
stantial restrictions on parental choice in schooling than Kukathas's
August 1998 491
Can the Liberal State
Support
Cultural
Identity
Schools ?
arguments would suggest. As Kymlicka argues, the right to freedom of
association protects not only freedom of cultural groups from political
intervention by the state, but also the right of individuals to exit associ-
ations they find uncongenial (1992, pp. 59-60). Furthermore, a substan-
tial right to exit may be violated for children whose education makes
them incapable of revising, reflecting on, and rejecting their inherited
cultural identity. Thus, strong cultural identity schools would be ruled
out by a liberal right to freedom of association that entails a substantial
right to exit.
Second, much of the libertarian position's appeal may result from a
misunderstanding of the implications of autonomy-based liberalism.
Kukathas charges that liberalisms that privilege personal autonomy as a
principle for determining the political legitimacy of parental demands
must generate an aggressively interventionist liberal state which imposes
liberal values in ways that are inevitably insensitive to the values of local
cultures and communities (1992, p. 121). But, as Kymlicka
notes, and as
my discussion of context earlier in this section emphasizes, an autonomy-
based liberal state need not impose liberal values on communities and
parents who disagree with them. First, even in cases where parental de-
mands involve (strong) cultural identity schools that violate liberal prin-
ciples of autonomy, and therefore lack political legitimacy, a liberal state
may refrain from imposing liberal values and may even have practical
reasons to support such demands. This may be the case, for example,
where available or feasible alternatives seem likely to be even more cor-
rosive of personal autonomy than would strong cultural identity schools.
Second, as Kymlicka points out, not all cases of state intervention are
coercive or impositional. As we saw above, the state may sometimes offer
incentives designed to encourage parental support for the aims of mod-
erate cultural identity schools where such support may otherwise be
lacking.
Conclusion: Separate Schools and the
Liberal Ideal of a Multicultural Society
In this article, I have tried to highlight the tension in a liberal response
to the claims of minority cultural groups-a tension between a policy of
protecting, and in some cases supporting, the survival of minority cul-
tural groups, on the one hand, and a policy of encouraging or requiring
cultures to respect the complex, cosmopolitan identity of their indi-
vidual members, on the other hand. This discussion suggests that a lib-
eral response to such questions should not arise out of an attitude of
conservative nostalgia or the desire for preserving cultures-pure or
492 American Journal of Education
McDonough
otherwise-as if culture were ajar of pickles. Some cultures should be
discouraged, penalized, transformed, or prohibited-because they ad-
vocate attitudes of hatred and intolerance, for example. Others may
simply die out, becoming assimilated to other cultures. AsJ. Raz (1994)
says, "So long as the process [of assimilation] is not coerced, does not
arise out of lack of respect for people and their communities, and is suf-
ficiently gradual, there is nothing wrong in it" (p. 74). Just as people die
natural deaths, so may cultures.
The absence of conservative nostalgia also characterizes the attitude
of liberals toward the aim of respecting, or tolerating, the aims of mi-
nority cultural survival in educational contexts. But in educational con-
texts, a special concern needs to be articulated. Here the interaction and
conflicting demands of different cultures reflects not only a natural pro-
cess of change, but a series of grievous inequalities, to which demands
for separate schools are one response. For example, the very formula-
tion of the problem of black male schools or separate schools for First
Nations people already implies a context of deep racial and social class
inequality. In the case of single-sex schools for African-American males,
the expressed desire for separation is from black females. But disadvan-
taged black males might be expected to benefit at least as much from the
absence of white males and females in educational contexts. However,
this is rarely articulated because blacks as a group are already frequently
relegated to segregated, unequal schools (Kozol 1991). And thus the
claim for separation may obscure deeper issues of economic and social
injustice.
Thus, while the goal of a multicultural, cosmopolitan society may be a
laudable liberal ideal, the existence of a cacophony of demands by mi-
nority groups for separate schools does not reflect the fact that we are
approaching that ideal. Rather, it often reflects the fact that separate
schools are the lesser of two evils in a society where public schools of-
ten fail members of minority groups. Separate schools may be, in some
cases, a useful way of attempting to rectify the disadvantages faced by
cultural minorities. However, separate schools designed to protect sepa-
rate cultural identities should be viewed as only one limited set of means
among others of attaining a broader liberal ideal of social and economic
equality in a culturally diverse society.
Notes
I would
like to express
deep gratitude
to Eamonn
Callan,
Walter
Feinberg,
and
four AJE
referees
for their detailed,
thoughtful,
and challenging
comments.
All
of these commentators
forced me to clarify,
expand,
and
improve
my arguments.
August 1998 493
Can the Liberal State Support Cultural Identity Schools?
None of them are responsible, of course, for errors and weaknesses that remain.
I have also benefited from discussions of earlier versions of this article at the
University of Delaware, Ball State University,
and McGill University.
1. Here I have in mind claims to separate schools by advocates of deafness as
a cultural identity, which contrasts with claims based on deafness as a medical
condition or disability.
For an interesting discussion, see Dolnick (1993).
2. For a particularly impressive and comprehensive review of scholarly-
mainly sociological and anthropological-literature on American Indian/First
Nations education, see Deyhle and Swisher (1997).
3. Some prominent liberal proponents reject personal autonomy as the core
value on which liberal determinations of political legitimacy must depend. For
example, Dworkin argues that the source of political legitimacy in a liberal society
is a certain conception of equality, while Isaiah Berlin holds that individual rights
are of supreme value in liberal politics (Kekes 1993, p. 202). Most notably, the
title of John Rawls's recent and massively influential book, Political
Liberalism,
marks his defense of a liberal theory which rejects all such comprehensive moral
ideals as the source of political legitimacy in a liberal society. For Rawls, all such
"comprehensive liberalisms" affirm overriding moral ideals that must inevitably
denigrate the integrity of some citizens in pluralistic societies who will inevitably
reasonably reject the privileged comprehensive moral ideal in question (Rawls
1993, p. 37). For Rawls, such ideals must also be divisive and thus threaten the
unity and stability of the polity (Rawls 1993, pp. 37-38). Nevertheless, there are
good reasons for thinking that the contrast between non-autonomy based liber-
alisms (such as Rawls's
political liberalism, as well as the liberalisms of Dworkin
and Berlin), on the one hand, and autonomy-based liberalism on the other hand,
is less dramatic than it first appears. As Waldron points out, despite the apparent
diversity and complexity of the liberal tradition, liberal politics can be distin-
guished from other ideologies by its distinctive "view about the justification for
social arrangements" (1993, p. 36). This view holds that "all aspects of the social
world should be made acceptable to every last individual" (Waldron 1993, p. 37).
Such a view is affirmed on the grounds that it is necessary in order to ensure
respect for individual integrity of all persons and for securing political stability
in
culturally and ethically pluralistic societies. But such a view of liberalism also nec-
essarily puts a premium on political and educational institutions that develop
reflective capacities of judgment and deliberation that are constitutive of per-
sonal autonomy. Thus, even liberalisms that treat comprehensive ideals other
than autonomy as overriding, as well as liberalisms that claim to reject all overrid-
ing comprehensive moral ideals, must "impose educational requirements that
bring autonomy through the back door" (Callan 1996, p. 21; see also Gutmann
1995).
4. In another article, I have traced the cultural recognition thesis through an
examination of Taylor's
work (McDonough 1998).
5. This is not to deny, of course, that there are important differences between
the theories developed by different liberal theorists. For example, Kymlicka's
ar-
guments are designed to apply mainly to disadvantaged cultures whose survival
is
threatened by a dominant culture. In contrast, Tamir and Taylor view the appli-
cation of liberal group rights as having a wider application to groups whose iden-
tity is not currently seriously threatened. Another difference worth noting be-
tween Kymlicka
and Tamir also has to do with the narrower scope of the former's
theory. In his most recent work, Kymlicka argues that his conception of group
rights is meant to apply only to "societal cultures." A societal culture is one that
494 American Journal of Education
McDonough
"provides its members with meaningful ways
of life across the full range of human
activities, including social, educational, religious, recreational, and economic
life, encompassing both public and private spheres. These cultures tend to be
territorial concentrated, and based on a shared language." Furthermore, these
cultures "must be institutionally embodied-in schools, media, economy, gov-
ernment, etc." (Kymlicka 1995, p. 76). Tamir's theory, on the other hand, would
assign cultural rights of much wider scope-including societal cultures as well as
isolated individuals living in an alien society-and whose identity is therefore
completely unsupported by an institutionally embodied societal culture (Tamir
1995, p. 43). I should stress, however, that I do not intend a comprehensive re-
view of the differences across these theoretical positions. My point here is that
whatever differences may exist, they are not as fundamental as they may first ap-
pear since they are essentially differences of interpretation about the practical
implications of the cultural recognition thesis and do not affect its justificatory
structure or force.
6. Similarly,
Tamir argues that her defense of a "right to culture" for members
of minority national groups depends on a conception of an "autonomous person
who can reflect on, evaluate, and choose his conception of the good, his ends,
and his cultural and national affiliations, but is capable of such choices because
he is situated in a particular social and cultural environment that offers evaluative
criteria" (Tamir 1995, p. 33).
7. My purpose here is simply to draw out some of the educational implications
of a conception of self-respect that is linked closely to the notion of personal
autonomy. Of course, there are other ways
of conceiving the value of self-respect,
some of which may conflict with Kymlicka's. For example, self-respect may be
understood as requiring a strong rootedness in a particular set of received cul-
tural values and roles. Such a conception of self-respect would almost certainly
entail educational practices that sharply diverge from those of proponents of the
cultural identity thesis. For an excellent discussion of this issue, see Tomasi
(1995).
8. A superficial reading of Taylor's (1994) work might misleadingly suggest
that his own
justification for group rights diverges fundamentally from Kymlicka's
and Tamir's
at this point. Taylor encourages this misconception in an interesting
footnote in which he explicitly contrasts his own position with Kymlicka's.
Here
he argues that Kymlicka's justification for group rights, unlike Taylor's
own, can
guarantee state protection only for "existing people who find themselves trapped
within a culture under pressure, and can flourish within it or not at all. But it
doesn't justify measures designed to ensure survival through indefinite future
generations. For the populations concerned, however, that is what is at stake"
(Taylor 1994, pp. 40-41). This is quite similar to the criticism of Kymlicka's po-
sition that I develop later in this article; however, the criticism also applies to
Taylor's own position even though he does not acknowledge it. The gap in
Kymlicka's justification is supposed to lie in Kymlicka's
reliance on the connec-
tion between cultural identity and personal autonomy. But, groups whose indi-
vidual members made choices over time that led them to gradually assimilate to
another culture may well end up inadvertently or otherwise destroying their par-
ticular cultural identity. Taylor
seems to regard his own position as diverging from
Kymlicka's in two respects: first, by offering a deeper justification for group
rights-deeper in the sense that it is based on the value of cultural survival for its
own sake, independent of the value of personal autonomy; second, byjustifying
permanent cultural survival rather than cultural survival merely for existing
August 1998 495
Can the Liberal State Support Cultural Identity Schools?
people. However, Taylor himself avers a notion of authenticity that prizes the
individual's ability to create and shape her own identity within a cultural com-
munity or through a process of intercultural reflection and examination that
leads to a "fusion" of cultural horizons. Thus, Taylor's notion of authenticity
brings in Kymlicka's
notion of autonomy through the back door. Furthermore, it
is this sense of "authenticity" that is supposed to generate the value of cultural
identity in the first place-at least for individuals whose identities are shaped
under cultural conditions of modernity, which for Taylor includes almost all of
us. Once we see this we can also see how his position converges with Kymlicka's
in both of the senses in which Taylor
supposes they diverge. First,
cultural survival
is justified by its connection to authenticity, which is also to say by its connection
to the value of personal autonomy. Second, Taylor fails to acknowledge that a
multicultural education designed to cultivate authentic, "multicultural" identi-
ties at the individual level cannot also "ensure that there is a community of
people here in the future that will want to avail itself of the opportunity to use
the French language" (1994, p. 40). Nor can it seek to ensure that there will
forever be a community of people who wish to avail themselves of, say,
Quebecois
cultural traditions. Of course, this is not to deny that cultural communities may
persist in spite of the stresses and strains of cultural change.
9. In his most recent work, Kymlicka (1995, pp. 35-44) develops a useful dis-
tinction between ensuring cultural survival through "external constraints" and
"internal constraints." Internal constraints involve restrictions on individual lib-
erty within the community (e.g., restrictions on religious liberty, freedom of ex-
pression). Kymlicka opposes these, except in extreme cases, on the grounds that
they endanger individual autonomy. External constraints are constraints on the
rights of outsiders (e.g., wealthy corporations who seek to exploit aboriginal
lands for commercial purposes). Kymlicka supports these for the same reasons
he rejects internal constraints. When external constraints are removed, or never
imposed in the first place, the survival of minority cultural structures may be
threatened, thereby threatening the autonomy of individual members. Cultural
identity schools might be seen, from one angle, as an "external constraint" since
they can be viewed as necessary to protect the identity and autonomy of children
from the hegemonic educational practices that may serve the well-being of out-
siders, but which often have harmful, destructive effects on children within mi-
nority cultures. But, seen from another angle, cultural identity schools have the
potential to impose internal constraints. For example, the aims of a cultural iden-
tity school designed to foster a monocultural identity in the senses discussed
above would have to seek to preclude children's autonomy, thereby largely de-
priving them of the capacity to exercise their individual rights to freedom of ex-
pression, religion, and so on. Thus, the distinction between internal and external
constraints implies at most a narrowly circumscribed right to cultural identity
schools. That is, the distinction suggests that liberalism can justify such schools
only to the extent that they are necessary as an external constraint to protect
children's identity and autonomy; but a liberal state must forbid such schools to
the extent that they impose an education that constitutes an internal constraint
on children's liberty and autonomy. All proponents of the cultural recognition
thesis must similarly acknowledge this implication of their position insofar as
their justification for group rights rests on a notion of respect for persons as au-
tonomous individuals capable of revising and rejecting their inherited cultural
roles and commitments.
10. As I argue above (see n. 9), Taylor's attempt to justify rights to "cultural
496 American Journal of Education
McDonough
survival" of this kind must fail for the same reasons that Kymlicka's
does. For a
fuller discussion of Taylor's
position see McDonough (1998).
11. Of course, some cultures may and do withstand and adapt to the corrosive
effects of the autonomous choices of their members. The point is simply that
cultural survival
cannot be a prior aim of liberal education.
12. The phrase "ingenious repetition" is used by F. Lizhi and P. Link in an-
other context to refer to the restrictive conditions under which scientists must
perform in China. The authors elaborate this notion by analyzing a case in which
a Communist Youth League member questioned party orthodoxy by endorsing
"independent thinking" and is reprimanded by a party secretary. As Lizhi and
Link put it, "To develop skill in political study, said the secretary, one had to
achieve 'ingenious repetition' . . . of the tenets of Marxism, but not to question
the tenets themselves. Individual creativity
was appropriate only in finding differ-
ent ways to express known truths" (1996, p. 44).
13. Although they do not cite research on schools to this effect, Deyhle and
Swisher cite evidence from several studies that suggests that "successful youths
came from moderate homes that valued many of the Navajo traditions while ad-
hering to many modern notions, while students who dropped out of school
tended to be more assimilated to the culture of the dominant society" (1997,
p. 137).
14. Furthermore, a considerable amount of empirical work remains to be
done before we can have the contextual understanding needed in order to de-
velop educational policies that can generate a tolerably determinate equilibrium
between principle and practice in this area.
15. For example, Asante also says that multicultural phenomena will necessar-
ily be viewed "from the perspective of the African person," thereby assuming the
ongoing and permanent identification with ancestral identity. His writings are
replete with comments to this effect.
16. Asante is a good example in this respect. The reasons for Asante's equivo-
cation on this point are undoubtedly complex, but one possible reason is that he
often seems to regard the possibility of a distinction between strong and moder-
ate cultural identity schools as a form of logical trickery and thus incoherent
(1991, p. 219). But if this is his considered view then he will necessarily be blind
to any tendencies in his own work toward the evils of strong cultural identity
schooling.
17. See also Deloria's exhortation to a group of First Nations college students:
"You are an Indian first, last and always.
You may have a degree in anthropology,
law or nuclear engineering, but that is your profession, and how you make your
living; it is not you!! ... Your first responsibility is to be a human being, an Indian"
(cited in Simonelli 1991, p. 15).
18. Of course, it is possible for educators to treat the alienation experienced
by such children as a form of severe psychological dissonance that results from
an upbringing within a cultural context that is constituted by the conflict between
two distinct (monocultural) structures. Based on this diagnosis, the same educa-
tors might seek to alleviate or prevent the alienating psychological dissonance by
initiating children into one of the previously clashing cultural structures in a way
that forecloses meaningful examination of the options available within the other.
But the point I stress here is that such an approach would be morally disastrous
from the perspective of educators dedicated to the development of children's
capacity for self-determination.
19. Or, at least substantial chunks of it may be irrelevant. Some of Deyhle and
August 1998 497
Can the Liberal State Support Cultural Identity Schools
?
Swisher's comments support this interpretation. For example, "In their class-
room, the Slug Woman story represented neither the world they knew nor the
world in which they lived" (1997, p. 166).
References
Alston, K "Community Politics and the Education of African American Males:
Whose Life Is It Anyway?"
In New Politics
of
Race and Class,
edited by C. Marshall.
London: Falmer, 1993.
Asante, M. K. "The Afrocentric Idea in Education." Journal of Negro
Education
60
(1991): 170-80.
Callan, E. "Common Schools for Common Education." CanadianJournal
of
Edu-
cation
203, no. 3 (1995): 251-71.
Callan, E. "Political Liberalism and Political Education." Review
of
Politics
58, no.
1 (1996):5-39.
Chubb, J., and T. Moe. Politics,
Markets and America's Schools.
Washington, D.C.:
Brookings Institute, 1990.
Danley,
J. R. "Liberalism, Aboriginal Rights, and Cultural Minorities." Philosophy
and Public
Affairs
20 (1991): 169-84.
Desai, A. "Sitting Pretty."
New York Review
of
Books,
May 27, 1993.
Deyhle, D., and K. Swisher. "Research in American Indian and Alaska Native
Education: From Assimilation to Self-Determination." Review
of
Research in Edu-
cation
22 (1997): 113-94.
Dolnick, E. "Deafness as Culture." Atlantic
Monthly,
September 1993, pp. 37-53.
Dworkin, R. "Liberalism."
In Public and Private
Morality,
edited by S. Hampshire.
Cambridge: Cambridge University Press, 1978.
Foster, M. "Savage Inequalities: Where Have We Come From? Where Are We
Going?"
Educational
Theory
43 (Winter 1993): 23-32.
Gardenswartz,
D. "Public Education: An Inner City Crisis!
Single-Sex Schools: An
Inner City
Answer?"
Emory LawJournal
42 (1993): 591-646.
Gutmann, A. "Civic
Education and Social Diversity."
Ethics 105 (April 1995): 557-
79.
Henze, R. C., and L. Vanett. "To Walk in Two Worlds-or More? Challenging a
Common Metaphor of Native Education." Anthropology
and Education
Quarterly
24, no. 2 (1993): 116-34.
Hollinger, D. Postethnic America:
Beyond
Multiculturalism. New York:
Basic, 1995.
Karp, S. "Is All Black and All Male All Right?" ZMagazine, June 1991, pp. 87-91.
Kekes, J. The Morality of Pluralism. Princeton, N.J.: Princeton University Press,
1993.
Kozol,J. Savage Inequalities:
Children in America's Schools. New York:
Crown, 1991.
Kukathas, C. "Are There Any Cultural Rights?"
Political
Theory
20, no. 1 (1992):
105-39.
Kymlicka,
W. Liberalism,
Community
and Culture. Oxford: Oxford University Press,
1989. (a)
Kymlicka, W. "Liberal Individualism and Liberal Neutrality." Ethics 99, no. 4
(1989): 883-905. (b)
Kymlicka, W. "The Rights of Minority Cultures: Reply to Kukathas." Political
Theory
20, no. 1 (1992): 140-46.
498 American Journal of Education
McDonough
Kymlicka,
W. Multicultural
Citizenship:
A Liberal
Theory of Minority Rights.
Oxford:
Clarendon, 1995.
Lizhi, F., and P. Link. "The Hope for China." New York
Review of Books, Oc-
tober 17, 1996.
Longboat, D. "First Nations Control of Education: The Path to Our Survival as
Nations." In Indian Education in Canada. Vol. 2, The Challenge,
edited byJ. Bar-
man, Y. M. Hebert, and D. McCaskill. Vancouver: University of British Colum-
bia Press, 1987.
McClaughlin, T. H. " 'Education For All' and Religious Schools." In Education
for
a Pluralist
Society:
Philosophical
Perspectives
on the Swann
Report,
edited by G. Hay-
don. Bedford Way
paper no. 30, University of London, Institute of Education,
1987.
McClaughlin, T. H. "The Ethics of Separate Schools." In Ethics,
Ethnicity
and Edu-
cation,
edited by M. Leicester and M. Taylor,
pp. 114-36. London: Kogan Page,
1992.
McDonough, K. "Minority Rights, Liberal Autonomy, and North American In-
dian Schools." Ph.D. dissertation, University of Illinois at Urbana-Champaign,
Department of Educational Policy Studies, 1994.
McDonough, K. "Cultural Recognition, Cosmopolitanism, and Multicultural
Education." In Philosophy of Education 1997, edited by L. Stone. Urbana, Ill.:
Philosophy of Education Society, 1998.
Nozick, R. Anarchy,
State and Utopia.
New York:
Basic, 1974.
Nussbaum, M. "Patriotism and Cosmopolitanism." In For Love
of Country: Debating
the
Limits
of
Patriotism,
edited byJ. Cohen. Boston: Beacon, 1996.
Olneck, M. "Terms of Inclusion: Has Multiculturalism Redefined Equality in
American Education?" AmericanJournal of
Education 101 (May 1993): 234-60.
Rawls,J.
A Theory ofJustice.
Cambridge, Mass.: Harvard University Press, 1971.
Rawls,J.
Political Liberalism.
New York:
Columbia University Press, 1993.
Raz, J. "Multiculturalism: A Liberal Perspective." Dissent 38 (Winter 1994):
67-79.
Richardson, B. People
of Terra Nullius. Toronto: Douglas & MacIntyre, 1993.
Rushdie, S. "Imaginary
Homelands." In Imaginary
Homelands:
Essays
and Criticism,
1981-1991. New York:
Penguin, 1991.
Simonelli, R. "Indian Education in America: Summarizing a Collection of Essays
by Vine Deloria,Jr." Winds
of Change
(Summer 1991): 12-17.
Tamir, Y.
Liberal Nationalism.
Princeton, NJ.: Princeton University Press, 1995.
Taylor, C. Multiculturalism and "The Politics
of Recognition.
"Princeton, N.J.:
Prince-
ton University Press, 1994..
Tomasi,John. "Kymlicka,
Liberalism, and Respect for Cultural Minorities."
Ethics
105, no. 3 (1995): 580-603.
Waldron, J. "The Theoretical Foundations of Liberalism." In Liberal
Rights:
Col-
lected
Papers,
1981-1991. Cambridge: Cambridge University Press, 1993.
Waldron,J. "Minority Cultures and the Cosmopolitan Alternative." In The
Rights
of Minority Cultures,
edited by W. Kymlicka. Oxford: Oxford University Press,
1995.
Waldron,J. "Multiculturalism and Melange." In PublicEducation in a Multicultural
Society: Policy, Theory, Critique,
edited by R. Fullinwider. Cambridge: Cambridge
University Press, 1996.
August 1998 499
