1. Top of page

The issue of culture2 was outside the purview of traditional mainstream liberalism. One of the core assumptions of liberalism is that we are first and foremost free, rational and equal individuals and should be treated as such. The pervasiveness of cultural influence in all aspects of our lives that marks us as members of distinct cultures was by and large set aside by liberalism in its preoccupation with “universal” traits shared by all of us. This dominant trend in liberalism, however, has been challenged in recent decades by some liberals who advocate multiculturalism which aims to protect diverse minority cultures in the Western context. Will Kymlicka's works in particular have been considered as the most sophisticated and systematic liberal treatment of multiculturalism.3 In this new phase ofliberalism, Kymlicka and other multicultural liberals acknowledge the significance of culture and contend that multiculturalism is not only consistent with, but is in fact entailed by the mainstream liberal emphasis on autonomy.4

I welcome this development as a move in the right direction within liberalism and fully acknowledge Kymlicka's enormous contribution to bringing liberals' attention to an important, yet hitherto ignored, issue of multiculturalism. In this article, however, I shall argue that an uneasy tension exists in Kymlicka's conjunction of liberalism and multiculturalism, which generates inconsistent multicultural recommendations for “national minorities” and immigrants, the two main categories of cultural diversity in the West. I shall attempt to uncover the source of such inconsistencies in Kymlicka's liberal multiculturalism by carefully analyzing reasoning steps involved in Kymlicka's conceptual marriage of liberalism and multiculturalism. I shall argue that the root of the problem can be traced to Kymlicka's usage of “freedom,” the key concept that forms the indispensable bridge between liberalism and multiculturalism. In particular, I shall argue that Kymlicka's recognition of only one sense of freedom, although at least two distinct senses of freedom are involved, in his construction of liberal multiculturalism lies at the source of the problem.


  1. Top of page

The distinctly liberal justification for multiculturalism, according to Kymlicka, is based on the assumption that the flourishing of culture is a necessary “precondition” for individual freedom. The freedom of choice presupposes that there are meaningful alternatives to choose from, and meaningful choices can be made only against a backdrop of a certain cultural context in which we are deeply immersed. In other words, freedom can be meaningfully exercised only within a “societal culture,” whose shared vocabulary we possess through earlier inculcation and habituation, that “provides its members with a meaningful way of life across the full range of human activities, including social, educational, religious, recreational, and economic life, encompassing both public and private spheres.”6

Societal culture is valued in liberalism not in itself, but only because it provides enabling conditions for the autonomy of its members.7 However, this recognition of the significance of societal culture for the exercise of individual freedom is taken to provide a decisive justification for the liberal endorsement of multiculturalism. This is so because it reveals the predicament faced by members of minority cultures residing, for one reason or another, in liberal Western societies. Adult members of minority cultures are deprived of their own societal culture. Yet they do not have easy access to the liberal culture of the host country because the “complexity and the density of [the dominant culture's] details defy explicit learning or comprehensive articulation.”8 Under these circumstances, securing a favorable condition for exercising autonomy of these individuals requires multicultural measures to sustain their own societal culture within the Western context. Therefore, the endorsement of multiculturalism is entailed by liberalism that advocates individual autonomy.

In order to enable minority members to sustain their societal cultures in the dominant liberal society, liberal multiculturalism would consist in granting various “group-differentiated” rights to minority cultures. Kymlicka distinguishes between two broad categories of cultural diversity in the West, “national minorities” and immigrant ethnic groups. National minorities are groups of people who have a territorial base, share a societal culture, and have a sense of common “national” identity, but for various reasons find themselves incorporated into an alien majority culture, often against their will. Typically they aspire to maintain the survival of their distinct culture through various forms of self-government. Immigrants, on the other hand, are people who voluntarily come to the West and do not in general wish to establish a separate and self-governing nation.9 While they may want greater recognition of their cultural identity, they are not in principle opposed to integrating into the society at large.

Group-differentiated rights appropriate for each group differ accordingly. National minorities may justifiably enjoy, Kymlicka argues, powerful self-government rights that will protect them from interventions by the larger society. They should be able to form an independent political unit with a separate societal culture and enjoy “sovereign powers as a matter of legal right,” entitled to exclude the central government from reclaiming such powers.10 Immigrants, on the other hand, should be granted “polyethnic rights,” intended to enable immigrant ethnic groups to “express their cultural particularity and pride” while allowing them to fully participate in the larger economic and political contexts. Such rights would entitle them to affirmative action programs, a certain number of seats in the legislature or government advisory bodies, the revision of history and literature curricula within public schools to give greater recognition to historical and cultural contributions of immigrant groups, the accommodation of immigrant religious holidays, antiracism educational programs, funding for ethnic studies programs, and bilingual education programs for their children, among others. The primary purpose of these rights, though, is to aid the “integration” of these groups into the larger society by providing them with fairer terms.11

There are, however, limits to accommodating minority cultural groups in liberal societies. While Kymlicka advocates strong group autonomy that precludes interventions by the dominant society with respect to national minorities, Kymlicka and other liberal multiculturalists are quite clear, in conformity with the traditional liberal line that sought the assimilation of cultural minorities, that they are not endorsing the value of all cultures in their advocacy of multiculturalism. From the liberal perspective, some minority cultures are clearly “illiberal” or “inferior.”12 Since multicultural rights for cultural minorities are to be endorsed only insofar as they are “consistent with respect for the freedom or autonomy of individuals,”13 many liberal multiculturalists seem to agree, especially with respect to immigrant cultural minorities, that the dominant liberal society may be selective in granting group-differentiated rights, making sure that these rights are not used to inhibit the autonomy of individual members of such groups.14 In other words, the basicposition of liberal multiculturalism seems to be that for some inferior and/or oppressive immigrant minority cultures, “liberalization” may not be an entirely illegitimate imposition, provided that this is done gradually.15 Although “tolerance” is considered a core liberal value, liberalism does not endorse limitless tolerance. “What distinguishes liberal tolerance is precisely its commitment to autonomy.”16


  1. Top of page

Yet as long as Kymlicka and other liberal multiculturalists assume that non-liberal cultures can be oppressive to their members and are inferior to liberal cultures in accommodating the freedom of individuals, a perplexing puzzle arises with respect to their multicultural recommendations for cultural minorities. If indeed liberal multiculturalists' position is that the eventual liberalization of some non-liberal minority cultures is a legitimate liberal goal, consistent with the traditional liberal stance, why opt for multiculturalism that advocates quite extensive group-differentiated rights?

As has been stated, Kymlicka believes that national minorities, unlike immigrants, should be treated much like foreign sovereign nations and left alone, even if they are non-liberal. To impose liberal principles on national minorities would be “a form of aggression or paternalistic colonialism.” If certain national minorities decide to reject liberalism in favor of their non-liberal culture, then there is no choice for the liberal majority but to “learn to live with this, just as [it] must live with illiberal laws in other countries.”17 This advocacy of strong self-government rights for national minorities seems to be inconsistent with the basic liberal position, especially when societal cultures of such national minorities are non-liberal. If the right of individuals to make their own choices is a core liberal value, and if liberal multiculturalists endorse the flourishing of minority cultures only to the extent that they are “themselves governed by liberal principles,”18 then, it seems reasonable to argue, cultures that do not respect their members' freedom of choice are, in Kymlicka's own words, acting “unjustly” and therefore should not be tolerated, even if they happen to be national minorities.19

Kymlicka may defend his advocacy of the group autonomy of national minorities, though, by claiming that they have a land base on which a reasonably well functioning self-government is in place, although in most cases publicly unrecognized. Therefore, he may continue, intervention by the dominant society would be tantamount to “aggression” or “paternalistic colonialism.” In other words, the central government “cannot ‘reclaim’ the powers possessed by the federal subunits, because those powers never belonged to the central government” in the first place.20 This may be true of relatively well-established federalisms in Canada and Switzerland, for example. However, with respect to most colonized national minorities in other liberal Western states—for example, Native Americans in the U.S.—the current status of their “nations” is quite precarious and at the mercy of the central government, whatever its history may have been. Hence, it is perfectly consistent for liberals to ask this question: Since it is the dominant liberal society that grants such multicultural rights to national minorities to begin with, why should it give up its prerogative to revoke them in extreme situations that involve national minority groups imposing internal restrictions on their members?

The advocacy of polyethnic rights for immigrants also generates quandaries for Kymlicka. Most contemporary immigrant groups come from non-liberal Third World countries, and Kymlicka's polyethnic rights are meant precisely for such groups. Contrary to Kymlicka's facile assumption that most ethnic minorities “share the basic liberal principles,”21 however, many societal cultures of non-Western immigrant groups do not advocate liberalism, as they do not affirm the inalienable rights of individuals to make choices concerning values and projects solely on the basis of their own individual judgments. Indeed, many, if not most, societal cultures of contemporary immigrant groups discourage such an individualistic notion of autonomy because it entails the critical scrutiny of constitutive values of their societal culture, which are often viewed as “an ancestral inheritance to be cherished and transmitted as a matter of loyalty to their forebears.”22 In other words, these cultures advocate communitarian views of the self that Kymlicka unequivocally repudiates as inveterately “illiberal.”23

This said, it is not clear why the dominant liberal society should let the members of non-liberal immigrant groups preserve their communitarian ideals by giving them government funding at public schools to educate their youngsters about their non-liberal homeland societal culture and language. The public funding of ethnic studies programs and revisions of history and literature curricula in public schools to advertise the values of minority cultures, included among Kymlicka's polyethnic rights, perpetuate cultures that are perhaps better “extinct” from the liberal perspective, as they adhere to communitarian ideals that Kymlicka insists are at odds with liberalism. If liberal multiculturalism endorses multicultural rights of immigrant cultural groups only to the extent that they conform to the liberal ideal of individual freedom, and most contemporary immigrant minority cultures are non-liberal, imposing internal restrictions on their members,24 why allow them to propagate and flourish by offering various polyethnic rights?

Perhaps the model of multiculturalism most consistent with this liberal intuition would be what Spinner calls “pluralistic integration.”25 This model takes the liberal principles as the unifying principles of social organization and demands that citizens of a liberal society “learn to accept [only those] ethnic practices that are compatible with liberalism.” The authority to judge whether a culture with “mysterious practices” is compatible with liberalism lies with the liberal state itself. “Practices that violate liberal principles should be either discouraged or forbidden, even if they are not completely understood by the larger political community.” In this setting, the “weakening [of] ethnic ties” is to be expected, and the liberal society, “[i]nstead of trying to become a culturally plural society,” should strive toward pluralistic integration which maintains its distinct liberal political structure.26 A pluralistically integrated society, then, seems ultimately incompatible with true cultural diversity. Instead, it seems like a modified form of traditional liberal assimilationism.27 If this pluralistically integrated model is the vision of a multicultural society most consistent with liberalism, then why has Kymlicka departed from this “true” liberal path?


  1. Top of page

I believe the reason has to do with the way in which Kymlicka has constructed his liberal multiculturalism. The philosophical union of liberalism and multiculturalism relies on a concept that functions as the main bridge in this connection, freedom. On the one hand, the core value of liberalism is an individual's freedom to choose any end solely on the basis of her own critical judgment. This sense of freedom I shall call “liberal autonomy,” and this is what Kymlicka explicitly endorses. On the other hand, Kymlicka professes to value culture because it is conducive to the members' exercise of freedom. Culture is indeed conducive to the members' exercise of freedom, but only in the sense that it provides a context in which members can make meaningful choices. In order for Kymlicka's argument to be consistent, the concept of freedom has to be unified in both contexts of individual and culture. There is no indication that Kymlicka doubts the consistency of his liberal multiculturalism. I shall argue, however, that Kymlicka's liberal multiculturalism is not consistent because the sense of freedom relevant to the context of culture, which I shall call “generic valuational agency,” is not equivalent to liberal autonomy. I believe that Kymlicka, by failing to recognize these two distinct senses of freedom, unwittingly equivocates on the notion of freedom in his construction of liberal multiculturalism. In what follows, I shall elaborate on these two distinct senses of freedom and the ways in which they are conflated in Kymlicka's construction of liberal multiculturalism.

1. Liberal Autonomy

The notion of autonomy advocated by Kymlicka refers to freedom to choose any conception of good according to one's individual judgment without undue external influence. It encompasses two “preconditions for leading a good life,” according to Kymlicka. In order to lead a good life, first, we ought to live our life “from the inside” according to our beliefs about what is valuable in life, and, second, we ought to be free to “question those beliefs, to examine them in light of whatever information, examples, and arguments our culture can provide.” The reason why liberalism is a superior political system, as well as a way of thinking, is because it allows each individual to maximize the chances of leading a good life, understood in this sense, by granting “a very wide freedom of choice in terms of how they lead their lives.” In particular, liberalism “allows people to choose a conception of the good life, and then allows them to reconsider that decision, and adopt a new and hopefully better plan of life.”28

A fundamental assumption presupposed here is that “revising one's ends is possible, and sometimes desirable, because one's current ends are not always worthy of allegiance.”29 In other words, we are not only capable of “detaching” ourselves from any particular values and ends previously adopted, but, should they turn out to have been unduly imposed from without, which is the case with most of our values and ends, we ought to be willing to review them critically, ready to revise or reject them. How extensive and thorough should be the critical review? In one of his later works, Kymlicka bluntly states that “Nothing is ‘set for us’.” This implies that, in principle, no value or end is exempt from an individual's critical scrutiny and revision/rejection.30 Two main characteristics, among others, of liberal autonomy can be identified from this account. First, it is individualistic and, second, it endorses extensive and thorough scrutiny of values and projects inherited from one's family and community. This second characteristic, in particular, retains the traditional liberal aspiration for the transcendence of contingency.31 Let me elaborate on them in turn.

First, liberal autonomy, conceived in this way, is individualistic. Being “individualistic,” however, does not necessarily imply being self-centered or egotistic. It simply means that, within the moral confines of liberal principles, an individual should have the final authority in the adoption of ends and values, as she reflects on them and determines their content. A liberally autonomous individual should be the final judge who has the sole authority regarding her ends and values in two senses. One sense is that the reasoning process by which an agent adopts ends and values should be in principle a personal and individual process under her control. This, of course, is not to deny that in real life, liberal agents may be influenced by the community or various others, such as family and friends, who have meaningful relationships with her. Regardless, in liberalism, it is ultimately the individual herself who should bear the right as well as the responsibility to conduct the reflective/deliberative process leading to the choice of her values and ends. The community or meaningful others play at best merely an “advisory” role in this process, and the individual herself alone reserves the right to seek or avoid, and, in the former case, to adopt or reject their advice.

Another sense in which liberal autonomy endows an individual with the final authority has to do with the content of ends chosen. Liberal autonomy imposes on an individual no substantive condition regarding her values and ends except that she adhere to liberal moral principles. As long as she is moral in the liberal sense and conforms to her perfect/negative duties,32 liberal autonomy does not require that her chosen ends incorporate communal or relational values. This is not to deny that liberal agents often do incorporate such values in real life. Liberalism certainly allows individuals to opt for community or relation-oriented life plans. The point, however, is that liberal autonomy is perfectly compatible with a life plan that is devoid of such socially oriented values. Exhibiting communal or interpersonal virtues is at best only a conditional obligation.33 Those who decide to pursue a solitary and isolated existence can be liberally autonomous, as their primary moral obligation is to perform their negative duties in accordance with liberal moral principles. As long as they fulfill this obligation, they are free to choose any personal plan of life.

The second crucial characteristic of liberal autonomy endorsed by Kymlicka is the requirement to engage in extensive and thorough scrutiny of one's values,plans, and projects that may lead to radical revisions or rejection of such values and ends. This is not to say that persons who subscribe to liberalism always scrutinize their values and ends in such a scrupulous manner in their actual lives. They mostly do not. However, liberalism not only assumes that they are capable of rigorous and extensive critical examinations of their various values and ends, but further requires them as a precondition for full autonomy. The required scrutiny should be extensive not only in terms of the number of values and ends, but, more importantly, also in terms of the range of values and ends subject to it. That is, we ought to scrutinize for possible revision/rejection not only relatively trivial values and projects, but also those that we take to be constitutive of our identity/self. Perhaps this is just as well, since Kymlicka denies that any value or end is constitutive of the self. For Kymlicka, “No particular task is set for us by society, no particular practice has authority that is beyond individual judgement and possible rejection.”34

This endorsement of extensive scrutiny for possible revision/rejection of values and ends, however, is not to be understood as an advocacy of “freedom for freedom's sake,” Kymlicka cautions. He recognizes that we cannot conceive ourselves as “unencumbered,” and that something must be taken as a “given” in making value judgments. Yet, Kymlicka insists, we have an ability to “detach ourselves from any particular social practice” and to retreat to a neutral standpoint where we can dispassionately and rationally assess various encumbered potential selves available from the outside as alternatives. Practical deliberation, in fact, involves a process of “comparing one ‘encumbered’ potential self with another ‘encumbered’ potential self.” While we must start with a set of givens in such deliberation, no particular potential self or end needs to be taken as a given with the self. In short, we may be encumbered, but we are certainly not constrained by our encumberedness.35

I believe, however, that Kymlicka's above qualification does not amount to much. Kymlicka's emphasis on our being able to imagine ourselves with any different givens as if “nothing is set for us” is not much different from advocating freedom for freedom's sake, if not literally, then surely in practice. A self for whom a given has no power to bind in any way might as well be unencumbered. Whether or not Kymlicka's liberal autonomy is tantamount to freedom for freedom's sake, at least one thing is clear. In order to be liberally autonomous in Kymlicka's sense, an agent must be willing to engage in rigorous and extensive critical self-examinations with respect to any value or end inherited from the outside, which is the case with most, if not all, of our values and ends. In this process, she is required to scrutinize her values and ends, no matter how deeply ingrained they may be, and be willing to revise or reject them without reservation, should they turn out to have been unduly imposed. In this way, a liberal agent may achieve the transcendence of contingent circumstances, including cultural limits.

2. Freedom as Generic Valuational Agency

When Kymlicka discusses the value of culture, the notion of freedom shifts without his acknowledging it. It is no longer liberal autonomy as elaborated above, but now it merely refers to an ability to make choices among meaningful options. Culture is important in liberal multiculturalism because “it is only through having access to a societal culture that people have access to a range of meaningful options.”36 I fully agree. As members are enculturated in a particular societal culture, they assume a particular cultural framework that will enable them to make meaningful choices among a range of feasible encumbered potential selves and viable options available within the culture. However, culture also functions to constrain the scope of meaningful choices.37 One way in which this occurs has to do with how culture circumscribes our imagination of the conceivable. Hence, encumbered potential selves that lie completely outside one's culture would be inconceivable. For instance, to be a businesswoman in corporate America would be an inconceivable option, and hence not an option at all, for a Hmong woman in Southeast Asia who has never left her rural village.

Still, some encumbered potential selves of outside cultures may be conceivable for some people, perhaps due to watching a foreign film or relocating to a new culture. However, as long as attitudes and values presupposed by foreign potential selves are not compatible with their own cultural framework, such potential selves would not be practicable alternatives, however appealing and worthwhile they may be in theory. Let us take the case of an adult Hmong refugee woman relocated to America, whose socialization took place in Southeast Asia. For such a person, becoming a businesswoman in corporate America would be at least conceivable, since she would be exposed to the popular media that portrays such people. However, it is still not a live practicable option for her, not only because acquiring skills, attitudes, and values necessary for becoming a businesswoman in corporate America exceeds the abilities of an ordinary human being brought up in a preindustrial world, but also because such skills, attitudes, and values may be incompatible with her cultural outlook. Indeed, innumerable encumbered potential selves “available” in foreign cultures may not be real options for those enculturated in radically different cultures, not only because cultural barriers, whether linguistic, religious, or cultural, are too great for an effective adoption of a new culture,38 but also because the abdication of one's old cultural framework, if it is at all possible, signifies abandoning a constitutive element of one's identity.

Being embedded in a culture and confined by a culturally specific “horizon of significance,” then, while presenting members with meaningful options, at the same time restricts their options. Being autonomous in the liberal sense, on the other hand, implies that persons are capable of “detaching” themselves from any set of givens and expanding options beyond their contingent circumstantial—including cultural—constraints. If so, culture seriously curtails members' capacity for liberal autonomy. As restricting as culture is, however, the fact that it enables members to make meaningful choices within the cultural context is by no means insignificant. Indeed, what seems to me distinctive about human agents is that they are capable of exercising freedom in this limited sense. I shall call this sense of freedom bound by culture “generic valuational agency” (generic agency, for short), to indicate its compatibility with a wider range of cultures than liberal culture. As I shall later argue in the conclusion, liberal autonomy is only a subset of generic agency.

How is generic agency to be understood? I shall define generic agency as the ability of human agents to embrace certain fundamental values or ideals that they consider superior to all other values—“hyper goods”—and to make qualitative choices with respect to life plans, projects, or ends, in accordance with such values in a consistent manner.39 Hyper goods are values of the highest order that function as the criteria by which we make second-order justifications concerning evaluative judgments. Hyper goods, however, are not some transcendental and universal values that all rational agents can endorse by virtue of their rationality, but rather cultural values circulating in a particular societal culture that persists, while constantly evolving, throughout a sustained period of time in a certain locality.40 Examples of cultural values prevalent in the West are individual freedom, equal opportunity, inalienable individual rights, to name a few. Some examples of cultural values prevalent in more traditional and non-liberal cultures are social harmony, common good, Godliness, reverence for nature, etc.

Cultural values, from which members' hyper goods originate, are in themselves highly general and abstract values or ideals in need of fuller elaboration in order to serve as practical guides to those who embrace them. Hence, by the time they are adopted as personal hyper goods, they must be wedded to a certain interpretation, ranging from unsystematic and elementary to highly structured and sophisticated, depending on the level of maturity and intellectual acuity of the subscriber. Multiple interpretations may be available in a culture with respect to any set of hyper goods that subscribers can adopt and elaborate on, but all such interpretations would be culturally specific, as they are rendered in a specific language, predicated on a culturally specific horizon of significance. Since hyper goods, interpreted in a certain way, structure our lives and define who we are, our culture, inscribed in hyper goods interpreted in a culturally specific way, is constitutive of our identity. Culture, then, is not a non-essential baggage that we can unload as we choose.

The cultural dimension of identity, however, is by no means predicated on a monolithic cultural essence branded in the identities of cultural insiders, as some liberals allege.41 Culture is ineluctably hybrid due to various cultural influences, although the specific mode of hybridity may vary with locality depending on its indigenous traditions and the manner of interaction with other cultures. This distinct hybrid cultural plexus typically associated with a specific locality is taken by the members as their unique societal culture. Culture is also complex, containing not only multiple and potentially conflictual cultural values but also their numerous interpretations. Some cultural values may be more prevalent or pronounced, even promoted as the “official” cultural values, with concomitant institutional support. Even with such prevalent cultural values, multiple interpretations may be vying for dominance within the culture. As a result, configurations of the cultural identity of the members may vary as they subscribe to different sets of available cultural values and interpretations.

Having clarified the intimate connection between culture and hyper goods, we are in a better position to see the distinctiveness of generic agency in relation to liberal autonomy. Generic agency, as the capacity to embrace certain cultural values as one's personal hyper goods and to make qualitative choices in accordance with them in a consistent manner, diverges from liberal autonomy in both aspects that characterize the latter, namely its individualism and the requirement of extensive and thorough critical scrutiny of ends and values.

First, generic agency in most cultures, with a notable exception of the modern West, is not predicated on the final authority of an individual, whether in conducting the reflective and deliberative process by which the agent adopts or in the content of her values and ends. Cultural values of many, if not most, non-Western cultures are communally oriented and encourage the members to engage actively in communal institutions and practices that uphold such values. As members participate in such collective deliberative processes, the community and meaningful others would play a significant role in the formation and adoption of their personal ends and values. Such external influences from others, however, are not necessarily coercive, for members may voluntarily seek and embrace them, whether out of deference, admiration, or respect. Also, the content of the members' own ends and values are typically communal and relational, since cultural values that members adopt as their personal hyper goods tend to promote the well-being of communal and interpersonal entities.

Second, and more importantly, extensive critical self-examination is not a requirement of generic agency. This is the respect in which generic agency crucially differs from liberal autonomy that requires extensive and critical reflection and deliberation concerning one's ends and values, including hyper goods, and thereby urges us to overcome the boundary imposed on our freedom by our specific culture. The most important aspect of generic agency is its valuational aspect that involves the commitment by an agent to a certain set of culturally embedded hyper goods that she deems as unqualifiedly superior to all other values. Once this commitment is made, hyper goods, interpreted in a certain way, are typically secure in their foundational status and exempt from critical and extensive scrutiny. Accordingly, generic agency is decidedly limited by a particular cultural horizon, as hyper goods are themselves cultural values interpreted in a culturally specific way.

This does not mean, though, that agents who exercise generic agency would not engage in any kind of reflection or deliberation concerning various aspects of their hyper goods or interpretations. Reflective and deliberative capacities are indeed uniquely human capacities, and humans have been exercising such capacities regardless of time and place. Extensive critical scrutiny required by liberal autonomy is not the only way to employ human reflection and deliberation. There are at least two aspects of generic agency that require reflection or deliberation. One aspect has to do with the fact that personal hyper goods are cultural values accompanied by a certain culturally specific interpretation. As such, those who subscribe to a set of hyper goods must have a relatively solid understanding of the interpretation they have adopted and reflect on its various implications. The other aspect has to with the application of hyper goods in everyday decision-making. As agents attempt to evaluate, plan and act in a manner consistent with hyper goods, some introspection and deliberation are surely necessary. In a more communal environment, agents may reflect or deliberate about them in conjunction with others that they trust and respect.

Some liberal critics may be concerned about the relative unimportance of extensive critical scrutiny in generic agency. They may argue that the requirement of extensive critical self-examination of liberal autonomy is necessary in order to prevent agents from subscribing to immoral or otherwise problematic cultural values, especially rampant in non-liberal cultures. As I have argued elsewhere, however, this worry is often predicated on unwarranted negative stereotypes of non-liberal cultural values than on reality. Most non-liberal cultural values of Third World origin are typically respectable moral values, entailing certain basic moral injunctions concerning the treatment of others, if rightly interpreted.42 Still, it is undeniable that some cultural values, whether in liberal or non-liberal cultures, may not be respectable moral values. “Purity of the German Race,” embraced by some Germans around the Second World War, is a case in point. Even cultural values that are respectable may be wedded to an inconsistent, distorted, or outright immoral interpretation, leading the subscriber to an immoral path. “Godliness” wedded to the interpretation, “Godliness requires you to either forcefully convert or eliminate pagans from the face of the Earth” is such an example. For those whose hyper goods are problematic in such ways, it may seem that exercising generic agency involves acting against basic moral injunctions.

Not so. Human agents with average reasoning capacities and a normal emotional constitution would have a robust moral sense that would generate in them emotional and/or psychological disturbances when they recognize problematic implications of either their hyper goods or interpretations. Even in cases where immoral actions are not directly entailed, agents may no longer feel content with their original hyper goods or interpretations due to transformational experiences that change their understanding of themselves and the world. As generic agency is predicated on a wholehearted and sincere conviction that their hyper goods and interpretations are indeed superior to all other values and interpretations, generic agency requires in such cases deeper reflection as well as more extensive deliberation about one's current hyper goods and interpretations, at least until the equilibrium of their moral sense is restored. In most cases, the appeasement of the moral sense may be achieved by revisions of various degrees at the interpretation level. However, the process of deeper reflection and more extensive deliberation may at times lead agents to a supersession of not only interpretations but also hyper goods.43 Liberals may take such incidents as exemplifying cultural transcendence and posit them as desired end-states in the exercise of liberal autonomy. In the exercise of generic agency, though, they are mere incidental eventualities that may or may not occur, as agents attempt to find a set of hyper goods and interpretations to which they can unqualifiedly commit themselves. Also, such cases do not emblematize cultural transcendence, as the new set of hyper goods or interpretations replacing the old set remain culturally specific. Cultural transcendence is a liberal myth.

So far, we have been focusing on agents with average human capacities who will register and respond appropriately to warning signs emitted by their well-functioning moral sense. Unfortunately, as history has shown time and time again, there will always be a minority of hard-core “fanatics” who are committed to extreme hyper goods or interpretations without reservation, ready to jump at an opportunity to act on their poisonous ideals. Many such people are undaunted by obviously immoral implications of their hyper goods, as their moral sense is either non-existent or emaciated to the extent that it is no longer functioning properly. Liberal critics may allege that my position is at its weakest with regard to such people because it encourages embracing hyper goods without critical self-reflection. I will admit that my position does not have a solution to “convert” such fanatics. But let us be clear here that we are now focusing on those whose moral sense is somehow faulty. If so, I do not think that the liberal position fares any better. With respect to humans whose moral sense is malfunctioning, whether due to genetic or environmental causes, I doubt that the liberal admonition to engage in thorough and extensive self-examination for its own sake would make much difference.

Although this is a difficult problem, I believe that my position does provide an effective way to deal with such fanatics, aside from converting them. The solution lies in internal democracy entailed by my conception of culture. In long-standing cultures that are complex and emergent, multiple cultural values and their interpretations abound. Hence, members are unlikely to agree on a single set of cultural values as their common hyper goods. Even when a large majority happens to endorse a single common set of hyper goods, their interpretations may differ. Under these circumstances, cultural identities of the members would vary and their political, social, or economic views will diverge as well. If all culturally embedded insiders are allowed to participate in cultural discourses to negotiate such differences in peace, then their cultural system is bound to shift toward egalitarianism and inclusion. I have elsewhere referred to this as the “dynamic” dimension of cultural membership, and its realization calls for robust internal democracy, understood as a process of equal participation of culturally embedded and self-identified members in internal contestations and negotiations concerning cultural values and a cultural system to actualize them.44 In a culture that values democracy in this sense and embodies it in its institutions, fanatics with extreme views will have a hard time winning supporters. Even if they happen to achieve power temporarily, it would be difficult for them to maintain power in true democracy. Hence, although my position does not solve the problem of fanatics at the individual level, it does offer a solution at the communal level.


  1. Top of page

Human agents as cultural beings are free to the extent that they exercise generic valuational agency. Being free in this sense, however, is not to be liberally autonomous. Subjecting most of one's ends and values, including hyper goods, to extensive and critical scrutiny and thereby transcending one's culture is not only unrealizable for most, if not all, human agents but, more importantly, unnecessary for moral agency. Then, questions arise as to where liberal autonomy stands. Is liberal autonomy somehow above and beyond generic agency? Is it really possible for members of liberal cultures to transcend their cultural embeddedness by exercising liberal autonomy?

Answers to both questions are in the negative. Let me begin with the first question. Liberal autonomy is not above and beyond but rather a species of generic agency. The exercise of liberal autonomy, much like generic agency, is predicated on embracing, whether explicitly or implicitly, a certain cultural value, interpreted in a certain way, as hyper good beyond scrutiny—the liberal value of individual freedom interpreted as having and exercising the capacity for liberal autonomy—and making qualitative choices in conformity with it. From the core value of individual freedom, interpreted as liberal autonomy, other liberal political values, such as civic equality, equal opportunity, and inalienable individual rights are derived as measures to protect the former. The collection of these forms a set of cultural values upheld by a large number of members of liberal societies as their personal hyper goods. Evidence for the foundational status of liberal autonomy as hyper good can be found in this paradox: If liberal autonomy requires rigorous critical scrutiny of all kinds of ends and values, then the value of liberal autonomy itself must be subjected to such a humbling procedure. However, liberals are averse to do so, as such a practice would undermine the primary status they attribute to liberal autonomy. To the extent that liberals uphold liberal autonomy as the foremost “universal” ideal beyond critical scrutiny, it qualifies as their hyper good.

Though a species it is, liberal autonomy is conceptually distinct from generic agency due to its paradoxical nature. Although it is itself a cultural value, it endorses critical self-reflection by the subscriber that would supposedly enable her to transcend contingent external constraints, including cultural limits. It is self-contradictory and is therefore impracticable. But why do many members of liberal societies, including liberal theorists, advocate it as a feasible ideal? The reason is that the advocacy for liberal autonomy, as ideological as it is, is entrenched in the public culture of liberal societies and, as is the case with every culturally entrenched ideology, misleads its subscribers to believe that liberal autonomy is practicable. Among those who embrace liberal autonomy as their hyper good in liberal societies, most, if not all, do so, not because their rational and critical reasoning has proven it to be feasible or superior to other cultural values, but rather because they find it “familiar and attractive” in the process of participating in common liberal public institutions that endorse its fundamental value.45 The flip side of this, of course, is that those who are not enculturated in liberal societies would not find liberal autonomy as familiar and attractive. Upholding liberal autonomy and attempting to live by it, then, does not exemplify cultural transcendence but rather cultural embeddedness in a liberal culture, despite its rhetoric.46

Having clarified the relation as well as differences between liberal autonomy and generic agency, let me conclude this article by summing up my main argument: The conception of freedom to which Kymlicka subscribes is liberal autonomy, and he thereby maintains his allegiance to mainstream liberalism. Kymlicka advocates “liberal” multiculturalism by arguing that minority cultures are conducive to the freedom of their members. Cultures in general are conducive to their members' exercise of freedom, but only in the sense that they provide a horizon of personal hyper goods that are essential to making meaningful options. I have referred to this sense of freedom as generic valuational agency. In most cultures, with the exception of contemporary liberal cultures, hyper goods that anchor generic agency are incompatible with liberal autonomy. Hence, the equivalence between liberal autonomy and the freedom of choice enabled by culture, necessary for the success of Kymlicka's union of liberalism and multiculturalism, turns out to be spurious.

  • 1

    I wish to thank Marilyn Fischer and Paul Benson not only for their helpful comments on previous drafts of this article but also for their encouragement and moral support while the theme of this article was first being conceived and developed.

  • 2

    “Culture” as used here refers to a comprehensive way of life found in a certain locality that has endured over time, predicated on a shared language, value/religious system, and history. Typically, one is a member of a culture not out of choice but by birth, although one may become a member through marriage or adoption. I shall discuss it further in section IV.2. Will Kymlicka's “societal culture” and Margalit and Raz's “pervasive” culture are similarly used. Will Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford: Clarendon Press, 1995) 76; Avishai Margalit and Joseph Raz, “National Self-Determination,” Journal of Philosophy 87 (1990): 444; see also, Joseph Raz, “Multiculturalism: A Liberal Perspective,”Ethics in the Public Domain: Essays in the Morality of Law and Politics (New York: Clarendon Press, 1994) 176–77.

  • 3

    For this reason, my primary focus is on Kymlicka, although I shall refer to other liberal theorists when relevant. For examples of liberal theorizations of multiculturalism, see Will Kymlicka, Liberalism, Community, and Culture (Oxford: Clarendon Press, 1989); Kymlicka (1995); Will Kymlicka, “Minority Nationalism and Multination Federalism,”Politics in the Vernacular: Nationalism,Multiculturalism, and Citizenship (New York: Oxford UP, 2001a); Will Kymlicka, “The Theory and Practice of Immigrant Multiculturalism,”Politics in the Vernacular: Nationalism, Multiculturalism, and Citizenship (New York: Oxford UP, 2001b); Raz (1994); Jeff Spinner, The Boundaries of Citizenship: Race, Ethnicity, and Nationality in the Liberal State (Baltimore, MD: Johns Hopkins UP, 1994); Jeff Spinner-Halev, “Feminism, Multiculturalism, Oppression, and the State,” Ethics 112 (2001): 84113.

  • 4

    Kymlicka (1995): 75.

  • 5

    Liberal multiculturalism is a term proposed by Raz (1994). Kymlicka himself does not use this term.

  • 6

    Kymlicka (1995): 76, see also footnote 2.

  • 7

    Ibid: 83.

  • 8

    Raz (1994): 177; see also, Kymlicka (1995): ch. 5, sect. 4.

  • 9

    For critiques of Kymlicka's liberal multiculturalism and Kymlicka's response, see Will Kymlicka, “Do We Need a Liberal Theory of Minority Rights? Reply to Carens, Young, Parekh and Forst,” Constellations 4 (1997): 7287. , in particular, criticizes Kymlicka's dichotomous categorization of cultural diversity in the West for assuming the “voluntariness” of immigrants.

  • 10

    Kymlicka (1995): 167; for a similarly strong position on national minorities, see also Spinner-Halev (2001): sect. V. For more on national minorities, see Kymlicka (1995): 10–13, 30, 79–80, 100–01; Kymlicka (2001a): 95; Will Kymlicka, Contemporary Political Philosophy (New York: Oxford UP, 2002) 330.

  • 11

    On polyethnic rights, see Kymlicka (2001b): 163–65 and Kymlicka (1995): 31. For more on immigrants, see Kymlicka (1995): 31; see also Kymlicka (2001b): 160–61; Kymlicka (2002): 168.

  • 12

    Raz (1994): 185. Although Raz criticizes the view that judges other cultures as “inferior” on p. 183, he himself employs the adjective as he discusses “oppressive,” “repressive,” or “inferior without being oppressive” cultures (p. 185).

  • 13

    Kymlicka (1995): 75, see also p. 153; Spinner (1994): 62. Raz (1994) also states that liberal multiculturalism recognizes and respects cultures “because and to the extent that they serve true values” (p. 183, emphasis added).

  • 14

    Kymlicka (1995) argues that multicultural rights are to be used for the sole purpose of the “external protection” of minority cultural groups from interventions by the dominant society, and not for the purposes of the “internal restriction” by the groups of their own members. Internal restrictions of the members can sometimes be condoned, but only as “unavoidable by-products” of an effort to protect the group from external incursions (p. 44).

  • 15

    See Raz (1994): 182–84; Kymlicka (1995): 75, 77, 94, 95, 153, 170; Kymlicka (2002): 168, 352; see also Spinner (1994): ch. 4. David Miller's republicanism is another example in which the liberalization of minority cultures is advocated. See David Miller, “Group Identities, National Identities, and Democratic Politics,” Citizenship and National Identity (Malden, MA: Polity Press, 2000): 62–80.

  • 16

    Kymlicka (1995): 75, 153, 158. For more on liberal constraints of multiculturalism, see, Kymlicka (1995): 94–95; Kymlicka (2002): 352; Raz (1994): 183–84.

  • 17

    Kymlicka (1995): 167.

  • 18

    Ibid: 153.

  • 19

    Ibid: 168. Such a view is explicitly endorsed by an influential deliberative democrat Amy Gutmann, Identity in Democracy (Princeton, NJ: Princeton UP, 2003) ch. 2.

  • 20

    Kymlicka (1995): 167; Kymlicka (2001a): 95.

  • 21

    Kymlicka (2002): 339, cf. 338, 343; Kymlicka (1995): 153. At other times, Kymlicka himself admits that some national minorities' cultures are indeed illiberal. Kymlicka (1995): 38, 43, 165; see also, Kymlicka (2002): 340.

  • 22

    Bhikhu Parekh, “Dilemmas of a Multicultural Theory of Citizenship,” Constellations 4 (1997): 59.

  • 23

    Kymlicka (1995): 163; Kymlicka (2002): ch. 6, sect. 6.

  • 24

    Will Kymlicka, “Liberal Complacencies,” Is Multiculturalism Bad for Women? ed. Joshua Cohen, Matthew Howard, and Martha Nussbaum (Princeton, NJ: Princeton UP, 1999) 32; see also Spinner (1994): 69.

  • 25

    In fact, Kymlicka endorses Spinner's pluralistic integration when he talks about immigrant multiculturalism, see Kymlicka (2001b): 168. Interestingly, Spinner-Hale (2001) advocates quite extensive “group autonomy” when it comes to national minorities. See footnote 10.

  • 26

    Spinner (1994): 62, 72 (emphases added), 175. Spinner does not see the necessity of providing multicultural rights such as bilingual education or the revision of school curricula (pp. 175, 177), which are among Kymlicka's polyethnic rights. Hence, Spinner's pluralistically integrated society would be somewhat different from Kymlicka's “polyethnic state,” Kymlicka's endorsement of pluralistic integration notwithstanding.

  • 27

    Avery Gordon and Christopher Newfield, “Multiculturalism's Unfinished Business,” Mapping Multiculturalism, ed. Avery Gordon and Christopher Newfield (Minneapolis, MN: U of Minnesota P, 1996) 81; Parekh (1997): 60.

  • 28

    Kymlicka (1995): 81.

  • 29

    Ibid: 82, italics added.

  • 30

    Kymlicka (2002): 224; see also p. 221. One might point out that Kymlicka's (2002) book, in which he gives a fuller account of liberal autonomy, is written as a textbook and that Kymlicka does not formally identify his own view. However, Kymlicka explicitly recognizes that his book is meant as “an introduction to the best answers we have” to main questions in contemporary philosophy (Preface to the Second Edition, p. xii, emphasis added). Also, the best answers that Kymlicka supports in this book are in line with Kymlicka's own position discussed elsewhere (1995, p. 81).

  • 31

    While some liberals have criticized Kymlicka for advocating overly stringent scrutiny of one's inherited ends (John Christman, “Liberalism, Autonomy, and Self-Transformation,” Social Theory and Practice 27 [2001]: 18890), similarly rigorous self-examination has been endorsed by other liberals, such as Joseph Raz, The Morality of Freedom (New York: Oxford UP, 1986) 382; Jeremy Waldron, “Moral Autonomy and Personal Autonomy,”Autonomy and the Challenges to Liberalism: New Essays, ed. John Christman and Joel Anderson (Cambridge: Cambridge UP, 2005) 315–16; and The discussion of liberal autonomy in this article, however, would not be affected by this debate among liberals, as my focus is on Kymlicka's own conception of autonomy, which lies at the foundation of his liberal multiculturalism. Accordingly, my reconstruction of Kymlicka's conception of autonomy is based on Kymlicka's own statements.

  • 32

    I am following a more traditional interpretation of autonomy by taking liberal autonomy to be a species of moral autonomy, as opposed to those who are moving toward a “personal” conception of autonomy, which is subject to criticisms raised by Sarah Buss, “Valuing Autonomy and Respecting Persons: Manipulation, Seduction, and the Basis of Moral Constraints,” Ethics 115 (2005): 195235.

  • 33

    For example, virtues pertaining to communal and personal relationships can all be accounted for either as special instances of “supererogation” (L. Kohlberg, C. Levine and A. Hewer, Moral Stages: A Current Formulation and a Response to Critics [Basel, Switzerland: S. Karger AG, 1983] 22–27), or as values capable of being promoted as “impartial” requirements for those who choose to enter into such relationships (George Sher, “Other Voices, Other Rooms? Women's Psychology and Moral Theory,” Women and Moral Theory, ed. Eva Kittay and Diana Meyers [Savage, MD: Rowman & Littlefield, 1987] 186–87). This aspect of liberal autonomy has been subject to feminist criticisms by Care Ethicists. See Annette Baier, “The Need for More than Justice,” Science, Morality and Feminist Theory, ed. Marsha Hanen and Kai Nielson (Calgary, AB: U of Calgary P, 1987) 47, 49, 53; see also Virginia Held, Feminist Morality: Transforming Culture, Society, and Politics (Chicago, IL: U of Chicago P, 1993) 187, 212. Proposals for a “relational” conception of autonomy have been made based on such feminist concerns. See Catriona Mackenzie and Natalie Stoljar, “Introduction: Autonomy Refigured,” Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self, ed. Catriona Mackenzie and Natalie Stoljar (New York: Oxford UP, 2000).

  • 34

    Kymlicka (2002): 223–24, emphases added; see also Kymlicka (1995): 163.

  • 35

    Kymlicka (2002): 223, emphasis added; 225.

  • 36

    Kymlicka (1995): 83, emphases added.

  • 37

    See also Hartmut Rosa, “Cultural Relativism and Social Criticism from a Taylorian Perspective,” Constellations 3 (1996): 41.

  • 38

    Kymlicka and other liberal multiculturalists concede this point. See, Raz (1994): 185; Margalit and Raz (1990): 447–49; Kymlicka (1995): 89.

  • 39

    “Hyper good” is Taylor's term. See Charles Taylor, Sources of the Self: The Making of the Modern Identity (Cambridge: Cambridge UP, 1989) 63; see also 27–28, 42–44. My account of generic valuational agency is inspired by Taylor's discussion of “strong evaluation.” See Charles Taylor, “What Is Human Agency,” Philosophical Papers 1 (Cambridge: Cambridge UP, 1985).

  • 40

    In the contemporary context, this locality will typically be a nation. Exceptions do exist in many sub-Saharan African nations, where the societal culture is still tribal. National boundaries are only nominal, as they were arbitrarily imposed by European imperialists at the turn of the last century. Obviously, national minorities are also exceptions.

  • 41

    Gutmann (2003): 39–42, 47–48.

  • 42

    Ranjoo Seodu Herr, “In Defense of Non-liberal Nationalism,” Political Theory 34 (2006): 31920.

  • 43

    For a similar discussion, see Taylor (1985): 39–40 and Taylor (1989): 35–38, 64–65, on “radical visionaries.”

  • 44

    See, Herr (2006): 317–18.

  • 45

    Joshua Cohen, “Democracy and Liberty,” Deliberative Democracy, ed. Jon Elster (New York: Cambridge UP, 1998) 189–90.

  • 46

    For a related discussion, see Taylor (1989): 78–80, 94.