|| 5 || The Distinctiveness of Religious Liberty VÍCTOR M. MUÑIZ-FRATICELLI 1. Introduction The model of religious liberty that has become dominant in contemporary political theory borrows increasingly from the multicultural paradigm. Religious liberty is presented as an individual right, a sub-class of the right to a free individual conscience, while the privileges and exemptions often enjoyed by religious associations – including recognition of the direct claims of authority that they make upon their members – are justified as a necessary or convenient derivation of the individual rights of the congregants, either as concessions of the state or as products of voluntary agreement among members. The model corresponds to a conception of religious freedom grounded on normative individualism that emerges from Enlightenment liberalism and displaces the previously dominant model of freedom of religion – libertas ecclesiae or freedom of the church – that was prevalent in the medieval period.1 The Enlightenment strain of religious freedom is an important historical achievement, one that enormously reduced persecution and sectarian violence and allowed individuals to conscientiously and sincerely abide by their deepest commitments.2 But it is neither correct nor desirable to argue from this historical achievement that a respect for individual conscience is all there is to religious liberty, and thereby to exclude the older strain – the concern for institutional autonomy of religious organizations – from our understanding of freedom of religion. It is not correct, first, because the historical development of religious liberty cannot be adequately explained without reference to its medieval antecedents;3 and second, the current complex of legal institutions that reference religion cannot be understood from a perspective exclusively concerned with individual practitioners.4 And it is not desirable first, because an account of religious liberty that references only the individual conscience severely distorts, at best, the self-understanding of religious adherents and, at worst, demands (coercively or not) that religious doctrine conform to the moral standards of the state;5 and second, because in doing so, an important institutional counterweight to state authority is mollified, a consequence that should concern not only liberals but all opponents of arbitrary power.6 I will not dwell on the normative desirability of a complex conception of religious freedom—that is, one that sustains that the importance of institutional autonomy is perhaps not on par with individual conscience, but is certainly not excluded by any concern for it. My intention is twofold: first, to explain that the wholesale incorporation of the religious liberty paradigm into the multicultural paradigm is an institutional, historical, and conceptual mistake, and that it distorts our understanding of the institutions that enshrine religious liberty and underlie our justification of them. The Western paradigm of religious liberty is a complex product of diverse historical conflicts and political traditions, and only contingently overlaps the multicultural argument.7 The purpose of this essay is to differentiate religious liberty from multiculturalism as theoretical categories, and to at least identify some of the consequences of this differentiation. Religious claims have a different historical origin from multicultural claims, and this historical origin marks both the subjects of religious freedom and the types of claims that it endorses. This origin accounts for the different institutional embodiments of religious freedom and how they differ from those of multicultural accommodation. It also conditions the conceptual or formal structure of the claims made in the name of religious freedom as opposed to those made in the name of multicultural accommodation. No one aspect is necessarily dispositive, but together they show that the Western idea of freedom of religion is more nuanced and complex than ordinarily presented, even by its defenders, and lies partly outside the spectrum of multicultural philosophy and policy. 2. The Multicultural Paradigm There is an obvious challenge to the distinction between multiculturalism and religious liberty as paradigms for the accommodation of difference: these concepts are ambiguous and contested, and they depend, moreover, on the even more ambiguous and contested concepts of culture and religion.8 To make a normative distinction between multiculturalism and religious freedom seems to require that one lay out a theory, or at least a comprehensive definition, of the concepts to which they refer. But this, I suspect, cannot be done with regards to these concepts, except perhaps through stipulation and surely not by philosophical or empirical argument. It would be simple to say that religion and culture are paradigmatic examples of ‘essentially contested concepts’ (and I think they are), but this only points to a broader methodological position: even concepts that are essentially contested in the abstract may be less so when addressed in more concrete discursive contexts or with reference to discrete historical strains.9 The difficulty of providing comprehensive definitions should not preempt the more modest endeavor of distinguishing between the kinds of claims that are made in the name of culture and religion, and clarifying at least the historical, formal, and institutional differences between them. This exercise necessarily involves some idealization, but I think that it is justified both because the history, conceptual structure, and institutional response to multicultural accommodation and religious freedom bear out a distinction, and because the distinction is prospectively useful in articulating legal and political responses to these claims. We may not be able to define culture and religion comprehensively, nor distinguish between them at an abstract level, but we may be able to distinguish between culture as used in multicultural discourse and religion as used in the discourse of religious freedom, and that more narrow and ‘situated’ or ‘contextualized’ definition is useful for explaining the appropriateness of rendering one concept in terms of the other. Given the extent and complexity of the literature on multiculturalism, I will draw in broad strokes. Multiculturalism refers, on the one hand, to a wide variety of policies adopted mainly (though not exclusively) by post-industrial liberal-democratic states in the last decades of the twentieth century, as they faced growing immigrant communities that seemed more resistant than previous migrants to cultural assimilation. On the other hand, multiculturalism is also ‘a body of thought in political philosophy about the proper way to respond to cultural and religious diversity’, marked by a rejection of strategies of ‘mere toleration’ and by ‘recognition and positive accommodation of group differences … through “group-differentiated rights”’.10 The political and philosophical dimensions of multiculturalism are closely related, and have informed each other over the years, even enjoying a parallel wax and wane of popularity. While something recognizable as multiculturalism is an inherent part of any imperial enterprise, and can thus be projected over thousands of years of human civilization and conquest,11 multiculturalism as explicit policy only comes into being in the 1970s and 1980s, and only becomes a major theme of self-conscious political theory in the last decade of the twentieth century.12 Here, I will limit myself to the philosophical literature on multiculturalism, and forgo an analysis of historical antecedents or empirical policy comparisons. The history that I take as subject is the history of ideas, because of what it clarifies about the distinctness of religious liberty. Two motivations spur the emergence and growth of the multicultural literature: the first, just mentioned, is the pursuit of policies of cultural accommodation by several post-industrial liberal-democratic countries who were important recipients of immigration. The second is the publication and subsequent reaction to John Rawls’s A Theory of Justice, the touchstone of twentieth century liberal political philosophy.13 Theorists like Michael Sandel and Charles Taylor – who are rather sloppily labeled as communitarians, a moniker most of them rejected –object to what they perceive to be an impoverished conception of the person in Rawls’s work, and take issue especially with the perceived lack of attention to the constitutive commitments that individuals draw from their cultural, religious, and other ethical communities.14 The definitive response from the Rawlsian camp was Will Kymlicka’s Liberalism, Community, and Culture – definitive because later endorsed by Rawls as representing his position in the socalled liberal-communitarian debate.15 Kymlicka understands the problem of multiculturalism in distinctly liberal terms: ‘[a] liberal democracy’s most basic commitment is to the freedom and equality of its individual citizens.’16 Drawing on the different strains of the liberal tradition, he articulates ‘a distinctively liberal approach to minority rights’17 that takes an individual’s societal culture – one ‘which provides its members with meaningful ways of life across a full range of human activities, including social, educational, religious, recreational, and economic life’18 – to provide an interpretive context that gives meaning to the different choices with which a person is confronted in a diverse liberal society. The exercise of freedom presupposes such a context, which is why minority cultures need protection, as the loss of their context of choice would prove traumatic to their members because of its high cost and the offense to members’ self-identity and so disable them from exercising their freedom meaningfully, from cultivating the virtues of citizenship. Multiculturalism is, then, first and foremost, a subsidiary strategy adopted by liberal-democratic societies in order to further the primary or dominant strategy of individual autonomy, and more specifically the autonomy of the individual as a citizen of the liberal-democratic state. Multicultural policies can be constructive or remedial: constructive when the cultural context that is preserved by multicultural policies then forms the basis for a deeper identification with a liberal and egalitarian public culture, made all the richer because it is sustained through an overlapping consensus of reasonable views, many of which are culturally rooted; or remedial when multicultural policies address and attempt to ameliorate historical or structural injustices that undermine the self-respect of members of disadvantaged cultural communities and thus prevents them from standing as equals in a common condition of citizenship. 3. The Limits of Multiculturalism The concern of liberal multiculturalism is an important one, and I do not question that some accommodation for societal cultures is a requirement that arises from a respect for human decency.19 Yet the point of departure of liberal multiculturalists—the autonomous individual—skews the perception of the groups that are objects of recognition or accommodation. The nature of religious groups – my concern in this essay – is especially distorted by the choice of examples that are fielded when discussing religious accommodation. As Sarah Song observes, ‘[m]ost of Kymlicka’s examples [of poly-ethnic rights20] involve religious practices’.21 The cases she alludes to are ‘exemption from Sunday closing or animal slaughtering legislation … from motorcycle helmet laws and from the official dress-codes of police forces … the right to wear the yarmulka during military service … [and] exemption from school dress-codes so they can wear the chador.’22 This is triply troubling: first, because the examples confuse the causal connection between religion and ethnicity by selecting examples where religious and ethnic membership overlap considerably though not completely. Ethnic claims for exemption not couched in religious garb are less likely to receive as much deference.23 There are sound strategic reasons for this. Many constitutions have express constitutional provisions protecting religious freedom (the First Amendment to the United States Constitution, section 2(a) of the Canadian Charter of Rights and Freedoms, etc.) while few dispositions elevate cultural protection to the same level. An interesting case is section 27 of the Canadian Charter, which mandates that the instrument ‘shall be interpreted in a manner consistent with the preservation and enhancement of the multicultural heritage of Canadians’. But the courts have not derived independent rights from section 27; they have only used it to reinforce other rights, among them freedom of religion.24 Second, the equivalence of religious and multicultural claims exoticizes demands for religious accommodation by highlighting cases where members of the religious group visibly distinguish themselves in ways the dominant culture finds unusual and conspicuous. These distinctions are often ‘alien’ because of cultural or temporal distance, or both, from the dominant society. Consider Hasidic Jews, whose garb marks temporal more than ethnic distance, or the Amish or Hutterites, whose exoticism is both temporal and ethnic (due to their German origins), and extends also to their aversion to technology. The exoticization of religious claims by those intent on bringing them under the multicultural paradigm obscures many claims that are also religious but not ethnically minoritarian: conscientious objection by members of many Christian denominations in Christian countries, self-government by mainline churches including, both Roman Catholic and Protestant, tax exemptions for religious denominations, etc. The Roman Catholic Church was claiming religious liberty against the Holy Roman Emperor as early as the fifth century, when its membership, with few exceptions, was coextensive with the population of the Empire (I return to this case in the next section). Third, poly-ethnic rights are ‘usually intended to promote integration into the larger society, not self-government’.25 But in many cases involving religious freedom, it is precisely self-government that distinguishes religious claims from mere cultural claims. It is not (as I argue below) that religious freedom is only about corporate autonomy for churches – that would be to ignore four centuries of development of the doctrine of freedom of conscience – but rather to say that there remains, as Perry Dane has put it, a ‘specter of intractability’ over the boundaries of religious freedom.26 And this is not captured by the multicultural paradigm. Indeed, religious groups consistently fall outside multicultural categories. The distinctively collective claims that religious groups make – group rights proper, as opposed to group-differentiated rights – are seen as an exception to the general category of cultural rights, which are ordinarily bestowed on individuals.27 Certainly some cultural groups make collective claims as well – claims to self-government, external rules limiting non-members liberty, internal rules respecting members’ conduct, and recognition and enforcement of traditional norms – but these are usually constituted as nations, peoples, or quasistates; Kymlicka’s distinction between multiethnic and polyethnic states speaks to this.28 In this classification, religious groups do not easily fit: they are associations in the state, not aspirants to territorial self-government, although they often exercise jurisdiction albeit along personal, not territorial, lines. They provide contexts of choice to their members, but do so through explicit structures and positive norms which may be more or less hierarchical, more or less restrictive, but distinguish members from non-members in a much more formal way that the ambiguous and porous boundaries of culture. Now, not all religious associations are formal to the degree of the Catholic or the Anglican churches, with their various gradations in episcopacy and elaborate systems of canon law. But, almost by definition, religious associations have discrete boundaries, which makes them qualitatively different things from other ‘groups’ like cultural minorities. Moreover the various strains that form the paradigm of religious freedom in the West take the formal character of the religious association as paradigmatic. This may have a normalizing effect on other religious associations, which conform, through various legal mechanisms, to the structure of more formal denominations, and should not thereby be taken as an endorsement of a certain organizational principle.29 But neither should it obscure the observation that the development of religious liberty in the West has been shaped by the confrontation of an organized church and an organized state. While confrontation also shaped the development of multiculturalism (consider most civil rights movements) it was generally aimed at the recognition of individual dignity or, at most, groupdifferentiated rights, not institutional autonomy. 4. Libertas Ecclesiae The historical sources of freedom of religion—especially the Abrahamic religions in the West—are twofold, and while each is well known, the importance of one or another source is usually downplayed by advocates of a particular conception of religious liberty. The older source is ultimately traced to the conflict between secular and religious authority dating to the end of the Roman Empire and extending through the so-called twelfth century Renaissance. The first important expression of it is in Pope Gelasius’ letter to Emperor Anastasius in 494: Two [elements] there are indeed, Imperatur Augustus, by which this world is principally ruled: the consecrated authority of the priests and the royal power … For if, in matters pertaining to the realm of public discipline, the religious dignitaries, recognizing that royal power was given to you by divine disposition, for their part obey your laws, lest they seem to neglect and reject your decree in worldly affairs; then with what fervor, I pray you, ought one to obey those who have been charged with celebrating the holy mysteries?30 The later medieval world was characterized by many overlapping jurisdictions, of which the Pope and the Holy Roman Empire were the most prominent. Their conflict came to a head during the Investitures Conflict in the eleventh century, which was resolved in the Concordat of Worms in 1122. It recognized as a constitutive element of the political order the existence of the Church as an autonomous authority, independent of the Empire. Harold Berman and others have called this principle – libertas ecclesiae – one of the foundations of Western freedom, as it established external limits to state authority, namely the authority of another sovereign body.31 Richard Garnett and others have explained not only the foundational character of this concept, but also its explanatory potential, as it seems to make sense of the latitude given to churches and other religious authorities in liberal-democratic states, even following widespread secularization.32 A general attitude not to interfere with the internal deliberation of religious bodies, and to defer to their judgments in matters concerning religious doctrine and religious law is prevalent, and extends, more concretely, to exemptions from certain anti-discrimination statutes, differences in the application of some provisions of the tax code, and special consideration during bankruptcy and tort proceedings.33 Only during the Reformation does religious freedom take an individualist turn, developing into a defense of freedom of conscience, which in turn becomes one of the central pillars of liberalism and the second source of the idea of religious freedom. This was a welcome turn and, as I have said, a definite civilizational achievement. Yet it obscured (although it did not replace) the previous collective or corporate strain in the tradition of religious freedom—the oppositional nature of Pope and Emperor, church and state.34 This strain remained in the discourse of religious bodies, but all but disappeared from liberal theory, especially because it seemed difficult to articulate in neutral or, at least, non-fideistic ways. Consider, for instance, Peter Berger’s defense of religious freedom: Religious liberty is not one of many benefits that the state may choose to bestow on its subjects; rather, religious liberty is rooted in the very nature of man and, when the state recognizes it, the state ipso facto bows before a sovereignty that radically transcends every worldly manifestation of power. For the religious believer, of course, this is the sovereignty of God; for the agnostic it will be the sovereignty of that mystery within man that ever thrives to go beyond the given—the mystery of man’s freedom.35 The language or divine sovereignty is excluded (with good reason) from the liberal justification of rights in diverse polities, but nothing is substituted for it except the language of individual freedom of conscience, an altogether different idea. At one level, the appeal to conscience helps us to make a distinction between culture and religion. As Sarah Song observes, [t]here is, however, an important sense in which religious belief and cultural attachments may be different. Both may be central components of people’s identities, but the tenets of religion, unlike the demands of cultural affiliation, are matters of conscience and obligation and therefore might be viewed as being of more fundamental importance. One might argue that adherence to a religious belief entails accepting the belief as being of fundamental importance and embracing the commitment to live by its dictates, whereas being a member of a minority cultural group does not necessarily entail accepting one’s cultural identity as being of fundamental importance or accepting any commitment to live according to whatever norms are associated with cultural identity.36 Claims of conscience, as claims of culture, do not necessarily presuppose duties towards an institutional authority, which facilitates the subsumption of religious claims under cultural claims: they are both deeply held, hard to revise, and central to identity. But in the process the distinctive origin of religious freedom is ignored. The theoretical dismissal of the institutional claims of religious authority, which are as important to the development of religious freedom as the later arguments about individual conscience, naturally leads to a mistake: the peculiar authoritative structure of religious practice is analogized to a more passive, more inert background which provides individuals ‘with an intelligible context of choice, and a secure sense of identity and belonging’.37 There are certain aspects to religious practice that are properly classed as religious culture: rituals, artistic imagery, particularly pungent profanities.38 But this is not all that religion is, and it is far short of what it has been throughout history. Religion is often more active than a background context of choice would imply, and makes emphatic assertions prescribing and proscribing conduct and belief. And it is also often less pervasive than the notion of culture would suggest, adapting practices to cultural context, as in the case of liturgical music or vernacular prayer. That some of these adaptations are strategic choices by religious associations, points to the historical reality that much of religion in the West is organized religion, and that this historical development affects the institutional concern of freedom of religion as well ass the formal or conceptual structure of religious versus cultural claims. 5. Religion as Practical Authority From Berger’s assertion that religious authority is ‘a sovereignty that radically transcends every worldly manifestation of power’ we can extract a more formal principle: that religion is a practical authority, in a way that culture is not. Now, culture and religion share in the form of practice, which Michael Oakeshott defines ‘as a set of considerations, manners, uses, observances, customs, standards, canon’s maxims, principles, rules, and offices specifying useful procedures or denoting obligations or duties which relate to human actions and utterances.’39 Like religious practices, cultural norms enable individuals to relate to each other as engaged in a common performance. These symbolic ‘procedures’ ground cultural meaning, enabling the context of choice that Kymlicka identifies, and may be conventional norms for participants in cultural interaction. They can therefore fall within the province of practical reason, and even help constitute the authority of certain actors in a determinate social context, while not themselves being practical authorities. There are several ways to approach this distinction. I will first address the institutional differences between culture and religion, again as the concepts are used in multiculturalism and religious freedom, and later discuss the idea of authority itself from a conceptual perspective. On one hand, neither a culture nor a religion as such can act as a legal, moral, or political agent. In this claim, I have in mind Hobbes’s famous definition of collective or corporate personhood, whereby, [a] multitude of men are made one person, when they are by one man, or one person, represented so that it be done with the consent of everyone of that multitude in particular. For it is the unity of the representer, not the unity of the represented, that maketh the person one. And it is the representer that beareth the person, and but one person, and unity cannot otherwise be understood in multitude.40 Hobbes defined ‘systems’ as ‘any numbers of men joined in one interest or in one business’41 and divided these into ‘regular’ and ‘irregular’ systems. Regular were those which had a representative (according to the definition given above) and could thus act with a discrete and common agency; irregular were leagues, factions, and alliances, but also individuals not formally organized who concurrently engaged in some activity, such as participating in a market. On this spectrum, both religion and culture fall on the irregular end: whatever else they might be, they involve a group of persons engaged in a common practice and sharing a common set of symbols, beliefs and practices, and perhaps interests. But as much as they may be personified in speech – Christendom, world Jewry, the Umma – cultures and religions are not persons in the sense of having discrete and common agency. By contrast, formally constituted cultural associations and religious associations are, by definition, capable of acting as agents – of being ‘represented’ – and of communicating to their members and to third parties in a determinate voice. The beliefs and practices advocated by these groups may correspond to a greater or lesser degree with the beliefs and practices of the universe of persons who self-identify (or are identified by others) as ‘belonging’ to the respective culture or religion. Of course, the boundaries of cultures or religions are often contested, and often by associations that claim to embody them. The boundaries of cultural or religious associations are more rarely questioned, if only because the conditions of membership often have a positive and formal quality that is missing in the former: they are set down in by-laws and constitutions, and when they are ambiguous and unclear, an affirmative decision on the part of the association’s officials will resolve the question, at least in the immediate case. But the difference between the religion and the religious group does not turn on the degree of inclusiveness of members of a certain class. It is a qualitative difference (one might even say an ontological one) and certainly one with profound institutional consequences. To identify a cultural association with the culture it celebrates or a religious organization with the religion to which it alludes is something like a category-mistake. They are simply not the same kind of institution. In brief, neither cultures nor religions as such are associations, although there are associations that are primarily organized around shared cultural or religious practices or traditions, and sometimes (usually unsuccessfully) claim to represent or police the boundaries of the entire tradition. Why is this relevant to the difference between multiculturalism and religion? Because the history of religious freedom developed by reference to distinct corporate institutional agents, not cultural groups, to ‘regular’, not ‘irregular’ systems; thus the concern of religious freedom for group-rights as well as group-differentiated rights, and the relative dismissal of the former by multiculturalists. The institutional focus of multiculturalism, its ‘social ontology’ as it were, does not contemplate institutionalized groups that exercise an authority similar to that of the state. 42 It is fundamentally individualistic in its institutional approach, even when it considers the importance of the cultural group in framing the individual’s identity. This is why the attempt to address issues of religious liberty by reference to multicultural categories ultimately creates an incomplete account. A particularly interesting example of this is the report of the Québec commission to study the problem of ‘reasonable accommodation’ in the province, authored by Gérard Bouchard and Charles Taylor (the latter, especially, an ardent advocate of a robust multiculturalism).43 The commission’s mandate instructed the Co-Chairs to take stock of accommodation practices related to cultural differences, analyse the attendant issues bearing in mind the experience of other societies, conduct an extensive consultation on this topic, and formulate recommendations aimed at ensuring that accommodation practices conform to Québec’s core values ... The analysis of accommodation practices related to culture, including religious life, and of related questions led us to directly question our society’s most fundamental sociocultural dimensions.44 The inclusion of ‘religious life’ as an aspect of ‘accommodation practices related to culture’ would go on to distort the subsequent analysis, even when the authors explicitly acknowledge the corporate dimension of religious liberty. Thus, they describe the separation of church and state ‘as a reciprocal autonomy. The State is free of all religious tutelage while religious associations are autonomous in their fields of jurisdiction, although they remain subject to the obligation to respect basic human rights and the legislation in force.’45 But the acknowledgement of the autonomy of religious bodies is not brought up again; instead the commissioners come to rely exclusively on [t]he subjective interpretation of religion … according to which freedom of religion must be regarded as an aspect of the broader category of freedom of conscience, which seeks to ensure that individuals are free to adopt the religious, spiritual or secular beliefs or fundamental reasons of their choice and that they are not compelled to act contrary to their convictions of conscience.46 I suspect that the reason for the reduction of religious freedom to only one of its strains – the individualist one that emerges after the Enlightenment – is, to a great degree, derived from the larger frame in which the question is put: as an aspect of deep-seated individual commitments similar to those of culture, which may be granted protection in group-differentiated rights, but which also make unintelligible a more robust conception of institutional religious autonomy. Final confirmation comes from Kymlicka who acknowledges (approvingly) that multiculturalism offers little support for the corporate strain of religious freedom and the institutional accommodations demanded in its name: ‘the real issue […] is the pre-modern legal doctrine of libertas ecclesiae, which gives religious organizations broad exemptions from equality rights, not the postmodern Multiculturalism Act, which firmly endorses the norms and principles of equality.’47 His assessment of the relationship between equality and religious autonomy is not entirely correct – to give just one example, while libertas ecclesiae has been raised to insulate churches from complaints of discrimination on the basis of gender and sexual orientation, it has also been used to protect immigrants though the provision of sanctuary – 48 but his identification of a premodern idea of religious liberty not comprehended in multicultural discourse is exact. They are different concepts and often pull in different institutional directions. 6. Religion as a normative system Some concepts in the philosophy of law may further illustrate the difference. Foremost is H.L.A. Hart’s distinction between primitive and fully-developed legal systems or, to use Hart’s less prescriptive language, the distinction between social structures composed only of primary rules of obligation and a society that adds to these secondary rules of recognition, change, and adjudication; the difference turns on Hart’s famous explanation of primary and secondary rules, and of legal orders as the union of both systems of rules. Primary rules are rules by which ‘human beings are required to do or abstain from certain actions, whether they wish to or not.’49 Secondary rules, by contrast, are rules about rules—rules that ‘specify the ways in which the primary rules may be conclusively ascertained, introduced, eliminated, varied, and the fact of their violation conclusively determined.’50 It is telling that Hart momentarily analogizes a system of primary rules alone to one of ‘custom’, but steps away from the analogy ‘because [the term custom] often implies that the customary rules are very old and supported with less social pressure than other rules.’51 But the momentary lapse is illustrative of a formal feature of customary and, by extension, cultural systems: they are normative systems, even authoritative systems, in that they prescribe the right ways of doing things, e.g. how to spice food, how to greet a family member or a stranger, how to dress for a funeral or a wedding. But what they lack is a way (or, for that matter, an authoritative agent) to identify a practice as part of the culture and discard another, to modify or abolish a cultural practice, or to identify cultural deviance in any manner that is generally acceptable in any but the most contingent way. Now, Hart writes that ‘only a small community closely knit by ties of kinship, common sentiment, and belief, and placed in a stable environment, could live successfully by such a regime of unofficial rules.’52 This may put a strain on the idea that cultural systems are similar to systems of primary rules, as cultural systems do not usually require such close connection and often operate on a much larger scale. Yet that would be too hasty a conclusion. The purpose of invoking Hart’s distinction is not to create a perfect equivalence, but to bring out a significant difference between cultural and religious norms. Nonetheless, some trepidation is warranted. The direction of the causal arrow in the relationship between the ‘close-knittedness’ and the ‘effectiveness’ of cultural systems is ambiguous, as one of the functions of such systems is precisely to create the kinds of shared meanings and commitments that create communal ties. These ties are not only ties of obligation, but often generate semiotic and hermeneutic rather than practically normative contexts, and are therefore not so profoundly affected, perhaps, by problems of enforcement. Lastly, none of this undermines the conclusion that cultural systems have no ‘prescriptive agent’ to determine the validity of cultural rules. We may be able to trace the emergence of cultural norms anthropologically, and even genealogically, but this will have no bearing on their normative status; by contrast, tracing the origins of a rule in a more developed legal system in a way that shows it not to be identified by a rule of recognition or sanctioned by a rule of change will rule it out as a rule of the system. Religious groups, by contrast, are constituted by secondary rules. This is true, in a sense, not only in their worldly structure but also, by projection or analogy, in the origin and form of divine intervention (even the deity, it seems, follows precise conventions when proclaiming the law). Religious belief often makes reference to creeds or canonical literature. Even non-credal religious movements – such as Unitarian Universalism or Reform Judaism – are, despite their self-assertions, at least passively creedal. To give an example, Reform Jews consider the bindingness of any given mitzvah (especially ritual mitzvot) to be a matter for the individual conscience to decide in light of the principles of Judaism, but that attitude towards halakha excludes from Reform Judaism anyone who believes that all the 613 mitzvot are binding on all Jews irrespective of individual conscience; Reform latitudinarianism does not encompass Orthodoxy. In order to identify these authorities, it must have something like a rule of recognition; in order to admit changes to these authorities (though they be changes at the margins, such as the recognition of a binding custom) or even to deny in principle the possibility of amendment, it must have rules of change;53 in order to determine the content of religious obligation it must have rules of adjudication. Especially in the case of the Abrahamic religions, all of which, in most of their manifestations, have designated a special class of officials to clarify matters of doctrine and decide on religious disputes, and the secondary rules both constitute and guide the authority of these officials. Cultural systems have no equivalent to this, and multiculturalism is ill-equipped to accommodate it, for reasons I will suggest in the last section. Another complementary way of way of describing the difference between cultural and religious norms builds on Joseph Raz’s account of the justification of authority.54 Authoritative reasons, Raz writes, are preemptive in that they replace, rather than add to, the reasons that a person has for acting; they can do this because they are predicated on reasons that apply to the person in any case, and the person is more likely to comply with those first order reasons if she accepts and tries to follow the directives of the authority rather than trying to follow the reasons that apply to her directly. This, Raz says, is the normal way of justifying authority. Now, both religious and cultural claims may be authoritative in that they prescribe certain directives to participants in a practice, and even indicate that adherence to the authoritative claim is more likely to realize the benefits of membership. But cultural claims are rarely preemptive in the way that religious claims almost always are. One reason is that the bindingness of an authority is institutionally constituted by power-conferring norms of the kinds that religious groups possess, but cultural groups do not. This line of reasoning was evident in Raz’s early work, but later waned in importance.55 It has, however, been recovered by Andrei Marmor, who emphasizes that authority is always situated in an institutional context: [t]he essential feature of any practical authority is that to have authority is to have power, in the normative sense of the term. A normative power is the ability to introduce a change in the normative relations (viz. rights, obligations, etc.) that obtain between those who are subject to the power under the relevant circumstances. The existence of power, however, is essentially an institutional matter, or so I shall argue here. Only rules or conventions of an institution, or a well structured social practice, can confer power. And this is why authorities are essentially institutional in nature, and the obligation to comply with their directives are institutional obligations.56 It is important to observe that some cultural norms are authorityconferring norms. Norms regarding gender relations, or the relation between parents and children, or elder and younger members of society, are precisely of this sort. In this ‘religion’ and ‘culture’ are similar, when considered as sociological or anthropological phenomena: religious practices create the institutional preconditions that confer the normative powers claimed by religious authorities, much like cultural practices create the conditions for the power of various other social institutions. But when considered as historical phenomena, the difference not between culture and religion, but between multiculturalism and religious freedom bears out the implications of the argument about practical authority. Modern multiculturalism has not been advanced by ‘cultures’ but by individuals burdened by restrictions on their cultural practices; in those cases when there has been a claim to respect the culturally-conditioned authority of male parents or elders or male heads of families, the multicultural argument has been derivative, and has taken the form of a group-differentiated claim to respect certain institutional practices because of the benefit they confer to individuals, where this benefit is evaluated from the perspective of the liberal-democratic state. This has, quite justifiably, led to significant restriction on such claims of authority. The distinction, then, is not categorical, but more subtle, and explained as much by historical claims as by conceptual analysis. The arguments that lead to claims of religious liberty, as I observed above, are of two streams: the older tradition of freedom of the church, and the later tradition of freedom of conscience. Freedom of conscience, despite some caveats, has some similarity with multicultural claims, especially as it placed the individual believer or practitioner at the center of inquiry. The tradition of freedom of the church, however, places the religious authority at the center of inquiry. In this earlier tradition, religious liberty has been advanced by religious organizations for whom the religious community is perfectly coextensive with the formal religious association. Extra ecclesiam nulla salus: this historical claim – which is still one of the demand of religious groups – grounds an important dimension to religious freedom that finds little parallel in multicultural claims. Multiculturalism may reference cultural practices that constitute practical authorities, but religious freedom – at least in the tradition of freedom of the church – references those authorities directly. In this way, religion (in the religious freedom paradigm) is a practical authority, while culture (in the multicultural paradigm) is not. It seems necessary to concede that, at the margins of culture and religion, there is certainly a blurring of the line between systems of primary rules and those of primary and secondary rules. There is much in religious practice that is culturally derived, not of doctrinal or dogmatic origin, although, except in religious communities that are largely coterminous with ethno-cultural communities, the cultural and religious elements seem to operate at different levels, or at least be recognized as distinct. This is perhaps most evident in the Abrahamic religions, and especially in the most proselytizing of these: charismatic Roman Catholic practice in rural Latin America is quite distinct from that of Tridentine parishes in the Italy, but both claims, with much justification, that this variance is due to cultural elements exogenous to orthodoxy and equally consonant with the Magisterium.57 Similar observations could be made of the differences in manner of dress, family structures, or modes of worship between Muslims in Jordan, India, and Indonesia. Even among the Jewish population, dietary and ritual divergence between the Ashkenazi and Sephardic communities (as opposed, say, to the difference between Karaite and Rabbinical Judaism) are understood to be exogenous to the religious requirements of halakha.58 The equation of religion and culture serves neither well, because it misclassifies both. What we see when we treat cultures as if they possessed secondary rules, is the calcification of the boundaries of the cultural group and its capture by official arbiters of culture; in essence, the transformation of a culture into a cultural association. In reverse, what we see when we treat religious groups as if they were social structures composed only of primary rules, is that we mischaracterize and even undermine the very religions that we may be inclined to protect. 7. Conclusion To sum up, the survey of multiculturalism and religious liberty suggests a distinction in the institutional implications of an approach grounded on one or another paradigm. Some claims made by religious adherents will be of a kind with the claims made in the name of minority cultures, and a common institutional framework may suffice for both. Other claims made in the name of religion, however, cannot be put in multicultural terms. Religious freedom differs from multiculturalism because the sorts of claims made by religious groups that have followed a certain historical trajectory are necessarily collective, proper grouprights, as opposed to collectively-derived yet individual group-differentiated rights. The reason is formal: religious liberty has as its object both individual practitioners and believers who can be analogized to members of cultural communities and religious associations which cannot be so analogized. The reason they cannot is that religious associations are the sort of corporate agent that can exercise a will while cultures are not. Drawing on Hobbes’ famous classification of ‘systems’ we can categorize religious groups as ‘regular’ systems and cultures as ‘irregular’ where regularity or its lack depend on the groups having ‘a sovereign representative,’ one that can self-consciously form its will and communicate it as the will of a collective.59 But against Hobbes, religious groups, to again echo Berger, claim an authority – a sovereignty even – that is external from the state. Multicultural claims are defended as a necessary corollary of respect for individual freedom and autonomy, which are the very values that undergird the legitimacy of the liberal state. But religious claims are, in principle, indifferent to these principles; they are, in principle, foundationally and, in practice, at least potentially incompatible with the state legal system, even if they coexist peaceably with it.60 No such claim can be made about culture, because it does not have the same kind of institutionally authoritative structure. Cultures may be authoritative in some vague way, but are not, because they are incapable of being practical authorities. But the distortion cuts both ways. The elision between religion and culture in the theory of multiculturalism has also had the effect of transforming the authoritative claims of cultural groups into the claims of cultural authorities. Anne Philips decries that ‘[i]n the political theorist’s understanding of culture, cultural group then becomes associated with a quasi-legal entity that has historically enjoyed or is now claiming jurisdiction over its members.’61 Cultural groups are not as formal as that, but many religious groups are. Granted, there are exceptions especially as one moves away from the Western experience, but I would argue that the formal distinction holds across a comparative analysis. It is not a tidy distinction, and there are various ways in which cultural practices may change more or less consciously in certain circumstances, and ways in which religious norms are predicated upon unspoken conventions and (often cultural) assumptions. But as an ideal type, I think it helps to clarify important differences. See B. Tierney, ‘Religious Rights: A Historical Perspective’, in N.B. Reynolds and W. Cole Durham (eds), Religious Liberty in Western Thought, (Atlanta: Scholars Press, 1996) 29-57. 2 So, for that matter, is multiculturalism, as a philosophical and political complement to liberalism that seeks to minimize instances of ethnic ‘violence, cruelty and humiliation’ (J.T. Levy, The Multiculturalism of Fear (Oxford: OUP, 2000) at 38), and even to actively protect cultural identity because of ‘the role that it plays in enabling meaningful individual choice and in supporting self-identity’ (W. Kymlicka, Multicultural Citizenship (Oxford: OUP, 1995) at 105). 3 Tierney’s work is a useful antidote to this, as is H. Berman, Law and Revolution: The Formation of the Western Legal Tradition (Cambridge, Mass.: Harvard University Press, 1983). 4 R. Garnett, ‘Two There Are: Understanding the Separation of Church and State’, in M. Hogan and L. Frederking (eds), The American Experiment in Religious Freedom, (Portland: Garaventa Center, 2008) 319-30 at 325. 5 See part one of N. Rosenblum, Membership and Morals (Princeton: Princeton University Press, 1998). 6 A. de Tocqueville’s observations on the relative position of churches in J.P. Meyer (ed), Democracy in America, translated by G. Lawrence (New York: HarperCollins, 1969) at 287-90) and in F. Furet and F. Mélonio (eds), The Old Regime and the Revolution, translated by A. Kahan (Chicago: University of Chicago Press, 1998) at 171-75) are an early testament to this. 7 When referring to the paradigm of religious liberty, I do not make universal claims, but limit my account and my examples to religious liberty as developed in the Western legal tradition. There are problems even with delimiting the boundaries of that tradition, but I think they do not affect the historical narrative or theoretical implications of this essay. 8 While multiculturalism and religious liberty are not only paradigms for the accommodation of difference, this is the main role that they play in liberal political theory. The stated liberal multiculturalist concern is to show ‘equal respect to citizens [by] recognizing and accommodating their cultural differences, insofar as it does not impact adversely on the rights and freedoms of others’ (J. Maclure, ‘Multiculturalism and Political Morality’ in D Ivison (ed.), The Ashgate Research Companion to Multiculturalism (Ashgate, 2010) 39–55, 40). And the liberal concern for religious liberty, even if directly motivated by respect for conscience, centrally involves accommodation (M. Nussbaum, Liberty of Conscience (Basic Books, 2010) 21). My contention is that liberal accommodation to individual claims of conscience neither exhausts or fully justifies religious liberty, but rather that acknowledgement of a distinct (and often corporate) authority is also an inherent part of the Western tradition of freedom of religion. 9 W. B. Gallie, ‘Essentially Contested Concepts’, 56 Proceedings of the Aristotelian Society, New Series (1956) 16798. Despite the dilution, through overuse, which Gallie’s category of ‘essentially contested concept’ has suffered since its introduction, it seems amply justified to resort to it in the case of religion and culture, not least because the original author referred to religion as one of his examples. 10 S. Song, ‘Multiculturalism’ in E.N. Zalta, (ed), The Stanford Encyclopedia of Philosophy, (Winter 2010 Edition) available online: <http://plato.stanford.edu/archives/win2010/entries/multiculturalism/>. 11 Susanne Hoeber Rudolph makes the provocative distinction that ‘[a] nation-state is a restricted territory in which there is a presumption or at least an aspiration of congruence between the state and a nation or people. By contrast, an empire is an extended territory comprising a group of states or peoples under the control or at least the suzerainty of a dominant power.’ ‘Presidential Address: State Formation in Asia—Prolegomenon to a Comparative Study’, 46:4 The Journal of Asian Studies 46 (1987) 731-46, at 736. The imperial label intuitively fits what Will Kymlicka calls ‘multination states’ more than it does ‘polyethnic states’ (W. Kymlicka, Multicultural Citizenship, supra note 2, at 11-26); but the dynamics of governing difference may be more similar than intuition intimates. 12 W. Kymlicka, Liberalism, Community, and Culture (Oxford: OUP, 1989); C. Taylor et al, Multiculturalism: Examining the Politics of Recognition (Princeton: Princeton University Press, 1994). There were antecedents, of course, especially in the cultural pluralism of H. Kallen, Cultural Pluralism and the American Idea (Philadelphia: University of Pennsylvania Press, 1956). 13 J. Rawls, A Theory of Justice (Cambridge, Mass.: Belknap Press, 1971). 14 M. Sandel, Liberalism and the Limits of Justice (Cambridge: CUP, 1982); C. Taylor, Philosophy and the Human Sciences (Cambridge: CUP, 1985). 15 Kymlicka, Liberalism, Community, and Culture, supra note 12; J. Rawls, Political Liberalism (New York: Columbia University Press, 1993) at 27, n. 29. 16 W. Kymlicka, Multicultural Citizenship, supra note 2, at 34. 17 Ibid, at 75. 1 18 Ibid, at 76. This is especially evident in Jacob Levy’s normatively minimalist defense of multiculturalism, supra note 2. 20 These Kymlicka defines as ‘group-specific measures … intended to help ethnic groups and religious minorities express their cultural particularity and pride without it hampering their success in the economic and political institutions of the dominant society.’ Multicultural Citizenship, supra note 2, at 31. 21 S. Song, Justice, Gender, and the Politics of Multiculturalism (Cambridge: CUP, 2007), at 65. 22 Kymlicka, Multicultural Citizenship, supra note 2, at 31. 23 See L. Sager, ‘The Free Exercise of Culture: Some Doubts and Distinctions’, 129:4 Daedalus (2000) 193-208, at 196; also cited by Song, Justice, Gender, and the Politics of Multiculturalism, supra note 20, at 65, n. 49. 24 Tellingly, however, the same court that used section 27 to reinforce freedom of religion also reduced the latter to ‘the notion of the centrality of individual conscience and the inappropriateness of governmental intervention to compel or to constrain its manifestation’ departing from (and indeed distinguishing from prior jurisprudence that supported) an institutional, corporate dimension to religious freedom. R v Big M Drug Mart Ltd, [1985] 1 SCR 295, at 115 and 121. 25 Kymlicka, Multicultural Citizenship, supra note 2, at 31. 26 P. Dane, ‘Constitutional Law and Religion’, in D. Patterson (ed), Blackwell Companion to Philosophy of Law and Legal Theory, (Oxford: Wiley-Blackwell, 1999) 113-25, at 128. 27 Kymlicka, Multicultural Citizenship, supra note 2, at 40-43; Levy, Multiculturalism of Fear, supra note 2, at 13738; L. Swaine, The Liberal Conscience (New York: Columbia University Press, 2008), at 116-17. 28 Kymlicka, Multicultural Citizenship, supra note 2, at 11ff. 29 There may be normative problems with this tendency, but I currently don’t have a formed opinion about their desirability. Neither do I claim to extend my account beyond the development of religious freedom and multiculturalism in the West; the dynamic between the state and religious associations is markedly different in other historical contexts. 30 Pope Gelasius I, ‘Letter to the Emperor Anastasius’ in K.F. Morrison et al (eds), University of Chicago Readings in Western Civilization: The Church in the Roman Empire, vol. 3 (Chicago: University of Chicago Press, 1986) 112. 31 Berman, Law and Revolution, supra note 3, at 87. 32 R. Garnett, ‘The Freedom of the Church’, 4 Journal of Catholic Social Thought 4 (2007) 59-86; R. Garnett, ‘Religious Liberty, Church Autonomy, and the Structure of Freedom’, in J. Witte, Jr. and F.S. Alexander (eds), Christianity and Human Rights: An Introduction, (Cambridge: CUP, 2010) 267-82. 33 P. Dane, ‘Constitutional law and religion’, supra note 25, at 129. 34 Kymlicka, for instance, situates the origin of religious tolerance in the Wars of Religion, Multicultural Citizenship, supra note 2, at 155ff. 35 P. Berger, ‘Afterword’, in J.D. Hunter and O. Guinness (eds), Articles of Faith, Articles of Peace (Washington, D.C.: Brookings Institution Press, 1990), 114-121, at 118. 36 Song, Justice, Gender, and the Politics of Multiculturalism, supra note 20, at 65. 37 Kymlicka, Multicultural Citizenship, supra note 2, at 105. 38 On the interjective use of religious language, see e.g. C. Legaré and A. Bougaief, L'Empire du sacre Québécois (Québec: Presses de l'Université du Québec, 1984). 39 M. Oakeshott, On Human Conduct (Oxford: OUP, 1975), at 55. 40 T. Hobbes, in E. Curley (ed), Leviathan: With Selected Variants from the Latin Edition of 1668, (Indianapolis: Hackett, 1994), at 104. 41 Ibid, at 146. 42 I borrow the term from P. Pettit, ‘Groups with Minds of their Own’, in F. Schmitt (eds), Socializing Metaphysics, edited by Frederick Schmitt (New York: Rowman and Littlefield, 2004), 167-93. 43 G. Bouchard and C. Taylor, ‘Building the Future: A Time for Reconciliation’, available online (archive.org): <http://web.archive.org/web/20110714070035/://www.accommodements.qc.ca/ >. 44 Ibid, at 36. 45 Ibid, at 176. 46 Taylor reiterates this view in J. Maclure and C. Taylor, Laïcité et liberté de conscience (Montréal: Boréal, 2010), 103-107. It bears noting that this interpretation is the one sanctioned by the Canadian Supreme Court in Syndicat Northcrest v Amselem, [2004] 2 SCR 551. I believe that the SCC also incorrectly ignores the libertas ecclesiae strain of religious liberty. 47 W. Kymlicka, ‘Disentangling the Debate’ in Uneasy Partners: Multiculturalism and Rights in Canada (Waterloo: Wilfrid Laurier University Press, 2007) 137-56, at 147. 19 R.K. Lippert, Sanctuary, Sovereignty, Sacrifice (Vancouver: UBC Press, 2005); L. Heredia, ‘From Prayer to Protest: The Immigrant Rights Movement and the Catholic Church’, in Kim Voss and Irene Bloemraad (eds), Rallying for Immigrant Rights, (Berkeley: University of California Press, 2011) 101-22. 49 H.L.A. Hart, The Concept of Law (2nd edn, Oxford: OUP, 1994) at 81. 50 Ibid, at 94. 51 Ibid, at 91. 52 Hart, The Concept of Law, supra note 48, at 92. 53 Even a rule that says that ‘till heaven and earth pass, one jot or one tittle shall in no wise pass from the law, till all be fulfilled’ (Matthew 5:18) is proclaiming a rule of change, albeit a quite restrictive one. 54 J. Raz, Practical Reason and Norms (New York: OUP, 1999). 55 Ibid., at 136. 56 A. Marmor, ‘The Dilemma of Authority’, 2 Jurisprudence (2011) 121-141, at 129-130 (citations omitted). 57 The status of Eastern Catholic churches is an interesting counterexample but it is, quite vividly, the exception that proves the rule. The special status of Eastern Catholic churches is as organizational as it is cultural. Its origin is found in the realignment of formal ecclesiastical communities which had historically followed rituals other than the Latin but aligned themselves with the Roman Pontiff following various crises, most important among them the Great Schism of the eleventh century. Its continued accommodation within the Roman Catholic communion necessitates a special code of canon law and special declarations during ecumenical councils (viz. Codex Canonum Ecclesiarum Orientalium, cc 1-6). 58 The permissibility of rice or nuts during Passover is often cited as an example that is allowed to Sephardim but proscribed to Ashkenazim for cultural, not religious reasons, but the halachic status of custom makes such a blanket judgment complicated. 59 Hobbes, Leviathan, supra note 39, chapter 22. 60 I borrow the language of incompatibility from Raz, who argues that ‘[a]ll legal systems … are potentially incompatible at least to a certain extent. Since all legal systems claim to be supreme with respect to their subject community, none can acknowledge any claim to supremacy over the same community which may be made by another legal system.’ Practical Reason and Norms, supra note 53, at 152. 61 A. Philips, Multiculturalism without Culture (Princeton: Princeton University Press, 2007) at 19. Philips attributes the quasi-legal view to such diverse theorists as Ayelet Shachar, Will Kymlicka, Charles Taylor, and Jacob Levy. 48