http://ptx.sagepub.com/ Political Theory http://ptx.sagepub.com/content/38/2/187 The online version of this article can be found at: DOI: 10.1177/0090591709354868 2010 38: 187 originally published online 6 January 2010Political Theory Corey Brettschneider Reasons for Rights A Transformative Theory of Religious Freedom: Promoting the Published by: http://www.sagepublications.com can be found at:Political TheoryAdditional services and information for http://ptx.sagepub.com/cgi/alertsEmail Alerts: http://ptx.sagepub.com/subscriptionsSubscriptions: http://www.sagepub.com/journalsReprints.navReprints: http://www.sagepub.com/journalsPermissions.navPermissions: http://ptx.sagepub.com/content/38/2/187.refs.htmlCitations: at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ http://ptx.sagepub.com/content/38/2/187 http://www.sagepublications.com http://ptx.sagepub.com/cgi/alerts http://ptx.sagepub.com/subscriptions http://www.sagepub.com/journalsReprints.nav http://www.sagepub.com/journalsPermissions.nav http://ptx.sagepub.com/content/38/2/187.refs.html http://ptx.sagepub.com/ Articles Political Theory 38(2) 187 –213 © 2010 SAGE Publications Reprints and permission: http://www. sagepub.com/journalsPermissions.nav DOI: 10.1177/0090591709354868 http://ptx.sagepub.com A Transformative Theory of Religious Freedom: Promoting the Reasons for Rights Corey Brettschneider1 Abstract Religious freedom is often thought to protect, not only religious practices, but also the underlying religious beliefs of citizens. But what should be said about religious beliefs that oppose religious freedom itself or that deny the concept of equal citizenship? The author argues here that such beliefs, while protected against coercive sanction, are rightly subject to attempts at transformation by the state in its expressive capacities. Transformation is entailed by a commitment to publicizing the reasons and principles that justify the basic rights of citizens. Keywords religious freedom, free exercise, citizenship, rights, accommodation I. Introduction Defenders of religious freedom often seek to protect existing religious beliefs from the influence of the state. They therefore tend to adopt a “static” con- ception of religious freedom: they assume that the current beliefs and practices of any given religion merit protection simply because they are reli- gious in character. On a static conception, state influences that lead to changes in these beliefs are at least presumptive violations of religious freedom. The 1Brown University, Providence, Rhode Island Corresponding Author: Corey Brettschneider, Brown University, Box 1844, Providence, RI 02912 Email: Corey_Brettschneider@brown.edu at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 188 Political Theory 38(2) “accommodationist” accounts of Michael McConnell and Martha Nussbaum are prominent examples of the static conception.1 Both of these theorists seek to protect existing religious beliefs from the influence of state policy, regard- less of whether change is sought intentionally or unintentionally by the state. Religious beliefs in general, they argue, have a special weight in the public forum because they are linked to the value of religion itself. For those with static conceptions of religious freedom, therefore, even unintentional bur- dens by the state on religious beliefs are problematic; intentional attempts at transforming these beliefs are an even more dangerous threat to religious liberty. The static view is often assumed, not only in constitutional jurisprudence, but also in normative accounts of religious freedom. For example, it is seen as being required by the account of legitimacy offered in John Rawls' theory of political liberalism. Many theorists argue that political liberalism should be concerned to preserve existing religious beliefs when forging an overlap- ping consensus between reasonable comprehensive views. Nussbaum has prominently defended the position that political liberalism requires what amounts to a static notion, and several others have advanced a similar con- nection.2 The static view is also assumed in prominent accounts of religious freedom that lie outside the political liberal framework.3 Fundamental to these accounts is the contention that the liberal state should, in addition to respecting religious practices, refrain from interfering at all with religious belief, intentionally or otherwise. In this essay, I appeal to a conception of political liberalism in arguing against static models of religious freedom and defend what I call a “transfor- mative” model of religious freedom. I argue that any robust conception of religious freedom will find itself at odds with some existing religious beliefs and that defenders of religious freedom should favor a role for the state in seeking to change some religious beliefs. In particular, I suggest that the state has an interest in actively promoting the shared values of free and equal citizenship—values that are rightly understood to underlie the rights of reli- gious freedom. I argue that when the state seeks to transform religious beliefs that are at odds with the shared reasons for rights necessary to the state’s legitimacy, it does not violate religious freedom or freedom of conscience when it complies with two limits. First, the state should limit its attempts at transformation to those beliefs that are fundamentally at odds with the shared reasons for rights. Second, it should pursue transformation through its expres- sive, rather than its coercive, capacities. My argument proceeds in three stages, each of which is designed to allay fears that a transformative account of religious freedom is antireli- gious and antiliberal. First, I argue that religious freedom is a principled at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 189 commitment that requires the transformation of religious viewpoints at odds with it. Second, I place this transformative account in the context of a wider theory of political liberalism. I argue that this transformative account does not require an all-or-nothing choice between a commitment to liberalism or to religion. Finally, I emphasize the compatibility of my transformative account of religious freedom with basic rights such as freedom of association. Throughout this article, I draw on U.S. Supreme Court cases related to the free exercise clause of the First Amendment. But I do so to develop an argument in political theory about how religious freedom should be understood, not to affirm existing case law or to develop a new free-exercise jurisprudence. II. Religion v. Religious Freedom: The Lukumi Principle On a static analysis of religious freedom, such as that suggested by accom- modationist thinkers, the first concern should be whether laws intentionally or unintentionally burden religious practices. On this view, an account of religious freedom must be sensitive not only to the dangers of the state’s directly attacking religious belief but also to unintended effects of policy on religious exercise. Accommodationists assume that religious freedom is endangered whenever existing religious beliefs are burdened or changed.4 Accommodationist thinking figured strongly in the Supreme Court’s juris- prudence prior to Employment Division v. Smith in 1990.5 A fundamental flaw in static views that associate religious freedom with the preservation of existing religious beliefs is that they cannot account for the potential tensions between these beliefs and the desire to protect religious freedom. Consider, for instance, the Court’s decision to strike down munici- pal legislation prohibiting animal sacrifice in Church of the Lukumi Babalu Aye v. City of Hialeah. In this case, the city of Hialeah, Florida, had inten- tionally targeted the practice of the Santeria religion, which involves the sacrifice of live animals. The legislation was designed to single out animal sacrifice in particular for criminal sanction, although similar forms of animal slaughter were permitted. A unanimous Court suggested that this legislation could not be explained on any basis other than animus toward the Santeria religion. I suggest that we study this example to understand the tensions pres- ent in thinking about the political morality of religious freedom. In particular, I wish to use this case because it is commonly thought to be a paradigmatic example of the judicial protection of religious freedom. I hope to push this common understanding of Lukumi to show that religious freedom at times at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 190 Political Theory 38(2) requires attempts at transformation. My aim is not to oppose the accommo- dationists’ claim that religious freedom requires some exemptions to general laws. Rather, I examine the philosophical inadequacy of such an approach when issues of equal citizenship are implicated. Since the ordinance at issue in Lukumi hindered religious practices, a static analysis would likely suggest that it should be struck down. Not only did the Hialeah city councilmen who passed the law limit the religious prac- tice of animal sacrifice, but they apparently did so with the intent of burdening the Santeria religion specifically. The flaws and limits of the static approach, however, can be found by looking at the transcript of the meeting in which the city as a whole discussed the ordinance. The transcript, quoted in Justice Kennedy’s majority opinion, reveals that the ordinance itself was religiously motivated. One councilman justified his opposition to the practice of animal sacrifice by stating, “I don’t believe the Bible allows that.” The chaplain of the Hialeah police department variously described Santeria practices as “an abomination to the Lord” and the worship of “demons.”6 Thus, whatever the Court’s decision in Lukumi, there would have been an adverse effect on some religion. The Santeria practitioners would have been adversely affected if the law had been allowed to stand. On the other hand, striking down the law would apparently burden the religious views of the councilmen and others, who hold that it is a Christian moral duty to ban animal sacrifice. In striking down the ordinance, the Court not only protected the Santeria religion from illiberal, coercive legislation. It also served to condemn the illiberal beliefs behind the legislation. The fact that these beliefs were motivated by religion does not immunize them from condemnation. As I see it, the Court’s decision to strike down this law also expressed the message that those particular views of the councilmen that led them to pass the ordinance, regardless of whether they were religious or not, have no place in a free society’s deliberations about coercion. We are thus confronted by a seeming paradox of religious freedom. It is usually understood that religious freedom protects certain religious practices. In a related vein, religious freedom is thought to protect the religious beliefs that underlie such practices. But not all religious practices, and not all reli- gious beliefs, should be protected. Some religious practices and religious beliefs are at odds with the principle that citizens should be allowed to prac- tice their religion and to believe what they wish free of state sanction. These beliefs are rightly condemned and discouraged by the legitimate state. In particular, the religious “practice” of the councilmen to ban the Santeria religion—as well as the religious beliefs behind this practice—are rightly subject to transformation. Specifically, the case of the Hialeah ordinance suggests that those who seek to use the power of the state to restrict a religious at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 191 belief or practice, even when motivated by their own religious beliefs, should be stopped as well as criticized. Of course, when a law such as the Hialeah ordinance is struck down on First Amendment free exercise grounds, those who passed the law are not sanctioned through criminal law. But the Court’s decision should be interpreted, on my view, to legally prohibit government officials from enacting policies at odds with the ideal of religious freedom. The decision also sends a message that the beliefs and arguments behind such statutes, even if religious in nature, are incompatible with religious freedom. This example suggests that an account of religious freedom must be able to distinguish between two sorts of existing religious beliefs: those that are inconsistent with the ideal of religious freedom—such as the councilmen’s views, which motivated them to pass the Hialeah ordinance—and those that are compatible with the ideal of religious freedom. Unlike the views of the councilmen, for instance, nothing in the Santeria practice of animal sacrifice suggested the need to use the force of law to impose beliefs on others. The Court’s invalidation of the Hialeah ordinance sends the message to its propo- nents that their reasoning fails to comport with the ideal of religious freedom. The example reveals how a principle of religious freedom might itself con- flict with existing religious beliefs. In the face of such conflicts, a principle of religious freedom is transformative with regard to religion. Regardless of Justice Kennedy’s personal intentions, I contend, Lukumi highlights why an account of religious freedom might conflict with, and require the transformation of, an existing set of religious beliefs. Toward the end of the opinion, Kennedy writes, The Free Exercise Clause commits government itself to religious toler- ance, and upon even slight suspicion that proposals for state intervention stem from animosity to religion or distrust of its practices, all officials must pause to remember their own high duty to the Constitution and to the rights it secures.7 The fundamental question is what we should do when “animosity to religion” is itself based in religion. I propose that, according to what I call the “Lukumi principle,” the commitment to “religious tolerance” is twofold. First, it entails the prohibition of any law that would use the state to limit the religious practices or beliefs of others—even if such hostility is based in religion itself. Second, the principle entails the condemnation of the beliefs that directly motivate such laws. When religious views conflict with religious freedom, therefore, the latter must trump the former. Importantly, the phrase “animosity . . . or distrust” cannot refer only to intentional hatred. Even if attempts to limit religious practice are based in the love that might at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 192 Political Theory 38(2) be said to come with a desire to convert others to a “true” religion, they still must be rejected as incompatible with religious freedom. The state therefore rightly challenges religion when religious views conflict with the principle of religious tolerance itself. I want to be clear about precisely which kinds of motivating beliefs are rightly subject to condemnation by the state. The councilmen in Lukumi believed that their general set of religious beliefs supported the ordinance; but it is not their general set of beliefs that should be condemned. Some of these beliefs can be set off from what they took to be the implication for out- lawing the religious practices of others. For instance, as I elaborate in the next section, a religious belief in God is compatible with the requirements of the Lukumi principle. The motivating beliefs that are rightly condemned can be defined as those that linked the councilmen’s general religious beliefs to the conclusion that Santeria should be legally prohibited. The councilmen’s belief that their religion required legally sanctioning the Santeria religion is incompatible with the ideal of free and equal citizenship, and is rightly sub- ject to transformation. Some might object that I am mistaken in discussing the beliefs of authors of and deliberators about law rather than restricting my concern to criticizing their actions after the fact. One could argue that what threatened religious freedom was not the beliefs of the proponents of the law per se but rather the fact that they specifically issued an ordinance restricting the practice of another religion. On this view, religious freedom is not about limitations on belief at all but rather about limits on the kinds of actions lawmakers can take—or, more specifically, on the kinds of laws they are allowed to enact. I think it is a mistake, however, to separate belief and action in this way. Although the Court is concerned to protect rights against certain kinds of state action, in its opinions and reasoning it also addresses the kind of prin- ciples that should underlie law. When it strikes down illegitimate laws, it does so often because the reasons and beliefs for such laws violate public principles central to the state’s legitimacy. These beliefs, moreover, are not only relevant after a law has successfully been passed. If, for instance, the Hialeah councilmen were unsuccessful in passing the ordinance at issue, their flawed beliefs about coercion would still rightly have been subject to public scrutiny. Therefore, in rulings like Lukumi, the Supreme Court acts as an “exem- plar” of public reason in two senses. First, it acts as an enforcer of public reason by striking down unconstitutional laws, such as those that constrain the practice of one religion because it is viewed as incompatible with another. Second, it acts as a model for the wider citizenry, including public officials who deliberate about and make law, when it explains why such laws at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 193 are illegitimate and when it speaks in defense of the values of free and equal citizenship.8 Supreme Court opinions certainly are concerned with the imme- diate task of whether to strike down law. But their audience also rightly includes all those potentially involved in lawmaking, including citizens con- cerned to think and deliberate publicly about lawmaking. In Lukumi specifically, the primary purpose of the Court’s decision was to strike down a law violating religious freedom. But it did more than this. The decision also expressed disapproval of the reasons its proponents gave for passing this law, and it issued a rebuke to them for violating their duty to respect religious freedom. The religious nature of the reasons advanced by the law’s proponents does not immunize them from this rebuke. Nor should they be shielded from criticism by their right of free-speech, which protects them from being punished merely for holding their views. They may still be criticized by those who defend the core principle of religious freedom. For example, the Court in Lukumi expressed a principled disapproval of the kind of religious reasoning that would use the state to impose religious beliefs on others; it therefore directly criticized the views of the Hialeah law’s propo- nents. The decision, moreover, is relevant for all those who would attempt to use similar reasoning to pass equivalent laws in other locales. It is especially important that decisions such as Lukumi serve as an example for the wider citizenry in a democracy in which citizens vote not only for repre- sentatives but at times directly for legislation. These opinions can warn citizens about the pitfalls of passing unconstitutional laws that would impose one set of religious beliefs on fellow citizens. However, given that most Supreme Court opinions are not widely read, the ruling in Lukumi is not itself sufficient to pro- mulgate the principle established in it. Wider publicity and promotion of the Court’s holding—and, as importantly, its reasoning—are necessary if its exam- ple is to transform citizens’ thinking. A variety of institutions within the legitimate state should take on this task of publicity and promotion. While it is commonly recognized that the content of rights should be widely promulgated, I want to emphasize why it is also important to promote the reasons for rights. These rea- sons include the Lukumi principle and the more general ideal of equal citizenship, which underlies rights to religious freedom. Such a task should not be left to any one organ of the state, including the Supreme Court. Rather, the promul- gation of the reasons that underlie rights is a diffuse duty incumbent on all state actors and citizens. When the state seeks to transform beliefs at odds with the values and rea- sons that underlie religious freedom and other basic rights, it should respect two distinct limits. The first concerns the method through which transforma- tion should be attempted and the principle of equal citizenship should be promoted. The second concerns limits on the content of the beliefs targeted at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 194 Political Theory 38(2) for transformation. The first, which I call the “means-based” limit, requires that the state not pursue transformation of citizens’ views through any method that violates fundamental rights such as freedom of expression, conscience, and association. For example, the state cannot prohibit meetings of the Anti- Santeria Society or threaten criminal sanctions against the Hialeah councilmen for holding and expressing beliefs that conflict with the Lukumi principle. How- ever, a public articulation of why the councilmen’s views are inconsistent with religious freedom would not violate their rights, even if this point were articulated by state officials. The Lukumi principle ideally should be defended by those in power as well as by citizens participating in democratic debate. Ideally, citizens and their representatives will successfully defend the Lukumi principle such that laws that would violate will not come before the court. As I have argued elsewhere, it is better for public principles to be defended through majoritarian processes than through judicial review.9 It is only as a last resort, when lawmakers and citizens fail to heed this duty, that they are rightly rebuked by the courts. On the view developed in this essay, the state can avoid violating the “means-based” limit by confining its attempts at transformation or promotion to its expressive rather than its coercive capacities. For example, public offi- cials and citizens may make arguments that defend the Lukumi principle when engaged in public discussion, even in the face of religious opposition. Moreover, as I suggest in the next section, there is a role for educators to teach the importance of religious toleration even if some parents hold reli- gious objections. The right to hold and express a belief at odds with the ideal of equal citizenship does not entail a right to hold it unchallenged. I take the position that mere reasoning by the state addressed to citizens is not tantamount to coercion. It is a familiar Millian point that the fact that I have a right to free speech does not keep you from convincing me that some of my opinions are wrong. Moreover, if you are successful in persuading me through reason, and my views are transformed, this is certainly not evidence that I have been coerced. To the contrary, we tend to regard such “transformations” as freely chosen. One need not be a full-blown Kantian to accept as much. When we move from the question of individual persuasion to state persua- sion, we should observe that the state has a particular interest in advancing principles, such as the ideal of religious freedom articulated in Lukumi, that are essential to its own legitimacy. It is also essential that the state use as much as possible the means of reasoning to defend its most fundamental principles. Like the Supreme Court, state officials should reason by means of an appeal to public principles. It might be objected that because of the state’s massive power, its own expression can be in a sense overwhelming. If the only viewpoint that at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 195 citizens were able to hear on the subject of religious toleration was the one articulated by the state, certainly state expression might be tantamount to propaganda. However, this merely suggests why the state’s voice must not preclude a diversity of opinion in the polity at large. The state should there- fore work to establish and protect public forums in which citizens are free to express views at odds with its own. Another objection might emphasize the fact that particular state officials will not always advance the Lukumi principle—indeed, they might contradict it. Yet this no more would be an argument against a role for state expression of core principles of legitimacy than the fact that state officials sometimes enact unprincipled law would be an argument against principled limits on state action. In sum, the means- based limit emphasizes the importance of reasoning on behalf of the principles of the legitimate state. The second limitation on attempts at transformation, which I call the “substance-based” limit, concerns the kind of beliefs the state is rightly con- cerned to transform, and when it is justified in doing so. It is necessary not only that the state use its expressive, rather than coercive, capacities in defending the Lukumi principle but also that it endeavor to challenge only those religious beliefs that both violate the ideal of free and equal citizenship. There are a variety of religious beliefs that might appear to violate the Lukumi principle, but which need not affect our capacity, to see each other as free and equal. For instance, one might believe that all non-Christians are condemned to hell but also that sinners cannot be saved through legislation and that therefore theistic reasoning has no role in lawmaking. Because this view says nothing about which beliefs should be imposed on others, it is not publicly relevant. The need for transformation arises only in regard to those views that conflict with the ideal of free and equal citizenship, including those views that would seek to impose by law religious beliefs at odds with this ideal. As such, it is not the belief in the superiority of one’s religion per se that transformation should target. Moreover, the substance-based limit suggests that it is only those beliefs most blatantly at odds with the principles of equal citizenship that are rightly subject to transformation. For instance, reasonable people might disagree about the implications of equal citizenship for issues such as affirmative action—and no position on this issue should rightly be subject to transformation. Rather, it is those beliefs, religious and otherwise, that are openly hostile to or implausibly consistent with the values of equal citizenship that the state should seek to transform. These two limits offer some traditional privacy protections. In emphasiz- ing why it is beliefs about coercion that are rightly targeted by the state, I have left immune from persuasive interventions many beliefs about, for example, the internal structure of the family and civil society. For instance, at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 196 Political Theory 38(2) the government’s scrutinizing my dinner guest list would be a clear misap- plication of the need to promote the Lukumi principle. Nor should the state, for example, spy on citizens to determine whether they hold beliefs that con- flict with the ideal of free and equal citizenship. Such a practice violates the means-based limit by employing coercive, not expressive, methods. But it also violates the substance-based limit because it unjustifiably targets citi- zens who have not demonstrated that they seek to use the force of law to “enforce” views that are antagonistic to public reason.10 As my discussion of Bob Jones University v. United States will illustrate, however, beliefs about internal affairs are not always immune from public scrutiny.11 It might be objected that, even if we accept the authority of the state to target beliefs at odds with core liberal principles such as a commitment to religious freedom, this need not commit us to a purposeful transformation of religious belief per se. One could argue, drawing on Rawls’ distinction between neutrality of effect and neutrality of intent towards religion, that the aim of state action should be to challenge illiberal beliefs at odds with the ideal of free and equal citizenship. On this account, we should think of the transfor- mation of religious beliefs as an acceptable but unintended consequence of challenging illiberal beliefs.12 But this doctrine ignores the fact that some illiberal beliefs are explicitly religious. The Hialeah councilmen’s arguments, for instance, are couched entirely in terms of the need to save practitioners of Santeria from a false religion. When one councilman said that he was “totally against the sacrificing of animals” but that kosher slaughter was acceptable because it had a “real purpose,” he was expressing a belief that denied the equality of Santeria’s adherents, and a belief that cannot be explained by any other than religious grounds. The city’s Police Department chaplain, too, was guilty of contradicting the ideal of free and equal citizenship when he made a leap from a statement that is protected by the Lukumi principle (“We need to be helping people and sharing with them the truth that is found in Jesus Christ”) to a statement that violates the principle (“I would exhort you . . . not to permit this Church to exist”).13 There is, in short, no way of getting around the fact that attempting to transform an illiberal belief is sometimes tantamount to attempting to inten- tionally transform a religious belief.14 Indeed, Rawls maintained, despite his claim of “neutrality of intent” toward religious doctrines, that he hoped unreasonable religions would become reasonable over time. But I have argued that the account of transformation should not, as Rawls suggests, be a mere “hope.”15 It is rather a commitment that is fundamental to the idea of religious freedom itself. at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 197 III. Dialectics of Religious Transformation In this section, I suggest how the transformation of illiberal religious beliefs could be achieved, and I explain why such transformation does not constitute an attack on religious freedom itself. Susan Okin, whose work on multicul- turalism is one example of a transformative account of liberalism, concedes that the promotion of certain liberal values, in particular those of equality, might cause some cultures—and, we can gather, some religions—to go “extinct.”16 A challenge to my view of transformation might suggest that, like Okin, I risk extinguishing religion. Moreover, does my view that the state should seek to transform certain illiberal religious beliefs force a choice between the secular and the religious? I respond to the concern about religious extinction by emphasizing why the transformative theory does not entail an all-or-nothing choice between religious freedom and the continued existence of religious beliefs. If it did, I would have undermined my own claim to be offering a defense of religious freedom. For two reasons, I contest the claim that any abandonment of “static” models of religious freedom is an attack on religion. First, I argue that the relevant type of transformation is dialectical in the sense that it does not force a stark either-or trade-off between religion and core liberal princi- ples. Rather, it suggests a synthesis between liberal principles and religious views. In addition, at times this synthesis might occur when religious tradi- tions help to clarify the very meaning of these principles. Second, the type of transformation called for by norms of legitimacy cannot be forced on any religious groups. Although religious freedom ideally results in dialectical transformation, I emphasize that it also protects religious groups that choose to resist such transformation. I therefore use “dialectic” in its traditional sense: to mean only that religious beliefs and the freestanding ideal of equal citizenship can result in new and varied combinations or “syntheses.”17 In elaborating on why the transformative account of religious freedom is dialectical, I want to begin by highlighting a distinction, suggested by Rawls, between discrete principles that are “freestanding” from particular compre- hensive conceptions and beliefs that are comprehensive in themselves. I say that the former principles are “discrete” to emphasize the fact that they can be separated from the web of wider beliefs with which they often intersect. Freestanding beliefs serve as the basis for a legitimate society because they are based solely on what is owed to citizens in virtue of their being regarded as free and equal. They thus stand in contrast to comprehensive beliefs, which are often linked to persons’ understandings of deep metaphysical and reli- gious questions.18 A commitment to religious freedom is itself an essential at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 198 Political Theory 38(2) aspect of a freestanding political conception. I want, however, to broaden the discussion to consider the broader and underlying freestanding commitment to an ideal of free and equal citizenship. On Rawls’s view, religious beliefs would ideally reinforce freestanding beliefs that are grounded in the ideal of free and equal citizenship. In such instances, religious conceptions and the fundamental commitments neces- sary for liberal legitimacy would be in harmony, and there would be no need for transformation of beliefs. Such an idealized view pervades much of the discussion about the role of religion in contemporary political theory. Rawls and Amy Gutmann, for instance, cite Martin Luther King, Jr.’s arguments against segregation as an example of how religious identity can enhance the liberal commitment to justice. Gutmann writes that King’s “civic genius was his ability to move between religious and secular sources of the same politi- cal argument in the course of communicating to his fellow citizens, showing that the religious identities of believers can be a powerful public force in sup- port of democratic justice.”19 King’s distinctive contribution appealed to Christian morality to promote norms of equal citizenship.20 Although King serves as an example of an ideal synthesis of equal status and religious values, many cases also arise in which particular religious views come into conflict with freestanding principles that are fundamental to the state’s legitimacy or indeed come into conflict with religious freedom itself. In such cases, the Lukumi principle should be invoked. The Lukumi principle demands that when religious views oppose the ideal of religious freedom itself—for example, because they demand the imposition of their views on nonbelievers—the state must challenge the beliefs in question. Moreover, the principle is a freestanding one that is not based on any com- prehensive doctrine. It is grounded in an attempt to think about the kind of religious freedom that would be protected in a society of free and equal citi- zens. By extension, when particular religious beliefs are at odds with freestanding principles of free and equal citizenship more generally, the state seeks to transform those existing beliefs to defend fundamental freestanding principles. Thus, while my theory retains a commitment to freestanding princi- ples and avoids appeal to comprehensive conceptions, it rejects the static claim that every existing religious view could serve to reinforce freestanding principles. The suggestion that religious freedom is a principle detachable from and inde- pendent of actual religious beliefs might seem at odds with what Rawls has called the “wide” or “inclusive” view of public reason. But I believe that the way this idea has sometimes been understood risks underplaying the transformative char- acter of these ideas. The idea of inclusive public reason is meant to suggest why religious claims have a place in public discussion. On Rawls’s view, there is no at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 199 reason to prevent citizens from making public arguments from religious view- points so long as they also provide arguments from freestanding principles. But it would be wrong to suggest that this principle somehow prioritizes existing religious beliefs in the manner of the static view. Inclusive public reason requires that arguments that are couched in reli- gious language also be translatable into freestanding arguments that address citizens solely in light of their status as free and equal, without appealing to religious or other comprehensive views. Freestanding public reason, there- fore, serves as a standard by which we can evaluate religious claims. For this reason, the idea of inclusive public reason reinforces, rather than undermines, my argument that the state must seek to transform existing religious beliefs when they conflict with freestanding principles. When religious arguments work in concert with arguments from the freestanding principles of free and equal citizenship, religious beliefs rightly serve to refine and reinforce free- standing principles. By contrast, when religious arguments work to undermine freestanding principles, the state rightly seeks to transform the existing reli- gious beliefs in question. For this reason, I suggest that my view is dialectical. At times, existing religious beliefs can serve to clarify and reinforce our com- mitments to freestanding principles of free and equal citizenship. At other times, existing religious beliefs will undermine those principles; in such cases, existing religious beliefs are rightly targeted by the state for transformation. Freestanding principles, then, have a moral priority over conflicting existing religious beliefs and might sometimes require their transformation. For this reason, as I have argued, my view is not a static one. This position conflicts with a common misunderstanding of Rawls’s theory, partially due to mislead- ing labeling. It has often been thought that dedication to an “overlapping consensus” requires basing public principles on an existing or “latent” intersec- tion of secular and religious views in the public political culture. Rawls explicitly rejects such an understanding, however, and his reasons demonstrate why his view is transformative and not static. In short, freestanding principles, such as religious freedom, are often at odds with currently existing views. Rawls’s idea of the overlapping consensus and the freestanding principles present within it are ideal political conceptions. When they conflict with existing religious views, therefore, we should “work backward” to seek to transform those views at odds with this ideal.21 Having argued that society’s commitment to freestanding principles of equal status for all citizens commits the state to seeking the dialectical trans- formation of religion, I turn now to the contention that my view constitutes an “attack” on religion. Does the commitment to freestanding principles of liberal legitimacy, such as religious freedom and the concern to treat citizens at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 200 Political Theory 38(2) as equals, require that we ask those who hold religious views incompatible with these principles to abandon religion itself? Some liberal political theorists in the Rawlsian tradition, such as Stephen Macedo and Rob Reich, have suggested that, at times, this might be the case. Reich, for instance, argues that citizens must be committed to the individual rights of children and that the “state must take its own side” when these rights conflict with the norms of identity groups.22 Reich points to examples that illustrate that the divide between the state’s rightful commitment to princi- ples of legitimacy and the commitment of some identity groups to other conflicting values might sometimes be unbridgeable. Macedo emphasizes why conflicts in education policy between religious views and secular com- mitments to free and equal citizenship might be irresolvable through deliberation. He argues that the case of Mozert v. Hawkins County Board of Education,23 for instance, pits the norms of a liberal society that requires respect for a variety of cultures against a religious worldview that would exclude even knowledge of these cultures.24 Such respect requires, at mini- mum, basic knowledge of other cultures, thus rendering the conflict between the religious viewpoint and liberal norms intractable. Although these liberal theorists might acknowledge that we should seek to avoid these confrontations through innovative policy, they help to elucidate the fact that, in some instances, there will not be an easy policy solution to conflicts between religious beliefs and the commitment to equal status. Even if we concede that Macedo and Reich are correct that conflicts between norms of legitimacy and some comprehensive conceptions are inevitable, however, such conflicts do not require that we sacrifice religious beliefs alto- gether when we attempt to transform them to be compatible with norms of legitimacy. That the comprehensive doctrines of some identity groups are in partial tension with freestanding principles such as religious freedom or equal respect does not mean they must be rejected entirely. Freestanding principles can be endorsed by and derived from a variety of comprehensive conceptions of the good, both religious and secular. The standard of legiti- macy thus may require the transformation of identity, but it does not demand the replacement of one identity with another. Again, the model here is not all-or-nothing but dialectical. The value of equal status might interact with religious norms to produce changes in identities that make them compatible with liberal values. Consider, for instance, the story told by Anne Fadiman in her book, The Spirit Catches You and You Fall Down, in which a Hmong family’s religious values led them to reject medical care for their daughter’s epilepsy.25 If we are committed to the individual rights of the child, it might be thought that we need to reject an identity featuring unbridgeable conflicts with principles of at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 201 legitimacy. It is important to notice, however, that although the state might require that medical treatment for a minor be imposed on a religious com- munity against its wishes, the members of the community might, over time, grow to accept such a requirement without abandoning their religious iden- tity. Ideally, they would learn to accommodate this specific policy based on their considered acceptance of the discrete principle, in accordance with lib- eral principles of legitimacy and not just religious views, that decisions about medical care must incorporate a concern to care for their children. Religious identity itself might change to accommodate modern medicine in such a case, but it is important to note that religious groups would not be required to aban- don their beliefs. Even if some change in identity were necessary, only very conservative notions of culture (e.g., Devlin’s) equate any change in a culture with its destruction.26 Religious identities might be transformed through a dialectic between specific norms of legitimacy and general comprehensive conceptions so as to become more liberal without being abandoned. Indeed, if we examine two of the most famously contested cases that pitted the freedom of religious belief against norms of equal status, we can see that these beliefs were changed rather than abandoned. I begin with the case of Bob Jones University v. United States.27 Bob Jones University is an evangelical Christian school in Greenville, South Carolina, that had banned interracial dating and maintained other racially dis- criminatory policies. Bob Jones administrators considered these racially discriminatory policies to be grounded in the institution’s understanding of the Bible. When the IRS instituted a new policy that withheld tax-exempt status from groups that discriminated based on race, it revoked Bob Jones’s nonprofit status. Defenders of the static view, who espouse the principle of accommoda- tion, might suggest that Bob Jones should have been given an exemption from the policy because its discriminatory policies were based in religion. I want to first suggest why the IRS action against Bob Jones is an example of the kind of transformation I have been defending here. Because this is a harder case than that of Lukumi, it will help to deepen and clarify my argu- ment for the Lukumi principle. Some might contend that, unlike the town council’s ordinance in Lukumi, which clearly involved state action, Bob Jones’s policy involved only a private belief. It could be argued, for instance, that even if these policies were morally objectionable, they were not grounded in a belief that the political rights or equal citizenship of others outside the university should be undermined. On this view, we might see Bob Jones’s policy as merely an internal matter of campus organization, not relevant to beliefs about public policy. A close look at the facts, however, gives one indication of why Bob Jones’s policy cannot be distinguished from public advocacy of beliefs at odds with at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 202 Political Theory 38(2) the ideals of free and equal citizenship. Namely, the school explicitly extended its policies to punish those who were members of groups that advo- cated interracial marriage in the wider society.28 This broad provision prohibited students not only from expressing support for interracial marriage while on campus but from even holding membership in a public organization such as the NAACP. I take such a position to be at odds with at least one core right of equal citizenship: the right of interracial marriage, protected by the Supreme Court in Loving v. Virginia.29 This would be a harder case if Bob Jones’s policies did not include such an explicit provision against public support of interracial marriage. It might be logically possible for a school like Bob Jones to have an internal policy that prohibited interracial dating and marriage but that said nothing about whether the university was opposed to the legality of interracial relationships in the wider society. Absent some explicit clarification, this policy would likely be perceived by students to express disapproval of the legal recogni- tion of interracial relationships even outside its campus. Given Bob Jones’s status as an educational institution and its concern to instill moral values, few students might distinguish an ostensibly internal message about race rela- tions from a claim about the status of blacks and whites as equal citizens. However, an inquiry into students’ perceptions about the wider implica- tions of the policy would not be determinative on this issue. A more conclusive inquiry should ask whether Jones could have some principled reason for believing its prohibition on interracial dating was compatible with an endorsement of the legality of interracial marriage. Such a principle would have to avoid any appeal to worries about so-called miscegenation in the larger society—worries that are incompatible with the ideal of free and equal citizenship. For instance, a Presbyterian institution might prohibit interfaith marriage among its students out of a concern to preserve its own community of faith without suggesting any implications about the acceptability of inter- faith marriage, which is protected by law, in the wider society. I doubt that an institution like Bob Jones has any such justification available. In sum, what makes Bob Jones’s policy subject to transformation is its direct affront to the ideal of equal citizenship. The university does not merely object to a law, or even a basic right, but rather, because of its racist policy, directly challenges the very idea of equal citizenship. It is important to note the significant difference between the kind of pres- sure an institution like Bob Jones experiences in the threatened loss of nonprofit status and the kind of pressure citizens might feel in the face of purely expressive state action. Bob Jones faced financial, not merely persua- sive, pressure to change its views. Yet despite the “quasi-coercive” nature of such pressure, it is still justifiable. Nonprofit status is a tax advantage that at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 203 should be linked at minimum to an institution’s willingness not to undermine the ideal of free and equal citizenship. On my view, institutions granted this status do not have to actively promote these public values, but they do have to be willing not to undercut them. Bob Jones might enjoy an intuitive advan- tage in the argument about its nonprofit tax status because its continued existence might seem to depend on this status. However, the intuition might push in the other direction if we ask whether an institution with these views has a right to be founded in the first place. There is a difference, moreover, between institutions that should be charged, at minimum, with not undermin- ing public values and, for example, a household. While the tax benefits enjoyed by a university can be linked to its obligation not to undermine public values, citizens are not assigned a tax status based on their beliefs about free and equal citizenship. In sum, although the pressure from the IRS is more intense than mere reasoning, the option to resist or ignore transfor- mation, on my view, means that its rights were not violated. In 2000, seventeen years after the Supreme Court ruled against the school, the president of Bob Jones University appeared on television to announce that the school was lifting its policy against interracial dating. In addition, a page now exists on the school’s Web site with a statement that expresses regret for having “allowed institutional policies to remain in place that were racially hurtful.”30 Despite the university administration’s rhetoric about how its discriminatory policies were based in fundamental religious beliefs, it is far from clear that Jones is a less religious institution because it now permits interracial dating. Unsurprisingly, these changes have not prevented the school from enforcing a strict code of sexual conduct. Regardless of the reasons why Bob Jones decided to change its policies, its transformation did not involve an all-or-nothing change. As was the case when Bob Jones began to admit black students, the decision to end the racially discriminatory policies in question did not require the abandonment of its religious belief altogether. Indeed, even if we concede that Bob Jones’s particular commitment to racially exclusive policies was genuinely religious and that the state intended to transform this religious belief, the university remains undeniably a religious institution, despite its abandonment of these particular views. So far, I have largely focused on the ways that conflicts between commit- ments to equal citizenship and religious identity can result in the transformation of particular group identities. As I have suggested, however, at times identity groups themselves can help clarify a society’s understanding of public values such as equal status. The case of Mozert cited by Macedo, for instance, could easily be seen as “all-or-nothing.” In that case, a mother objected to her child’s being subject to a certain curriculum, including a textbook that at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 204 Political Theory 38(2) included non-Biblical literature and information about other cultures. The court in Mozert refused to grant a free-exercise objection. The case is thus a classic instance of religious belief being pitted against public values, such as respect for and knowledge of other cultures. But the story does not end there. The litigant who brought the case continued to engage in public debate and has arguably come to hold a view that is compatible with my conception of religious freedom. As Archon Fung has described in a detailed look at the mother’s views, her new position was not that the textbook objectionably included non-Christian perspectives but that the curriculum failed to empha- size that women who choose to stay home to raise a family also have a legitimate conception of the good, which the state ought to recognize.31 I take this case to illustrate both that religious perspectives can change to become more publicly justifiable and that the religious perspectives them- selves can generate and clarify public values. The claimant in this case was right that the perspectives of homemakers should be included, and the text- book’s exclusion of this viewpoint may have resulted from a position that was not publicly justifiable because it was excessively secularist. One way of understanding this dialectic is to see it as an argument against forcing a choice between religion and public values. But I think this case demonstrates the opposite. The dialectic emerged precisely because the court forced the claim- ant to choose between public values and the discrete aspects of her religious beliefs that conflicted with those values. Her beliefs both became more rea- sonable and resulted in a position that educated the public at large.32 I now want to address one final challenge to my contention that my view is not hostile to religion itself. I have thus far sought to demonstrate that a religious viewpoint’s transformation does not entail its destruction. The evi- dence for this is that there are cases when religious viewpoints were able to change yet survive. But despite these specific instances of dialectical change, one might counter that part of the essence of religion is its insularity from public culture and principles. On this view, a religious doctrine could not, without destroying its religious character, be open to change by appeal to a specific discrete liberal principle. Suppose, for instance, that a particular citi- zen of Hialeah derived her commitment to outlawing the practice of Santeria from a particular passage in the Bible prohibiting idolatrous sacrifice. If this citizen’s religious views required a literal reading of the entire Biblical text, she could not abandon a discrete commitment to outlawing idolatrous sacri- fice without abandoning Biblical literalism itself and therefore her entire comprehensive religious view.33 Here, I believe, I must bite some bullets. If defenders of religion insist on defining it by its insularity, I respond that my view is clearly incompatible with some religions. As I have argued, however, many religions are at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 205 decidedly not marked by insularity and illiberality. Such religious views not only are entirely consistent with the principles of free and equal citizenship, but they also support them from within. I believe, moreover, that the static nature of such an insular account of religion ignores the reality that religions have survived for centuries precisely because they are able to evolve—not only to fit various cultural contexts but also to incorporate fundamental values such as those that I have suggested form the basis of a free society.34 IV. The Right to Resist Religious Transformation and the Freedom of Association I have suggested an alternative to static conceptions of religious freedom; on my view, the state should promote principles of free and equal citizenship even in the face of religious doctrines that oppose them. Because of the substance-based limit on such transformation, some groups whose internal policies make them appear hostile to these principles might not rightly be subject to transformation as long as they do not seek to impose their religious beliefs on others with the force of law. However, if we also adhere to what I have called the means-based limit, even those groups that do seek to impose their religious beliefs on others by law—those who are actively hostile to the ideal of free and equal citizenship—retain certain rights to resist transforma- tion. The persuasion and the financial inducements I have defended must stop short of coercion. Core liberal commitments to the freedoms of expres- sion, association, and conscience require that the state not use force to change religious viewpoints. Of course, the state should not view religious belief as justification for violent crimes, and it rightly punishes even those who give religious reasons for such actions. But policies involving coercive sanction cannot serve as the means of transforming illiberal beliefs. To ensure this, an account of religious transformation must endorse a robust conception of free- dom of association. The right to freedom of association entails what might be called a series of rights to resist or “opt out” of the process of liberal dialectic, which I defended in the previous section. For instance, it is essential that even associations that advocate views of coercion at odds with the Lukumi principle not be forced to publicize their membership lists or the content of their meetings. Subject to the reasonable limitations that accompany any rights, the rights to exist and to congregate should be granted to even deeply illiberal groups. More- over, absent public subsidy, private associations have a fundamental right to decide on their own membership and to keep their membership lists private.35 This right is particularly important when membership is connected to a group’s expressive purpose.36 For instance, I take it to be a right of religious at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 206 Political Theory 38(2) groups to exclude nonbelievers. Often, these rights protect matters of internal membership and organization, matters that do not necessarily violate the political ideals of free and equal citizenship. However, even for groups with views clearly opposed to these ideals, the rights of association must be pro- tected. I therefore stress in this section why, alongside the state’s role in promoting public values of equal citizenship through its expressive capacity and its role as an effective subsidizer of nonprofits, there exists a right for groups or individuals to resist transformation. This exit option is the comple- ment to my claim that the state should limit its pursuit of transformation to reasoning and financial inducement. To clarify this idea, consider a hypothetical university that, when faced with a choice between tax-exempt status and racial discrimination, chose the latter. This university would retain its right to promote values antithetical to liberal legitimacy. However, permitting the continued existence of such an institution does not imply that it should be “left alone”; I believe that institu- tions that retain racially exclusive policies, like our hypothetical university, should be widely condemned by citizens and representatives of the state. Yet their rights to freedom of association and expression require that they not be forcibly shut down. For this position to work, of course, it is important not to regard the concern to protect such a university’s rights as an endorsement of its values.37 An actual case involving a religious institution’s decision to opt out of the dialectical process concerns a recent controversy in Boston over gay adop- tion and the Catholic Church. In March 2006, Catholic Charities of Boston, which had long been a nonprofit entity licensed to facilitate adoptions, shut down its adoption services rather than comply with a state law requiring adoption agencies not to discriminate against gay families.38 While it might seem that this is an unacceptable instance of the state coer- cively preventing a religious organization from participating in charity work, a closer examination of the situation reveals otherwise. As I suggested earlier in my discussion of Bob Jones, Catholic Charities certainly must have felt more pressure than it would have if the state of Massachusetts had merely objected to its refusal to facilitate gay adoptions. But the state’s action is not objectionable in the way that it would have been if there had been an outright prohibition on Catholic Charities’ ability to operate under its favored polices. The issue here concerns Catholic Charities’ financial relationship to the state. Like Bob Jones, the group received nonprofit status. Moreover, the group directly received state funds for each adoption it carried out. The question of whether to force the organization to respect gay couples’ right to adopt, then, should be evaluated with regard to the state’s role as a potential subsidizer because the organization directly receives state funds. For this reason, the at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 207 state may permissibly seek to transform religious belief when it violates the Lukumi principle. In no way was Catholic Charities compelled to participate in a state program funding adoption. Indeed, if Massachusetts were to allow private adoption, Catholic Charities would be able to resume its adoption program, but not as a nonprofit licensee eligible for federal funds.39 We can illustrate this important point simply by imagining a hypothetical state discontinuing a subsidy to a faith-based adoption agency because of its noncom- pliance with an antidiscrimination law against gays but still allowing the agency to operate. In such a case, the state, as spender, decides not to subsidize a reli- gious policy that is fundamentally at odds with its commitment to equal status for gay citizens.40 At the same time, the adoption agency and its associated church are forced to give up their religious beliefs about the immorality of homosexual- ity or the importance of charity work. The state is thus able to express its commitment to equal status and its disapproval of the church’s actions without resorting to the kind of coercion that would come with an outright ban on adop- tion services that did not facilitate gay adoption. In sum, if religious groups that hold such illiberal discrete views wish to opt out of the dialectical process, they should be allowed to do so as a matter of law. These rights to opt out, however, do not imply an acceptance by the state of these views nor an acknowledgment that they are consistent with freestanding public principles fundamental to legitimacy. V. Conclusion In this essay, I suggested why static conceptions of religion neglect the fact that a commitment to religious freedom itself requires transformation of some religious beliefs. Religions, on my view, are not rightfully protected by a metaphorical “wall” or private space into which public values cannot interfere. On the con- trary, the “Lukumi principle” I have proposed suggests that when conflicts emerge between existing religious views and the freestanding public values cen- tral to legitimacy, the state should work to transform religious belief. I argued, moreover, that such transformation must respect both a “means-based” and a “substance-based” limit. The state should seek transformation through persua- sion, not coercion, and should attempt to change only those beliefs at odds with the shared values of free and equal citizenship. In recognizing these limits, the legitimate state can promote its own values while respecting the rights of all citi- zens. My approach therefore grounds a transformative theory of religious freedom in the promulgation and promotion of the reasons that underlie rights, including rights of religious freedom.41 at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 208 Political Theory 38(2) Acknowledgments I am grateful for opportunities to present this article at the Harvard University Center for Ethics, the Harvard Law School, Columbia University, and the University of North Carolina, Chapel Hill. I would especially like to acknowledge the helpfulness of comments from Dennis Thompson, Phillip Pettit, Jed Purdy, Jeff Spinner-Halev, John Goldberg, Richard Fallon, Melissa Schwartzberg, and Mitchell Cohen. I thank Amy Gutmann, George Kateb, Stephen Macedo, Nancy Rosenblum, David Estlund, John Tomasi, Sharon Krause, James Morone, and Michael McConnell for discussions about the themes and general argument of this article. Charles Larmore and Lucas Swaine kindly provided very helpful comments on a late draft. For excellent research assistance I thank Michael Bechek and Minh Ly. Declaration of Conflicting Interests The author declared no potential conflicts of interest with respect to the authorship and/or publication of this article. Financial Disclosure/Funding The author received no financial support for the research and/or authorship of this article. Notes 1. See Michael McConnell, “Free Exercise Revisionism and the Smith Decision,” University of Chicago Law Review 57 (Fall 1990): 1109. Nussbaum also defends a version of McConnell’s jurisprudence in her Liberty of Conscience: In Defense of America’s Tradition of Religious Equality (New York: Basic Books, 2008). The accommodationist approach was employed by the Supreme Court prior to its decision in Employment Division, Department of Human Resources of Oregon v. Smith, 494 U.S. 872 (1990). 2. Martha Nussbaum makes the link most explicit in the final chapter of Liberty of Conscience. Also see John Tomasi, Liberalism Beyond Justice (Princeton, NJ: Princeton University Press, 2001); and Lucas Swaine, The Liberal Conscience: Politics and Principle in a World of Religious Pluralism (New York: Columbia University Press, 2006). 3. See Nancy Rosenblum, Membership and Morals: The Personal Uses of Plural- ism in America (Princeton, NJ: Princeton University Press, 1998); Nancy Rosen- blum, “Okin’s Liberal Feminism as a Radical Political Theory,” in Toward a Hu- manist Justice: The Political Philosophy of Susan Moller Okin, ed. Debra Satz and Rob Reich (New York: Oxford University Press, 2008), 15-40. Rosenblum argues against a “logic of congruence” that suggests we should seek to promote liberal values “all the way down” in citizens’ personal morality. at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 209 4. Nussbaum has provided a book-length defense of the principle of accommoda- tion, which suggests that policies that adversely affect religious belief are limits on religious freedom, but she defends this principle by reference to a deeper value of equality. In contrast, I am concerned to highlight tensions between a commit- ment to equal citizenship and the principle of accommodation. Our projects are united, however, by an attempt to theorize about religious freedom through the lens of political liberalism. See Nussbaum, Liberty of Conscience. 5. Employment Division. 6. Church of the Lukumi Babalu Aye v. City of Hialeah, 508 U.S. 520 (1993). 7. Ibid., at 547. 8. John Rawls terms the Supreme Court an “exemplar” of public reason. It is clear he means to do so in my first sense, but it is unclear whether he would agree with my extension of this term to the second sense, that the Court should be an ex- ample for the wider citizenry. See Rawls, Political Liberalism, 2nd ed. (New York: Columbia University Press, 1996), 231. 9. I develop this argument at length in my Democratic Rights: The Substance of Self-Government (Princeton, NJ: Princeton University Press, 2007). 10. It is an open question as to the amount of transformation the Lukumi principle requires. In societies in which the views of the councilmen are widely invoked in public deliberation, there could be a fairly robust role for the state in seeking transformation. In societies in which the Lukumi principle is widely internalized, such a role might be minimal. I have argued elsewhere about why the ideal of public reason should not necessarily be thought of as a minimal principle. See Brettschneider, “The Politics of the Personal: A Liberal Approach,” American Political Science Review 101, no. 1 (February 2007): 19-31. 11. Both of these limitations help to clarify why the law targeting the Santeria vio- lated the Lukumi principle, while a concern to transform illiberal beliefs such as the councilmen’s is justified by it. The law banning animal sacrifice sought to change a Santeria belief not at odds with the ideal of free and equal citizenship. Moreover, the councilmen used outright coercion, not persuasion, to challenge this practice. In contrast, the Court’s message to the councilman did address a belief at odds with free and equal citizenship, as was evidenced by the statements of the proponents of the law quoted above. Moreover, this message was delivered through expressive, not coercive, means. 12. Rawls, Political Liberalism, 192-93. 13. Lukumi, at 541, 542. I assume that the we here does not refer to official state actors. 14. It is possible, moreover, that this belief in saving one’s fellow citizens through law was central to the beliefs of these councilmen. In my analysis, therefore, I do not appeal to a claim that these particular religious beliefs are tangential to those religions subject to transformation. at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 210 Political Theory 38(2) 15. Arguably, Rawls’s use of the word hope, despite his attempt to hedge here, implies that religious transformation is normatively desirable. See Rawls, Political Liberalism, 192. 16. Susan Okin, Is Multiculturalism Bad for Women? (Princeton, NJ: Princeton Uni- versity Press, 1999), 22. Okin certainly did not mean to say that such extinction should be the intention of any liberal regime—merely that it might be the byprod- uct of promoting certain values in the face of cultures that oppose them. 17. I do not mean to imply, in using the term dialectical, that religious reasons will have an equally weighted presence in the resulting synthesis. It is open to ques- tion whether this dialectic will play out more by having equality clarified by religious views, as in the case of King, or whether it will result in fundamental transformation of religious views. 18. On Rawls’s account, “Political liberalism, then, aims for a political conception of justice as a freestanding view. It offers no specific metaphysical or epistemologi- cal doctrine beyond what is implied by the political conception itself.” Rawls, Political Liberalism, 12. 19. Amy Gutmann, Identity in Democracy (Princeton, NJ: Princeton University Press, 2004), 160. 20. Christopher Eisgruber and Lawrence Sager note that another downside of insisting on literal and complete separation of church and state is that such a reading forbids the use of persuasive religious reasons in political arguments for policies. A literal interpretation of the “wall of separation,” then, might have precluded King’s use of Christian arguments for civil rights. See Eisgruber and Sager, Religious Freedom and the Constitution (Cambridge, MA: Harvard University Press, 2007), 49-50. 21. See Brettschneider, “Politics of the Personal.” For a good general discussion of how liberal citizens might seek to change the minds of illiberal citizens, see Swaine, Liberal Conscience, 134. 22. Rob Reich, “Book Review: Amy Gutmann, Identity in Democracy,” Theory and Research in Education 5 (2007): 247. 23. Mozert v. Hawkins County Board of Education, 827 F.2d 1058 (6th Cir. 1987). 24. Stephen Macedo, Diversity and Distrust: Civic Education in a Multicultural De- mocracy (Cambridge, MA: Harvard University Press, 2003), 157-63. 25. See Anne Fadiman, The Spirit Catches You and You Fall Down: A Hmong Child, Her American Doctors, and the Collision of Two Cultures (New York: Farrar, Straus and Giroux, 1997). The case is also discussed in Gutmann, Identity in Democracy, and in Reich’s review of Gutmann’s book. 26. See Patrick Devlin, The Enforcement of Morals (Oxford, UK: Oxford University Press, 1965). 27. Bob Jones University v. United States, 461 U.S. 574 (1983). at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 211 28. Provision 2 of the policy read, “Students who are members of or affiliated with any group or organization which holds as one of its goals or advocates interracial marriage will be expelled.” 461 U.S. 582. 29. Loving v. Virginia, 388 U.S. 1 (1967). 30. Bob Jones University, “Statement about Race at BJU,” http://www.bju.edu/ welcome/who-we-are/race-statement.php (accessed June 17, 2009). Despite this change, Bob Jones has not reclaimed its tax-exempt status. 31. Archon Fung, “Loosening the Constraints on Public Reason” (unpublished). Fung’s view differs from my own in that he seeks to make public reason less demanding on religion. Fung bases much of his analysis of the mother’s own views on quotations reported in Stephen Bates, Battleground: One Mother’s Crusade, the Religious Right, and the Struggle for Our Schools (New York: Henry Holt, 1993). 32. Another example of the role religious belief can play in educating the wider cul- ture about the requirements of public values can be drawn from an amicus curiae brief filed by the Mennonites in the Bob Jones case. Ultimately, the brief suggests that the Court should support the university’s free exercise claim. Yet the brief is also clear in its condemnation of the practice of racial discrimination. Indeed, the brief’s point is that, overall, there has been more promotion of the values of justice from groups that have been “let alone” by the state than from those that have faced state interference. The Mennonites cite, for instance, the passive resistance of their own fol- lowers to the Vietnam War. While it is an empirically contestable claim whether justice has been promoted more by the state or by religious groups that have separated themselves from the state, the Mennonites’ claim has certainly been true in some instances. The civil rights movement, for example, was largely incited by the black church, which was excluded for most of its history from equal status and recognition by American society at large. The flaw in the brief’s argument, however, is that it fails to distinguish be- tween instances when private religious groups promote public values and instances when they undermine them. The Mennonites clearly distinguish, on substantive grounds, the examples they cite from the case of Bob Jones’s discriminatory poli- cies. Yet they fail to see that this distinction should be the basis of policy. In this sense, their stance is a static one that takes an excessively monolithic view of religion. They advise against state meddling in religion because religions fre- quently promote social justice, but they do not recognize that the current state of religion as a whole, however beneficial individual religions may be to society, is not necessarily optimal. State efforts to change specific illiberal religious beliefs need not deprive religion of its power to work for justice and change. 33. One way of defending this insular understanding of religion as protected against incursions from discrete principles, moreover, might be to appeal to the long-term at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ 212 Political Theory 38(2) impact of such transformation. It could be contended that, without a commitment to insularity, religions would progressively weaken to the point of extinction. I have little confidence in anyone’s ability to make a long-term prediction with any degree of certainty whether this will in fact occur, but if the long-term outcome of making religions compatible with freestanding public principles is the decline of religion, we must view this as the inevitable result of living in a free society. 34. Consider, for example, the significant liberalization of the Catholic Church that was brought about by the Vatican II conference, partially in response to interac- tion with wider liberal society. Although some conservative factions within Catholicism have lamented Vatican II’s changes, these changes have largely met with approval by both laypeople and clergy within the Catholic faith. 35. NAACP v. Alabama, 357 U.S. 449 (1958). 36. See, e.g., Roberts v. United States Jaycees, 468 U.S. 609 (1984), in which the Court found that the Jaycees did not have such an expressive purpose in exclud- ing women, as opposed to Boy Scouts of America v. Dale, 530 U.S. 640 (2000), in which the Court found the Boy Scouts had an expressive purpose in excluding gays. The first question according to my account is whether excluding gays went beyond an internal policy and indeed rejected free and equal citizenship. But even if the latter were the case, my means-based limit would suggest that the organiza- tion might still have a right to exclude gays. It would be a different matter, on my analysis, whether those groups that received public subsidies could have them revoked. Still another question would be raised if such groups claimed a right to meet in public facilities. 37. One of the only universities that relinquished federal funding rather than comply with such legislation is Hillsdale College, a small, nondenominational liberal arts school in rural Michigan. Hillsdale’s administration justifies its decision on the grounds that federal guidelines that require the college to count students by race would violate the college’s commitment to nondiscrimination. 38. Originally, the board members of Catholic Charities had agreed that, as a state licensee, Catholic Charities should comply with state policy on the right of gays to adopt. The policy of the lay-dominated board came into conflict with Vatican doctrine, however, which opposes gay adoption. Following the failure of Boston bishops to obtain an exception to the law, the organization stopped facilitating adoptions. For an important discussion of this and related issues, see Martha L. Minow, “Should Religious Groups Ever Be Exempt from Civil Rights Laws?” Boston College Law Review 48 (September 2007): 781. I read Minow’s general argument to suggest caution about the state’s means of transformation. Even if my account of transformation is correct as a matter of political legitimacy and at the level of principle, I take Minow’s account to offer a warning about how imprudent means of transformation might at times backfire. at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/ Brettschneider 213 39. The fact that Massachusetts, unlike the large majority of states, requires the use of a licensed adoption agency for a child to be legally adopted complicates the issue. If Massachusetts allowed unlicensed adoptions, it could deny state funds to Catholic Charities; having denied funding, the state would have to allow Catholic Charities to continue operating. The fact that Massachusetts state law combines the licensing and subsidy-provision process makes the question more difficult. Should the state be allowed to forbid Catholic Charities from operating because that operation would automatically entail public funding, or should the state be forced to continue financing an organization that refuses to comply with antidis- crimination laws? In the particular case of Massachusetts, the state’s action may well qualify as action in its role as coercer rather than as spender and as a result may be illegitimate. Of course, all this analysis rests on the claim that gay citizens have a right to adopt, which I do not attempt to establish here. 40. Even in the Massachusetts example, the Church remains free to find potential foster families for children. Its lack of a license to facilitate adoption merely forces it to refer such families to other adoption agencies to perform the legal adoption process. 41. There is thus a contrast between my view and those that seek merely to show that a commitment to transformation—for instance, through education—does not entail a limit on rights. My account does not just posit compatibility but argues for a neces- sary moral connection between the protection of religious freedom and transforma- tion. For a narrow account of compatibility that admits to being open to the charge of ethical liberalism, see Macedo, “Liberal Civic Education and Religious Fundamen- talism: The Case of God v. John Rawls,” Ethics 105 (April 1995): 468-96. Bio Corey Brettschneider is associate professor (with tenure) of political science and public policy at Brown University, where he teaches courses in political theory and public law. He is the author of Democratic Rights: The Substance of Self-Government (Princeton University Press, 2007). The first article of his current book project, which offers a rethinking of the public/private distinction, is “The Politics of the Personal: A Liberal Approach,” in the American Political Science Review (February 2007). Brett- schneider’s case book on constitutional law will be published by Aspen Press. Brettschneider is the recipient of the Cornell University Young Scholar Award, a fel- lowship from the American Council of Learned Societies, and a Faculty Fellowship in Ethics from Harvard University. He has also been Visiting Associate Professor at Harvard Law School. He received a PhD in politics from Princeton University and a JD from Stanford University. at BROWN UNIVERSITY on September 20, 2010ptx.sagepub.comDownloaded from http://ptx.sagepub.com/