Are Religion and Autonomy Inconsistent?

An argument that religious purposes are likely to generate autonomy-promoting goods, and that this is part of the justification of the state's promotion of religious purposes via charity law, is plausible only if we accept that the human experience of religion can be autonomous. If religion and autonomy are, in fact, inconsistent—if religious lives (qua religious) are never, or even not generally, autonomous lives—then, from a liberal perspective informed by the ideal of autonomy, the better argument might be that religious purposes are not likely to contribute to the conditions of autonomy and that the promotion of such purposes via charity law is unjustified to that extent. For, to the extent that religion and autonomy are inconsistent, it is not obvious how such goods—whether private or public—as are likely to be realized by religious purposes might contribute to the conditions of autonomy, whether by helping to constitute options that may be the subject of autonomous choice or in other ways. Are religion and autonomy inconsistent? Arguments asserting that religion and autonomy are inconsistent may be found in some of the voluminous literature dealing with questions of religion and the state. The writers who make these arguments tend to emphasize the ways in which religion is experienced as unchosen, a matter of duty rather than preference, largely informed by traditions, commitments, obedience to authority, and revealed truth: in others words, it seems, the very opposite of the path of self-direction and self-disclosure that is an autonomous life.Footnote ³⁴ In order to get a better sense of the proper liberal approach to the promotion of religious purposes in charity law, we need to form a view on arguments that deny that religion is experienced autonomously.

There are several reasons to be skeptical of arguments asserting a general inconsistency of religion and autonomy. First, different religions and different denominations within religions accommodate autonomy in matters of belief and practice to different degrees. At one end of the scale are religious worldviews that emphasize tradition, submission to authority and the obedient discharge of duty, and in which very little (if any) value is placed on autonomy; at the other end of the scale are religious worldviews in which individual inquiry and engagement in matters of spirituality, as well as innovation and self-direction in religious practice, are tolerated and even celebrated.Footnote ³⁵ Given the diversity of religious beliefs and practices, I do not think it is possible with confidence to locate any one religion, or even any one denomination, at a particular point on this scale. For example, all fundamentalist Christian churches emphasize submission to the revealed word of God in scripture, but many encourage experimentation and self-expression in the forms of worship; equally, the old association of the Roman Catholic Church with unquestioning obedience to institutional authority is no longer, if indeed it ever was, sound.Footnote ³⁶ In some respects, Islam presents a set of duties—consider the demands of the Five Pillars of Islam—and yet Abdullahi Ahmed An-Na‘im opens his important book on Islam and the secular state by asserting that “the only way one can be a Muslim” is “by conviction and free choice.”Footnote ³⁷ But even if it is difficult to place various religions and denominations precisely on the scale, the very fact that it is plausible to think of religion in terms of such a scale suggests strongly that to assert a general inconsistency between religion and autonomy is to underestimate the extent to which religion accommodates autonomy. Moreover, even in cases where the demands of religion in matters of belief and practice are not experienced as invitations to exercise autonomy, those demands may be such that autonomous choice is required in realizing them; for example, a Christian may consider it her religious duty to give to the church, but also consider herself free to settle on the amount of the gift as a matter of personal choice.Footnote ³⁸

A second reason to be skeptical of arguments that assert a general inconsistency of religion and autonomy emerges from closer reflection on the nature of autonomy. At times, scholars who doubt that religion and autonomy can be consistent write as though autonomy entailed nothing more than individuals making unfettered choices in order to satisfy their personal preferences.Footnote ³⁹ If autonomy is thus understood, then religion is likely to appear inconsistent with it to the extent that religion is experienced in the forms of the claims of tradition, commitment, authority, and duty; on the view of autonomy as unfettered preference-satisfaction, religious claims figure as autonomy-impeding fetters. However, in Razian liberalism, autonomy need not, indeed should not, be understood as unfettered preference-satisfaction. For Raz, autonomy is an “enabling” value, the point of which is to “enable people to engage with other values.”Footnote ⁴⁰ In Raz's thinking, then, autonomy entails self-directed engagement with values whose goodness is established independently of the fact that they are autonomously engaged. In the case of autonomy, its value is derived from the value of options chosen. Given that autonomy entails self-directed engagement with independently established value, unfettered choices made in order to satisfy personal preferences and unresponsive to independently established value are not central cases of autonomy and, in some respects, are the opposite of what autonomy entails.Footnote ⁴¹ In asking whether religion and autonomy can be consistent, then, our focus should be not on whether the claims of religion fetter free choice, but instead on whether the choices entailed in responding to the claims of religion by accepting those claims are choices that engage with relevant values. I will consider some relevant values below in my discussion of the private and public goods associated with religion.

Still, a person who doubts the consistency of religion and autonomy might insist that people do not typically respond to the claims of religion—understood as claims of tradition, commitment, authority and duty—by making choices and that, to that extent, autonomy and religion are inconsistent. Undoubtedly, certain central experiences in the religious lives of at least some people are not characterized by choices among options: a Damascene conversion, for instance, can scarcely be described as an exercise in self-determination. And the sorts of commitments and practices typically entailed in religion often lead religious people to renounce autonomy in favor of other values. These are reasons to think that religious lives are not likely to be central cases of autonomous lives. But they are not reasons to think that religious lives are never autonomous. It is important to notice that, even though choice is central to autonomy, voluntary activity not amounting to choice can also contribute to self-determination in just the ways that manifest the value of autonomy.Footnote ⁴² To take two of many possible examples, a person may deliberately detach or distance herself from certain beliefs and commitments and try from that position to understand herself in relation to those beliefs and commitments, perhaps with a view to making a choice in the future; or she may subject to scrutiny what she feels are unchosen claims on her—say, the claims of family—and she may respond to those claims in some voluntary way, whether by identifying with them and “owning” them, or by rejecting them, or by adopting a more nuanced stance toward them. Raz himself writes at times as though voluntary activity not amounting to choice cannot be described as autonomous;Footnote ⁴³ however, there appears to be no reason to think that choice exhausts autonomy to the exclusion of other voluntary activity, rather than constituting autonomy's typical and central case. Thus, even the Damascene convert might—perhaps some time after her conversion—aim to reflect on her beliefs and practices, and might seek to engage with those beliefs and practices for the purposes of that exercise in ways that are self-determining and self-disclosing. If we must express this thought using the language of choice, then let us say that the Damascene convert might choose to appraise, for the first time, beliefs and practices that she previously adhered to in an unquestioning fashion. Even if that appraisal results in a renewed commitment to the beliefs and practices in question, it has still engaged the convert's will.

Against the backdrop of an autonomy-promoting public culture such as exists throughout the Western world, it seems plausible to suggest that the experience of religion for many and perhaps most religious people is a combination of autonomous and non-autonomous responses to the claims of religion and that, among the autonomous responses, a range of voluntary activity including, but not limited to, choice may be found. This is not to say that wherever an autonomy-promoting public culture is to be found, religious lives are invariably characterized by autonomy. Even within such a culture, there may be religious groups that insist on strict adherence to tradition and doctrine, or on submission to authority and the obedient discharge of duty in all aspects of life, and many of the individuals within those groups presumably lack autonomy. And there are bound to be people who have grown up in one or another religious tradition and have been taught unquestioningly to accept the beliefs and practices of the tradition in question, but have ceased to identify with those beliefs and practices and have become alienated from their faith. To the extent that, because of their upbringing, such people lack the inner capacities to detach themselves from their beliefs and practices and subject them to the sort of scrutiny that might trigger a renewed commitment or a clean break, they are not autonomous in any meaningful way. Nonetheless, given the diversity of religious beliefs and practices and the range of voluntary activity that typically characterizes autonomous lives, we have reasons to think that the religious lives of highly traditional groups and the helplessly alienated are not typical of religious lives played out against the backdrop of an autonomy-promoting public culture.

Religious Purposes and Autonomy-Promoting Goods

I hope to have said enough to justify proceeding on the assumption that religion and autonomy can be consistent, and on that assumption I now want to explore some of the ways in which goods associated with religious purposes might be autonomy-promoting. To begin with, I want to set to one side purposes that are pursued by people who are motivated by religious beliefs and commitments but the results of which relate only indirectly, if at all, to matters of religious belief or practice. For example, the ethical and moral teachings of most, if not all, religions demand care and concern for others, and this often leads religious people to establish programs for poor relief, education, health care, and support for the disabled, among other purposes. Although the inspiration for the establishment of such programs might be religious, and although the people responsible for them may regard them as out-workings of faith, the goods that they generate—goods to do with the alleviation of suffering or want, or the acquisition of knowledge or skills, and so on—are not directly related to religious belief or practice in that they would be goods even if they were the outcome of purposes pursued by people who were not religiously motivated. In this sense, they may fairly be described as nonreligious purposes.

In thinking about the extent to which goods associated with religious purposes are autonomy-promoting, then, we ought to be thinking about purposes relating to the maintenance and promotion of religious beliefs and practices.Footnote ⁴⁴ Turning first to purposes relating to the maintenance and promotion of religious beliefs, the obvious point to make is that such purposes have a propensity to generate substantial private goods in the form of the spiritual, emotional, and intellectual resources that religion offers to people who are searching for answers to questions about the origins and meaning of the universe, the relationship of the human and the divine, and ethics and morality. Not least among these resources is the phenomenon of faith; a phenomenon that plays a unique role in the epistemic lives of many people and that, it would seem, cannot be experienced except via religion.Footnote ⁴⁵ The spiritual, emotional, and intellectual resources of religion—centered around faith—help people to make sense of the world and their place in it and are the source and form of fundamental lessons about how to live and how to treat others. And these substantial private goods might be autonomy-promoting in at least two ways: first, they might contribute to the development of inner capacities—such as an ability to form commitments, reflect on the direction and purpose of one's life, and evaluate and respond to one's optionsFootnote ⁴⁶—that are necessary to autonomy; and second, they might help to constitute options—especially where they take the form of religious creeds, worldviews, or doctrines—that may be the subject of autonomous choice or other forms of voluntary engagement. The options of being a Jew, or a Buddhist, or a Druid, or an adherent of any other religious creed, worldview, or doctrine, depend on the private goods that are realized by the maintenance and promotion of religious beliefs, and in this way purposes that maintain and promote such beliefs make an important contribution to the conditions of autonomy.

Purposes relating to the maintenance and promotion of religious practices are also likely to generate substantial private goods; these are goods associated with belonging to religious communities, whether those communities are defined with reference to the maintenance of certain practices over the course of time, or with reference to shared practices in the here and now.Footnote ⁴⁷ Religious practices of one type or another make possible shared forms of worship, including the collective participation in religious rites that brings individual people into the experience of a larger spiritual unity; they also give form and substance to celebrations and rites that signify and interpret important events in life, such as birth, marriage, and death. As is the case with the private goods associated with the maintenance and promotion of religious beliefs, the private goods relating to the maintenance and promotion of religious practices are autonomy-promoting in important ways; they substantially constitute a range of options relating to membership in one or another religious organization or group or adherence to one or another religious tradition; they also make possible religious outlooks like iconoclasm and radicalism. This special contribution that religious practices make to the constitution of options also informs the distinction between “practicing” a religion on the one hand, and affirming the beliefs of a religion but not participating in its communal life on the other.Footnote ⁴⁸

As well as generating private goods, purposes relating to the maintenance and promotion of religious beliefs and practices might, under certain conditions, make a contribution to the realization of non-excludable public goods. It is possible to discern, in academic literature and case law on the topic of religion and the state, at least two types of argument drawing a link between religion and public goods. First, there is an argument that the ethical and moral teachings typically associated with religion contribute to social order and stability in a society populated to some sufficient degree by religious people who adhere to those teachings.Footnote ⁴⁹ For two reasons, this argument is problematic. First, although the connection between religion and adherence to moral teachings can scarcely be doubted, and although in many societies religious citizens do live in circumstances of social order, there are enough instances of sectarian division and conflict to call into question the proposition that there is in every society a strong connection between religion, adherence to moral teachings, and social order.Footnote ⁵⁰ Second, the argument assumes that social order is a public good, an assumption that, when applied to circumstances of religious diversity, must be qualified if it is to be reconciled with liberal commitments. Some degree of social order is undoubtedly a great good in the absence of which autonomous living would not be possible; to this extent, social order is a liberal value. However, social order—whether generated by coincidence of adherence to moral teachings by religious people or in some other way—can be at odds with autonomy and to that extent is not valuable from a liberal perspective. Thus, in a society in which most of the citizens are adherents of one religion and share common beliefs about ethics and morality, there might be a high degree of social order, but to the extent that this social order is achieved by sidelining or suppressing the unwelcome beliefs of a religious (or nonreligious) minority, it is not a public good according to a liberal conception of the good.Footnote ⁵¹

The first type of argument drawing a link between religion and public goods may thus encounter trouble in circumstances of religious diversity. In contrast, a second type of argument seeking to connect religion and public goods identifies religious diversity itself as a public good; according to an argument along these lines, the pursuit of religious purposes by the adherents of a variety of religions helps to ensure that those religions—including their constituent beliefs, traditions, practices, and so on—are at least sustained and ideally flourish over the course of time, and a culture characterized by such religious diversity is a public good, enjoyed necessarily by everyone irrespective of whether or not they are religious.Footnote ⁵² In light of Razian liberalism, the key question is whether the public good of religious diversity contributes to the conditions of autonomy, and if it does so, to what extent; and there appears to be an argument in support of the proposition that religious diversity makes a substantial contribution to those conditions. A culture of religious diversity gives people opportunities that they would otherwise lack to become familiar with religious beliefs and practices different from their own. Such opportunities enable people to compare their own religious or nonreligious beliefs and practices with those of others and thereby to understand them more fully and deeply; such understanding contributes to the development of the inner capacities that number among the conditions of autonomy. Moreover, in circumstances of religious diversity, it is through familiarity with religious difference that people are likely to acquire and cultivate attitudes like empathy, trust, and even respect in thinking about and dealing with different religious—and, for that matter, nonreligious—beliefs and practices.Footnote ⁵³ To that extent, autonomy is enhanced. For one thing, attitudes like empathy, trust, and respect contribute to the development of autonomy-enhancing capacities, largely through disposing their holders to grasp and respond to the value in different options. And such attitudes, where widely held, help to sustain social conditions that make religious diversity, and the range of options for autonomous choice that is entailed in religious diversity, possible. Those social conditions are often referred to as “freedom of religion.”

Conflicts between Religion and Autonomy

We have now seen that religion and autonomy can be consistent in the sense that religious lives (qua religious) can be, and perhaps in an autonomy-promoting public culture often are, characterized by autonomy. Moreover, we have seen that religious purposes have a propensity to generate private and public goods that make substantial contributions to the conditions of autonomy. That said, it is undoubtedly true that conflicts between religion and autonomy can arise, and that the pursuit of religious purposes sometimes impairs the realization of autonomy-promoting goods. Such conflicts might arise in a variety of ways. As charity law cases about closed monastic orders show, devout people might commit themselves once and for all to living in highly non-autonomous ways.Footnote ⁵⁴ A religious group might, in light of their religious beliefs or practices, refuse to allow their children to gain the sort of education that is necessary both to developing the inner capacities and to accessing the options on which autonomy depends.Footnote ⁵⁵ A set of religious teachings might require that heretics be “shunned” by the faithful, denying to those who receive this treatment the opportunity to enjoy continuing relationships with spouses and children and impairing autonomy in other severe ways.Footnote ⁵⁶ Members of a religious sect or cult might be subjected to manipulation and even coercion by a charismatic or overbearing leader. A strident preacher might preach intolerance or hatred of those with different beliefs and practices, undermining the culture of religious diversity that, as we have seen, contributes to the conditions of autonomy.

From the perspective of Razian liberalism, these conflicts between religion and autonomy should be resolved in favor of autonomy, especially where what is under threat is the future autonomy of children whose lives are directed in accordance with the religious beliefs and practices of their parents.Footnote ⁵⁷ In some circumstances, the resolution of such conflicts in favor of autonomy may be consistent with state toleration of religious beliefs and practices, for example, where a small religious group is organized according to beliefs that deny the value of autonomy, but keeps to itself and has no interaction with the wider society. In such circumstances, it might be heavy-handed and wasteful for the state to interfere with the group by exercising coercive power against it.Footnote ⁵⁸ In other cases—extreme instances of “shunning” and the preaching of intolerance or hatred might furnish good examples—the harm principle, interpreted along Razian lines in the way that I discussed in Part 2, may give the state good reason to exercise coercive power against and even suppress religious practices that conflict with autonomy. But whether or not the state ought to refrain from exercising coercive power against the holders of religious beliefs and practices that conflict with autonomy is one thing; whether or not the state ought to promote such religious beliefs and practices is another. The case against promoting religious beliefs and practices that conflict with autonomy is more easily made than the case for coercively interfering with such beliefs and practices and, given that the state promotes charitable purposes via charity law, this has a bearing on how the state ought to respond to conflicts between religion and autonomy in the setting of charity law. From the perspective of Razian liberalism, then, charity law should be generous in its acknowledgement of the extent to which private and public goods associated with religion contribute to the conditions of autonomy, but it should also be willing to withdraw its privileges in cases where religious purposes harm people by diminishing their autonomy.

Secularism

Razian liberalism is committed to what might be called “secularism,” not in the sense that it is hostile to religion,Footnote ⁵⁹ but rather in the sense that it seeks to appeal to nonreligious values and principles—chiefly autonomy—when thinking about the justification of state action. Of course, this sort of commitment to secularism is not peculiar to Razian liberalism; indeed, it may be described as a core commitment of liberals of all stripes, so that a person who rejected the notion of secularism altogether simply would not be a liberal.Footnote ⁶⁰ Nonetheless, in Razian liberalism the commitment to secularism is relatively strong, because Razian liberalism goes beyond the proposition that the state should remain in some sense neutral as between differing religious and nonreligious conceptions of the good; in Razian liberalism, the state should promote a nonreligious conception of the good—a conception of the good centered on the value of autonomy—and this should occur even though it might demand state disapproval or even suppression of religious conceptions of the good.Footnote ⁶¹ This sounds harsh and indeed almost illiberal; however, as we have seen, promoting the conditions of autonomy is actually compatible with promoting many (although not all) religious purposes and may be compatible with toleration of some others. Indeed, and perhaps ironically, the relatively strong secularism of Razian liberalism is consistent with a greater degree of state accommodation of religion than is the relatively weak secularism of anti-perfectionist liberalisms that rule out state recognition and promotion of religion altogether within certain spheres of action and reason.

A liberal ideal of public reason developed with reference to the ideal of autonomy must, if it is to be a liberal ideal, entail the commitment to secularism that characterizes the liberal conception of the good centered on the value of autonomy. Thus, according to an autonomy-inspired ideal of public reason, state officials may never endorse, or take as a reason for action, a proposition that may be embraced as true only in light of some religious belief; in other words, an autonomy-inspired ideal of public reason rules out any state establishment of religion. From this perspective, it is never permissible for state officials to support a particular religious organization on the basis that the religious teachings of that organization are true, and equally, it is impermissible for state officials to deny support to or to disadvantage an organization on the basis that its religious teachings are false. When looked at in light of an autonomy-inspired ideal of public reason, in some respects the history of charity law yields disappointing results. Perhaps most notably, prior to the advent of religious toleration in England, trusts for religious purposes other than those associated with the Church of England were unenforceable in law;Footnote ⁶² moreover, in some cases judges ordered cy-pres application of funds intended for such purposes with the result that those funds ended up in the hands of the established church.Footnote ⁶³ Arguably, it is unreasonable to evaluate such early cases in light of a liberal ideal of public reason given that, in many ways, the cases predate a recognizably modern liberalism. The same may not be said of the nineteenth-century case of Mitford v. Reynolds, in which Lord Lyndhurst expressed the view that an application of funds for “idolatrous purposes” such as building a Hindu temple would not be of benefit to the people of Bengal.Footnote ⁶⁴ This illiberal view almost certainly rested on the injunction of Exodus 20:4: “Thou shalt not make unto thee any graven image.”Footnote ⁶⁵ Nor need we be shy about applying liberal standards of public reason to the 1996 Report on the Law of Charities of the Ontario Law Reform Commission, in which the Commission stated that “[t]he good of religion lies in the worship of God, and a life of relationship and obedience to Him.”Footnote ⁶⁶ This statement may be accepted as true only by those whose religious commitments are both monotheistic and entail belief in the possibility of a personal relationship with God; to that extent, it cannot be reconciled with an autonomy-inspired ideal of public reason.Footnote ⁶⁷

That said, from the secular perspective that characterizes an autonomy-inspired ideal of public reason, there is also much to celebrate in the history of charity law. For that history reveals state officials repeatedly refraining from making value judgments about religion based on religious beliefs. As we have seen, during the seventeenth century the purposes of dissenting Protestant groups were not tolerated in English law, and trusts for such purposes were unenforceable. Nonetheless, once the Act of Toleration of 1689 removed the legal impediment to toleration, English law was quick to enforce, as charitable, trusts for dissenting Protestant purposes, even though such trusts were motivated by beliefs at odds with the teachings of the established church.Footnote ⁶⁸ The reasons for this acceptance of dissenting Protestant purposes as charitable are not spelled out in the scant law reports of the time, but we can fairly assume that they rested more on a perceived association between religion and charity than on any particular religious beliefs. A similar story may be told about trusts for the purpose of saying the Roman Catholic mass. In England, such trusts were struck down as “superstitious uses” until as late as the mid-nineteenth century,Footnote ⁶⁹ but once the doctrine of superstitious uses was finally laid to rest in 1919,Footnote ⁷⁰ the enforcement of trusts for masses as charitable soon followed, based at first on deference to the beliefs of Roman Catholic citizens and later on an appreciation of secular goods associated with the performance of religious rites (but only to the extent that the secular goods in question were sufficiently public in orientation).Footnote ⁷¹ And consider also the well-known case of Thornton v. Howe, in which Sir John Romilly had to decide whether a trust for the purpose of disseminating literature written by the controversial religious leader Joanna Southcote was charitable.Footnote ⁷² Although he assessed the religious claims made in that literature as “foolish” and “ignorant,” he refrained from making that assessment the basis of his decision;Footnote ⁷³ instead, he invoked a principle of neutrality, stating that the court ought to make no distinction between religions or between denominations within a religion, and he upheld the trust as a result.Footnote ⁷⁴

The well-known case of Gilmour v. Coats is often cited as the locus classicus of charity law's secularism when dealing with religious purposes.Footnote ⁷⁵ The case was about a trust for the purposes of a Carmelite order of nuns who spent their lives in intercessory prayer and other spiritual exercises within their convent. Three arguments were put to the House of Lords on the question of the likely public benefit of the nuns' purposes. First, it was argued that, according to the teachings of the Roman Catholic Church, the nuns' intercessory prayers were heard and answered by God and caused divine blessings to be bestowed on the world. Second, it was argued that the example of the nuns' pious lives brought spiritual edification to the wider church. And finally, it was argued that the convent provided an opportunity for women of the Roman Catholic faith to practice their religion in the distinctive way that characterizes the Carmelite order. The House of Lords accepted none of these arguments.Footnote ⁷⁶ I will discuss the rejection of the second and third arguments below. For now, I want to point out that, looking at Gilmour v. Coats from the perspective of a liberal ideal of public reason that entails a commitment to secularism, the House of Lords was surely right not to accept the first argument. To accept that argument would have meant accepting a proposition—that intercessory prayers are heard and answered by God—the truth of which may be agreed to only by Roman Catholics and other religious people whose beliefs coincide relevantly with Roman Catholic beliefs.Footnote ⁷⁷ It would have been an illiberal establishment of religion.

Interestingly, in applying the public benefit test in Gilmour v. Coats, the House of Lords had available to it another strategy for dealing with the argument about the efficacy of intercessory prayer; namely, it could have deferred to Roman Catholic beliefs on the subject, as state officials in Ireland and Australia have done.Footnote ⁷⁸ By refusing to defer to Roman Catholic beliefs about the efficacy of intercessory prayer, the House of Lords in Gilmour reinforced the emphasis on objectivity when applying the “benefit” component of the public benefit test that characterizes charity law.Footnote ⁷⁹ This refusal of the House of Lords to defer to Roman Catholic beliefs was, of course, consistent with a secularist ideal of public reason; however, it is worth pointing out that the path of deference, had it been chosen, might also have been consistent with such an ideal. Deferring to religious beliefs is not the same as accepting the beliefs in question or taking them as reasons for action, and it is possible to follow or endorse the practice of deference from a secular perspective, for instance, by viewing it as a means by which the preferences of the citizens whose beliefs they are may be satisfied. That the strategy of deference is consistent with secularism may be seen in the reasoning of Irish judges who have adopted that strategy. Perhaps the clearest statement of that reasoning is in the judgment of Fitzgibbon L.J. of the Irish Court of Appeal in the leading case of O'Hanlon v. Logue:Footnote ⁸⁰

In this passage, the link between deference and secularism is unmistakable.

Autonomy-Promoting Goods

In ascertaining the extent to which the practical reasoning of state officials with respect to “advancement of religion” has been consistent with a Razian understanding of the relation of religious purposes and autonomy-promoting goods, we might as well start by considering the extent to which state officials point to goods associated with religion in the “advancement of religion” setting before considering the extent to which these goods are autonomy-promoting. That state officials do point to goods associated with religion in the “advancement of religion” setting can scarcely be doubted. In general terms, the very fact that “advancement of religion” is identified in charity law as a distinct type of prima facie charitable purpose indicates that state officials readily accept the possibility that religious purposes produce goods of various sorts, even if this is not spelled out explicitly.Footnote ⁸² Moreover, in many “advancement of religion” cases, decision makers do explicitly point to goods associated with religious beliefs and practices in their practical reasoning. For example, endorsement of the proposition that religion is a private good for religious people may be found in the case law, although it is typically accompanied by endorsement of the proposition that it is better to be a religious person than a nonreligious person, a proposition that, for reasons I will explore below, is problematic from the perspective of Razian liberalism.Footnote ⁸³ In Crowther v. Brophy, an Australian case, Gobbo J. of the Supreme Court of Victoria referred to the private goods in the form of the “comfort and peace of mind” associated with intercessory prayer and the “edification” entailed in the celebration of the Roman Catholic mass.Footnote ⁸⁴ And it might be possible to infer from the espousal of a principle of neutrality in Thornton v. Howe, in which a purpose of disseminating religious literature was in view, an acceptance that there are private goods associated with reading, digesting, and discussing such literature.Footnote ⁸⁵

In other cases, decision makers point to public goods associated with religion when reasoning in the charity law setting. This happens in cases where the immediate consequence of some religious purpose is the realization of private goods lacking a sufficiently public character to satisfy the public benefit test of charity law, but those private goods are accompanied by incidental public goods that support a finding of public benefit. Two cases are worthy of comment in this regard. The first is Neville Estates Ltd. v. Madden, a case in which a trust for the purposes of a synagogue closed to the public was under consideration.Footnote ⁸⁶ Cross J. made a finding of public benefit on the basis that, notwithstanding that the synagogue was closed to the public, “some benefit accrues to the public from the attendance at places of worship of persons who . . . mix with their fellow citizens.”Footnote ⁸⁷ The second is Joyce v. Ashfield Municipal Council, a case about whether a hall used for private worship was used for charitable purposes. For the New South Wales Court of Appeal, the worship services, although conducted in private, nonetheless had “public value in improving the standards of the believer in the world.”Footnote ⁸⁸ The views expressed in these two cases suggest the argument associating religion, adherence to moral teachings, and social order that I discussed earlier. By contrast, in the Irish case of Attorney-General v. Becher, Gibson J. of the King's Bench expressed the view that “liberty and equality in religion, as well as in other fields of thought and action, are for the public good,” a view more in keeping with the argument—also discussed earlier—that identifies a culture of religious diversity as a public good.Footnote ⁸⁹

To reiterate: there is evidence that, in the charity law setting, state officials appeal to both private and public goods associated with religion in their practical reasoning. However, the evidence does not all point one way. Gilmour v. Coats suggests that, in some circumstances, decision makers are reluctant to recognize goods associated with religion. As we have just seen, in that case three arguments were put to the House of Lords on the question of the likely public benefit of the nuns' purposes; I dealt with the first argument, appealing to the efficacy of intercessory prayer, earlier. The second argument was that the example of the nuns' pious lives brought spiritual edification to the wider church; in effect, this argument pointed to private goods in the form of encouragement and inspiration for the holders of religious beliefs. The House of Lords rejected the second argument on the basis that the private goods associated with edification were “too vague and intangible” to support a finding of public benefit.Footnote ⁹⁰ The third argument was that, in carrying out their purposes, the nuns in Gilmour v. Coats made available to Roman Catholic women the opportunity to practice their religion in the distinctive way that characterizes the Carmelite order. This opportunity surely entailed private goods for Roman Catholic women. However, the House of Lords made no reference to such goods. Instead, in rejecting the third argument, the House of Lords reasoned that it could not benefit the public to make available to Roman Catholic women the opportunity to become a Carmelite nun, except where this resulted in some wider public benefit, and that there was no wider public benefit in the case at hand.Footnote ⁹¹

In many respects, the practical reasoning of state officials in the “advancement of religion” setting reflects a Razian assessment of the relation of religious purposes and autonomy-promoting goods. Given the propensity of religious purposes to generate a range of private and public goods that contribute in substantial ways to the conditions of autonomy, the willingness of state officials to recognize such purposes as prima facie charitable would, taken on its own, appear to be consistent with an autonomy-inspired ideal of public reason, at least up to a point. Moreover, in cases like Attorney-General v. Becher, decision makers making findings of public benefit point specifically to autonomy-promoting goods like religious diversity. Even the application of a presumption of benefit to certain religious purposes might be consistent with an ideal of public reason understood with reference to the liberal ideal of autonomy; in situations where the links between religious beliefs and practices and the private and public goods associated with the conditions of autonomy have been clearly demonstrated over the course of time, there might be a case for presuming the benefit of purposes relating to the maintenance and promotion of those beliefs and practices.Footnote ⁹² This is most clearly an argument for applying a presumption of benefit to religions that have enriched the lives of large numbers of adherents over long periods of time; implied in the argument is the suggestion that it might be appropriate, in the charity law of a liberal state, to require new religions to demonstrate via the public-benefit test the connections between their beliefs and practices and the goods on which autonomy depends. This suggestion is arguably reflected in the remarks of the Charity Commissioners for England and Wales in their 1999 decision about the Church of Scientology that public benefit may be more difficult to establish in the case of a new religion than an old one.Footnote ⁹³

In other respects, the practical reasoning of state officials in the “advancement of religion” setting may not be consistent with a Razian assessment of the relation of religious purposes and autonomy-promoting goods. In some cases, decision makers point to goods that appear to play no role in promoting the conditions of autonomy—the “comfort and peace of mind” alluded to in Crowther v. Brophy is an example—and in other cases, decision makers point to states of affairs—for example, the “improved standards” referred to in Joyce v. Ashfield Municipal Council—that are not obviously goods at all. However, in some cases in which decision makers make findings as to whether religious purposes are charitable based on reasons that are not consistent with a Razian conception of the good, their decisions may be justified in light of the demands of autonomy, even if their reasoning is not. For example, the decision of the New South Wales Court of Appeal in Joyce v. Ashfield Municipal Council might have been justified on the basis that religious diversity is an autonomy-promoting public good, even though the Court did not have regard to that good in reaching its decision. Similarly, while it is possible that the House of Lords was overly hasty in rejecting the second and third arguments on the question of public benefit in Gilmour v. Coats because those arguments pointed to genuine goods associated with religion, their Lordships' decision not to recognize the purposes of the nuns as charitable might nonetheless have been justified because, even if those purposes stood to generate goods associated with religion, those goods were not autonomy-promoting.Footnote ⁹⁴

Conflicts between Religion and Autonomy

Earlier, we saw that from the perspective of Razian liberalism, conflicts between religion and autonomy should be resolved in favor of autonomy. To what extent does the practical reasoning of state officials in the charity law setting reflect this proposition? The history of charity law shows some willingness on the part of state officials to refuse to recognize as charitable religious purposes that are likely to conflict with autonomy. In Thornton v. Howe, Sir John Romilly M.R. expressed the view that religious purposes “adverse to the foundations of all religion” or “adverse to all morality” ought not to be regarded as charitable in law;Footnote ⁹⁵ this view seems to recognize that, in some circumstances, religious diversity—which, as we have seen, is an autonomy-promoting public good—might conflict with the religious purposes of a particular person or group, and that such a conflict ought to be resolved in favor of religious diversity. And more specifically, section 3(10) of the Irish Charities Act of 2009 provides that a gift is not for the advancement of religion if it is for the benefit of “an organisation or cult . . . that employs oppressive psychological manipulation.” This provision appears to be clearly directed against interference with the freedom from coercion and manipulation that is a fundamental condition of autonomy. On the other hand, in the 1981 case of Holmes v. Attorney-General, Walton J. applied a presumption of benefit to the purposes of the Christian group, the Exclusive Brethren, notwithstanding evidence suggesting that the Exclusive Brethren might have engaged in harmful practices.Footnote ⁹⁶ To the extent that this evidence revealed that the practices of the Exclusive Brethren impaired the autonomy of members of that group, it is arguable that Walton J. decided Holmes v. Attorney-General by resolving a conflict between religion and autonomy in favor of religion. From a Razian perspective, the resolution of conflicts between religion and autonomy in favor of religion is unjustified, and the better response is one that vindicates the liberal ideal of autonomy, whether via a refusal to make a finding of public benefit by the application of a disqualifying rule like the one alluded to in Thornton v. Howe, or by assessing evidence of harms to autonomy and finding that these are outweighed by benefits that stand to promote the conditions of autonomy overall.Footnote ⁹⁷

Singling Out “Advancement of Religion”

“Advancement of religion” is almost always singled out in charity law as a distinct type of prima facie charitable purpose. By singling out “advancement of religion” in this fashion, state officials imply that certain goods are associated with religious purposes in a special or unique way that is not replicated in the case of purposes of other types. When “advancement of religion” is contrasted with “relief of poverty,” “advancement of education,” or indeed many other descriptions of prima facie charitable purpose in modern charity law, it is not usually difficult to grasp the ways in which goods associated with religious purposes are distinct from goods associated with purposes within those other descriptions. However, matters become more complicated when “advancement of religion” is contrasted with purposes associated with the maintenance and promotion of secular belief systems, including atheism, that play a role in the lives of their adherents functionally similar to the role of religion in the lives of the religious. As we have seen, a variety of autonomy-promoting goods, both private and public, are associated with religion. That said, there appears to be no reason to think that, with the possible exception of faith, the various autonomy-promoting private goods associated with religion—the spiritual, emotional, and intellectual resources made available by religious creeds, worldviews, and doctrines, and the goods associated with practicing religion, especially in communities—cannot be realized through the pursuit of purposes for the maintenance and promotion of secular beliefs and practices that are functionally similar to religion. Moreover, there seems no reason to believe that the conditions of autonomy are not maintained, indeed enhanced, in circumstances where religious diversity is accompanied by diversity in analogous beliefs and practices. All of this raises the possibility that, from a Razian perspective, state officials are not justified in singling out “advancement of religion” as a distinct type of prima facie charitable purpose, and that public reason might demand that “advancement of religion” be replaced with a broader general description that captures all of the purposes, religious and secular, likely to generate the autonomy-promoting goods associated with religion that I discussed earlier.Footnote ⁹⁸

The problem of singling out “advancement of religion” has not gone unnoticed in charity law. At times, decision makers seem to have responded to that problem by expressing the view that religious and analogous secular purposes should be treated similarly because they promise to realize similar goods.Footnote ⁹⁹ However, on other occasions, decision makers seem to have denied the proposition that goods associated with religion might also be realized by secular purposes: the best-known of these cases is Neville Estates Ltd. v. Madden, in which Cross J. stated that “[a]s between different religions the law stands neutral, but it assumes that any religion is at least likely to be better than none.”Footnote ¹⁰⁰ And in some cases, decision makers have sought to avoid the problem by defining religion in the setting of “advancement of religion” so as to exclude certain secular purposes, and then finding such secular purposes charitable under the catch-all description of prima facie charitable purposes that invariably completes the set of such descriptions in charity law. Such was the case in In re South Place Ethical Society.Footnote ¹⁰¹ There, Dillon J. said that it was desirable in a liberal society not to draw distinctions between “beliefs deeply and sincerely held, whether they are beliefs in a god or in the excellence of man or in ethical principles, or in Platonism or some other scheme of philosophy”; however, he also said that the purposes of a society formed for the “study and dissemination of ethical principles and the cultivation of a rational religious sentiment” were not within the description “advancement of religion” because they did not relate to “man's relation with God” and could only be charitable under the catch-all description of prima facie charitable purposes incorporated into the English charity law of the time.Footnote ¹⁰²

With an autonomy-inspired ideal of public reason in mind, the latter two of these strategies present certain problems. Denying the proposition that goods associated with religion might also be realized by secular purposes appears to be defensible in view of the unique value of faith, understood as a secular value in light of the contribution that it makes to the epistemic lives of the religious.Footnote ¹⁰³ However, from the perspective of a liberalism of ideals grounded in the value of autonomy, faith is almost certainly an insufficient basis for the seemingly illiberal statement that “any religion is at least likely to be better than none”; moreover, given that almost all of the autonomy-promoting goods associated with religion are also associated with secular beliefs and practices that are analogous to religion, there is reason to think that the unique contribution of faith to religious lives may not be of sufficient weight to justify singling out “advancement of religion” as a distinct type of prima facie charitable purpose. The strategy of defining religion in the “advancement of religion” setting, so as to exclude secular purposes, and then finding secular purposes charitable under the catch-all description, is also not without problems in view of an autonomy-inspired ideal of public reason. Even if a satisfactory definition of religion can be achieved—and the fact that the definition offered in In re South Place Ethical Society arguably excludes Buddhism indicates the difficulties in this regardFootnote ¹⁰⁴—the question of interest is not where to draw a conceptual line between religious and secular purposes. Rather, it is whether, by singling out religious purposes (once defined) and thereby separating them from secular purposes that stand to realize similar autonomy-promoting goods, state officials fall short of an autonomy-inspired ideal of public reason.Footnote ¹⁰⁵

That said, from a Razian perspective, a liberal justification for singling out “advancement of religion” as a description of prima facie charitable purpose might be found in one of two ways. First, such a justification might emerge from reflection on the special way in which religion combines a range of autonomy-promoting goods. Michael McConnell expresses this as well as anyone could:

To the extent that religion uniquely combines goods in the way McConnell describes, and to the extent that this unique combination of goods makes a special contribution to the conditions of autonomy that could not be achieved otherwise, then in light of an autonomy-inspired ideal of public reason, the state may be justified in singling out “advancement of religion” as a distinct type of charitable purpose in law. Whether or not McConnell is right about religion is a question that I will not pursue here; for now, I simply note that the answer to this question may yield a liberal case for singling out “advancement of religion” in charity law.

A second way in which a liberal justification of the state practice of singling out “advancement of religion” might be found is via an argument that an autonomy-inspired ideal of public reason demands that state officials reason about the options that constitute the conditions of autonomy in ways that reflect all or at least a majority of citizens' understandings of those options. Another way of putting this argument is to say that the ideal of public reason demands that the practical reasoning of state officials achieve some degree of faithfulness to the social and cultural meanings that have developed around various options over the course of time.Footnote ¹⁰⁷ According to such an argument, even if it is true that the autonomy-promoting goods associated with religious purposes are associated in the same way with analogous secular purposes, given that religion is widely understood to be a distinct phenomenon, the state has reason when legislating with respect to religion to do so in terms that reflect this widespread understanding, so long as the result of this is not an unacceptable cost to autonomy. The argument might be pressed further to assert that, where the state refers to options in terms that do not reflect widespread understandings of those options, citizens may come to be alienated from their law, with consequences that ought to be of concern from a liberal point of view. Thus, if all else is equal and the state must choose between a prima facie description of charitable purposes that refers specifically to religion on the one hand, and a description that refers to a category wider than but including religion on the other, there may be reason for the state to choose the category that refers specifically to religion, and to deal with the nonreligious purposes that would have been caught by the wider category in some other way.Footnote ¹⁰⁸ On this view, In re South Place Ethical Society might have lived up to a liberal ideal of public reason after all.