• 1 Introduction

  In Burwell v. Hobby Lobby (2014), the US Supreme Court had to determine whether a business association run by a religious family had the right to exclude contraceptive coverage from its workers as part of their employee-based health plan, contrary to the 2010 Affordable Care Act which greatly extended health insurance to US citizens.¹xBurwell v. Hobby Lobby, 573 U.S. (2014). The Court determined that this coverage did indeed burden Hobby Lobby’s free exercise of religion. Similar, in Hosanna-Tabor Church School v. EEOC (2012), the Court’s Justices unanimously found that a school was exempt from federal anti-discrimination legislation after it had dismissed a teacher for ‘insubordination and disruptive behaviour’, which, according to the school board, had damaged her ‘working relationship’ with the school.²xHosanna-Tabor Evangelical Lutheran Church and School v. Equal Employment Opportunity Commission, 565 U.S. (2012). A school, of course, is not a church, even though it may be a church school. By contrast, in Christian Legal Society v. Martinez (2010) the Court took a contrary view, ruling against a religious group.³xChristian Legal Society v. Martinez, 561 U.S. 661 (2010). It held that a public law school could permissibly refuse to register a Christian students’ association whose ‘Statement of Faith’ expressly forbade homosexual relations.

  These cases all involve accommodation for groups, not individuals, and the first two are the basis of Jean Cohen’s powerfully argued paper.⁴xJean L. Cohen, ‘Freedom of Religion Inc.: Whose Sovereignty?’ Netherlands Journal of Legal Philosophy 44(3) (2015). Whether groups per se may enjoy accommodation is a difficult question, and Cohen robustly criticizes some arguments which have been canvassed in favour of their doing so. My aim in this response is to reconstruct Cohen’s argument, noting some of the assumptions she makes and distinctions which are implicit in her argument, and then to consider what if anything may be said in favour of institutional exemptions for religious groups.⁵xCohen’s paper notwithstanding, group exemptions have received rather less scholarly attention than their individual counterparts. They are defended in Mark Rosen, ‘Religious Institutions, Liberal States and the Political Architecture of Overlapping Spheres’, University of Illinois Law Review 3 (2014): 737-83; while Richard Shragger and Micah Schwartzman, ‘Against Religious Institutionalism’, Virginia Law Review 99(5) (2013): 917-85 and Andrew Shorten, ‘Are There Rights to Institutional Exemptions?’ Journal of Social Philosophy, 46(2) (2015): 242-63 raise some important critical worries.

• 2 Against religious institutionalism

  Besides criticizing what Richard Schragger and Micah Schwartzman call religious institutionalism, Cohen also reflects upon the history of church-state relations and the diverse meanings of pluralist and monistic sovereignty. She traces with learning and authority the political theological arguments at work in these key cases through which Christian associations (or more accurately, associations run by Christians) have represented themselves as corporate communities and elevated their status as separate sovereign centres of power, insulating themselves from decisions taken by secular authorities. She traces this to the historic two-worlds theory of shared jurisdictional authority and shows the connection between that and the early-twentieth-century pluralist conception of the state in Britain and France where there were a multiplicity of associations, guilds and other politically empowered groups. Contrasted with the monistic, centralized conception of state sovereignty, even one perhaps that is partially democratic, the pluralist theory holds a certain attraction. As Cohen shows, the new religious institutionalists like to contrast their project from the total sovereignty which the pluralist view opposes. The latter is (potentially) anti-hierarchic, is localized, and disperses power. But the pluralist project only succeeds given certain assumptions about the internal democracy of myriad semi-sovereign associations and the way in which their powers balance each other, for example guilds or trade unions checking business interests. In the case of religious corporate authority this is not what we find. Religion is special, so the new religious institutionalists claim, as it is a sacred realm operating over and above human-made law, so there is no balance of powers. And religion means in practice Christianity, so there is no pluralism either. (It’s noteworthy that most accommodation cases in the US involve Christians, while in Europe cases have been brought by Christians, Muslims, Jews, Sikhs, and other faiths.)

  At the same time as tapping into the historic two-worlds theory, the new religious institutionalists help themselves to some distinctively modern political vocabulary. They make use of the constitutional and indeed human right to freedom of religious conscience, a right which only makes sense in a diverse society where not all citizens are of the same faith or interpret that faith in the same way. And they make use of the principle of non-discrimination to argue that secularism is a controversial conception of the good which the state should therefore not favour over its religious alternatives. Elsewhere, Cohen has shown how the Supreme Court’s endorsement test, which defended church-state separation and symbolically protected non-adherents from state sponsorship of religious purposes, has given way to a relativized neutrality of non-preferentialism between religious and secular activities.⁶xJean L. Cohen, ‘Political Religion vs Non-establishment: Reflections on 21st-century Political Theology: Part 1’, Philosophy and Social Criticism 39(4-5) (2013): 443-69, at 455.

  As Cohen argues, the dangers of the new religious institutionalism for the liberal constitutional project are twofold. First, it detaches democratic political authority from its ultimate authors, the people, relocating a portion of that authority to religious corporations conceptualized as self-authenticating sources of binding norms. While federal political arrangements have much to commend them, federated political units nonetheless draw their authority from just the same earthly sources as the republic itself, the democratic public. Religious institutionalism thus challenges the public justification of political power through its replacement by ‘meta-social’ sources of law. Secondly, by wrapping up political authority inside the black box of protected associations, religious institutions endanger the rights and freedoms of individual citizens which democratic constitutionalism is pledged to respect.

  This is what occurred in the Hobby Lobby and Hosanna-Tabor cases where employees were robbed of access to basic civic entitlements by religious associations. Those associations’ purposes are aided by the very expansive interpretation of the right to religious conscience employed by the Court. Moreover, state recognition of particular (i.e., Christian) religious identities, ceding them norm-making authority, and denuding citizens’ individual rights, work to depluralize associational life and underprotect religious minorities, just what the free exercise and non-establishment clauses were designed to forestall. The ideal of church-state separation, in Cohen’s view, has given way to an integrationist paradigm where religious and state powers are entangled.⁷xCohen, ‘Political Religion vs Non-establishment’, 456.

  In fact, I think there are two distinct ways that associations can acquire religious ideals. First, there are private associations which have an ostensibly non-religious purpose but which are imbued with a religious ethos. Though the purposes of private schools and business organizations – the two main types of association I have in mind here – are morally uncontroversial, the moral problem arises from the fact that these associations are affiliates of religious groups and are infused with their ethos and character. Hobby Lobby and Hosanna-Tabor belong to this category. Cohen points out how this was the first time the Supreme Court examined the right to religious freedom of for-profit associations. Even if there is substantial overlap between private and for-profit associations, the two categories are not identical. A social club, perhaps infused with a religious ethos, is a private association which need not be run for a profit.

  Secondly, there are some organizations – largely non-profit but also profit-driven – which are recruited by the state to achieve public purposes which could otherwise be undertaken by official state agencies. This is the second category, and it includes public schools, hospitals, and charities, which meet the welfare needs of children, the elderly, the poor, the homeless, and other vulnerable groups. State functions become devolved to such associations on a variety of grounds: for fiscal reasons when public funds are cut (as in David Cameron’s ‘Big Society’ in the UK), as part of a generalized impetus to re-energize civil society or as a way of trying to foster a public religious ethos through the back door in states which are officially separationist. As with private associations, public purpose associations raise issues to do with the tension between their ostensible purpose and their internal ethos, but there are also questions about states endorsing the ethos of public purpose associations, especially when the latter enjoy accommodations such as tax breaks or exemptions from non-discrimination legislation.

• 3 Justifying group exemptions

  Private and public purpose associations with a religious ethos can both seek accommodation from the law, but what they seek is a little different. In his recent examination of institutional exemptions, Andrew Shorten distinguishes between the structure, purpose, and ethos of an institution.⁸xShorten, ‘Are There Rights to Institutional Exemptions?’ The purposes of an institution are the aims its members set out to achieve, worshipping God in the case of a church, for example. Its structures are the internal rules which regulate its operation, and the way these assign individuals to institutional roles. An institution’s ethos are the values and norms which its members endorse and which govern the way it tends to carry out its purposes. The relationship between the purposes, structure, and ethos can be quite complex, but in general, for an institution to be properly functioning, its purposes and structure need to be in alignment with its ethos so that they mutually support one other. States, however, can seek to regulate the internal structure of an institution in such a way as is incompatible with their ethos. In the case of public purpose associations, this can happen when the state seeks to enforce non-discrimination legislation, for example, contrary to how members understand the ethos of their institution. This was what occurred for example when the IRS threatened to remove the tax exempt status of Bob Jones University which refused to admit black students. What Shorten calls structure exemptions enable institutions to maintain congruence between their structure and their ethos when the state seeks to reform the former. Thus, a school might be permitted to give preference to children of some Christian denomination in order to preserve its religious ethos. Structure exemptions principally concern private associations without an official public role. By contrast, purpose exemptions protect the external face of institutions.⁹xShorten also mentions a third, less important category of subsidy exemptions which principally consist of tax exemptions. They come into play when states put pressure on institutions to pursue purposes incompatible with their ethos. Purpose exemptions allow institutions to opt out of those purposes so as to preserve their ethos. They chiefly concern public purpose associations which act as proxies for the state in providing basic services to the public. In contrast with structure exemptions, purpose exemptions entitle institutions to discriminate between non-members. One instance is the reluctance of Catholic adoption agencies in both the UK and the US to place children with same sex couples. Another example is hospitals or clinics run by religious organizations which do not wish to provide abortions or contraceptive services. Religious accommodation for individuals raises the issue of congruence between their inner beliefs and the outward manifestation of their beliefs. Institutional accommodation, as Shorten’s view makes clear, involves the same kind congruence: between the ethos of an institution and its purposes or structure.

  How, then, should we assess institutional accommodation (if we reject the two-worlds theory)? Two sets of assumptions will help us to begin with. First, I shall assume, with Cohen, that there is nothing special and unique about religious belief which renders it peculiarly apposite for exemptions. Second, again with Cohen, we need to delimit the right to religious liberty so that it does not become an all-encompassing right and catch all ground for the accommodation of religious institutions. We do better to disaggregate the right to religious liberty into standard liberal rights to conscience, expression, privacy, association, and so on.¹⁰xFor this view see James Nickel, ‘Who Needs Freedom of Religion?’ University of Colorado Law Review 76 (2005): 941-64. This de-sacralizes the right to religious liberty, placing it on the same moral plane as earthly concerns and enabling religious adherents to engage in a process of public justification with their fellow citizens. On the disaggregation view, religious liberty cannot be mobilized as an objection to any kind of secular practice to which a religious person objects.

  With these two assumptions in mind, the next task is to analyse the nature of the institutions which claim accommodation, in order to tease out the sorts of interests they protect. We are already equipped with Shorten’s distinction between the structure, purposes, and ethos of religious institutions. Cohen provides us with a typology of three different types of corporate legal personhood with different normative features. First, the nexus of contracts theory says that associations are comprised of no more than their individual members who have consented to join in order to achieve some common purpose. As Cohen says, this is a nominalist account, the rights and duties the association bears are simply the rights and duties of its members; there is no mysterious group standing over and above the individual members. An association functions simply as a representative of its members’ interests. The concession theory, by contrast, says that corporations are ‘artificial beings’, groups in themselves above and beyond the aggregate of members. On the concession theory, a corporation enjoys rights and bears duties in and of itself, it is a sui generis collective legal person. But even if they are natural groups, they do not have a natural existence. On this account, corporations are brought into existence by the state in order to achieve certain publicly beneficial purposes. If this is so, then, as Cohen points out, it is hard to see how groups constituted by the concession theory can have rights against the very state that brought them into being. At any rate, I shall not discuss the concession theory further here. The third account of corporate personality Cohen discusses is the real entity theory, and although its implications can be benign, it nonetheless has worrisome implications for institutional exemptions. The real entity theory sees religious institutions as sui generis corporate persons, distinct natural kinds, irreducible to the individuals who nonetheless compose them and capable of holding rights against the state. On the real entity theory, religious institutions merit recognition as legal persons. This is not a metaphysical doctrine about the existence of groups over and above the aggregation of their members, but rather a legal doctrine about the proper subjects of rights. Religious institutions, just by themselves, are rights-holders, on the real entity view; the rights they enjoy are group rights, not the agglomeration of the individual rights of their members, as on the contract theory.

  Cohen thinks that the rights enjoyed under the nexus of contracts view are ultimately individual rights, not group rights. Moreover, the individualist antecedents of the notion of a contract might suggest that it is only individual rights that the model supports. However, there is no logical incompatibility between group rights and the nexus of contracts view, or at least something like it. For example, one can imagine each new resident of an apartment block entering into a contract with the other residents in the block, and the existing tenants holding a group right – one they exercise jointly – not to admit any new prospective resident whom they do not want as a neighbour. If the current tenants have an agreed upon decision procedure, such as majority vote, for exercising their right then it is one they exercise jointly; it is not merely the accumulation of individual rights of veto on who occupies the apartment next door. If this is correct, then the difference between the nexus of contracts view and the real essence view is not a difference over who is a rights-holder; it is the difference between a conception of group rights characterized by the rights-holders having a joint interest (sufficient to ground a right) and a conception characterized instead by their prior existence as a group.¹¹xThis distinction, between what he terms the corporate and the collective conception of group rights, comes from Peter Jones, ‘Group Rights and Group Oppression’, Journal of Political Philosophy 7(4) (1999): 353-77. On the former view, a group of rights-holding individuals come together to form a group and in doing so come to acquire joint interests grounding group rights which allow them to protect those interests. On the latter view, the group has prior interests as a group; it does not consist merely of individuals who acquire interests by virtue of combining in some way. This accords with the real entity view’s claim that groups are irreducible to the individuals who comprise them.

  The leaders of an institution who claim an exemption might argue the following. ‘Look, we concede we’re not internally democratic, but neither are many associations. We have a historic ethos. How that is interpreted is up to our leadership, not ordinary members, but if they do not like it they are free to leave. We want to claim an exemption to defend that ethos. It may not be shared by every member, but by joining the association, they acquiesced to it. In that sense, we speak for them’.

  This is the sort of argument that the Green family who ran Hobby Lobby might have made. Is it plausible? It is plainly not a case of individuals exercising their rights because (as we saw) institutional exemptions seek to defend a particular ethos while this argument concedes that ordinary members need not share that ethos. One can hardly exercise a right to protect an interest one does not share. I just said that the difference between the nexus of contracts view and the real essence view is the difference between a conception of group rights characterized by the rights-holders having a joint interest which they come together to advance and a conception characterized instead by their prior existence as a group with certain interests. The view we’re currently considering sits better with that latter alternative because it is too much of a stretch to say, for example, that employees of Hobby Lobby joined it in order to advance their religious purposes. They acquiesced to its rules, including those informed by its religious ethos, a far weaker idea. Thus, we do better, then, to endorse some version of the real essence view when considering this argument. The idea is that, in contrast to the joint interest view, an institution has a prior interest as a group where that interest is defined by its ethos and purpose and is one to which members acquiesce. With that acquiescence, those who control the associations now wish to claim a group right of either the purpose or structure kind.

  One prima facie plausible view is that for a group right to be possible, the members of an association have to be able to authorise those in charge of it to exercise agency on its behalf.¹²xThis is Shorten’s view. Shorten only sees agency being authorised democratically, a position I discuss below. How might that authorization work? One answer is democratically, and I shall consider that in a moment. For now our question is whether hierarchic associations can authorise their leaders to exercise a group right on their behalf. I think that in many cases acquiescence is sufficient to ground the requisite authorization. Suppose for example a company negotiates a contract with its suppliers. If the negotiators act on behalf of the company then I think we can say that the company as a whole enjoys a group right against the suppliers that they deliver the requisite good. Most employees of the company may know or care little about this negotiation. But their joining the company in the knowledge of its purposes can be taken as acquiescing to those purposes and hence acquiescing to the negotiators exercising a group right on behalf of them all.

  However, that a non-democratic association can enjoy a group right does not imply that it can enjoy a group exemption. For an individual to enjoy an exemption, she must be burdened in some way, and if we want to defend group exemptions, it seems reasonable to look for an analogous notion of a burden. In the case of Hobby Lobby, the Green family would point to the fact that the use of contraception was contrary to their evangelical Christian convictions. They would be appealing to the right to freedom of religious conscience, and perhaps, also the right to manifest one’s religious or moral convictions. But that is the Greens, one family. Unlike the case of the company negotiating a contract with its suppliers, the ordinary employees at Hobby Lobby’s stores did not acquiesce to the Green family’s stance; their interests in healthcare (specifically, contraceptive coverage) were set back, unless of course they shared the Greens’ view. Because ordinary employees of Hobby Lobby did have their interests set back by the Greens’ stance on the Affordable Care Act, it is difficult to see how Hobby Lobby as a whole could enjoy a group exemption; however, much the Greens themselves were burdened. Still, there might be a way of rescuing the real essence view. Suppose (a) that those in control of an association do find that their beliefs are significantly burdened by some rule or piece of legislation. Suppose further, that (b) the ordinary members of the association have consented to their control. It seems plausible to say that the ordinary members have thereby acquiesced to the association’s ethos, even if they do not actively support it. If, finally, (c) the proposed exemption which the leadership seeks would not set back the interests of ordinary members (unlike in the Hobby Lobby case) then it seems to me that an exemption may be permissible. This last condition, in particular, requires some elaboration. We need to know what sorts of interests ordinary members actually have, and how an association’s policies (over which they have little or no control) can advance, set back or remain neutral with respect to their interests. But, as an example, suppose that the leadership of a Christian school felt burdened by a legal requirement that they teach all religions rather than focus on Christianity. If non-Christian teachers joined the school, in full knowledge of its Christian ethos, and if (which admittedly could be contested) their interests are not set back by its focus on teaching Christianity, then arguably this exemption is acceptable.

  To be sure, conditions (a)-(c) concern whether a particular exemption claim is permissible; whether it is all things considered justified is a further question. We must also take into account the interests of third parties – in the above example, the children at the school and their parents – as well as any ‘compelling state interest’, in the words of the Supreme Court. Further, it is not clear that this view could justify structure exemptions, where a group wants to maintain discriminatory or inegalitarian practices since those would not appear to be in the members’ interests.

  That is one potential way – I have not spelled out the argument – to justify group exemptions claims. There remains the alternative strategy, which I mentioned above, of democratic authorization. This view makes most sense if we adopt the nexus of contracts view or something like it. Rather than valorizing a group’s ethos, the democratic view will see that ethos represented in the joint interests of its members. A group exemption remains a group right but it is a right grounded in the joint interests of the members of a group, one which they have come together to form. The cogency of the democratic view, however, will rest upon the critical notion of democratic authorization. The strongest notion of authorization will appeal to a formal democratic procedure. On this interpretation, the structure, purpose, and ethos of an institution will be the settled view of its members, analogous to the democratic will of a polity. A law which required an institution to change its structure or purposes so that they were at variance with the ethos of a group could be said to burden the members of a group insofar as they all agree to their group’s structure, purpose, and ethos and the way the former two reflect the latter. But even if robust in theory, this is a very idealistic view. It presumes that associations are internally democratic which very few associations are. And even if an association is democratic, there remains the question of dissenters from the majority view.

  To get round the minority dissenters problem, we could employ a weaker notion of consultation. If an association consults with its members through a recognized forum then perhaps that is sufficient to ground an exemption. If (once again) it does not set back the interests of those who dissent from it, then an exemption may be justified, depending once again on the nature and strength of third-party interests. I think this view offers a more robust route to exemptions than the acquiescence argument above for the simple reason that members of an association, or at least a majority of them, actively authorize the exemption, they do not just acquiesce to the ethos in whose name the exemption is claimed. But at the same time, the democratic view only makes sense if an association is genuinely democratic internally or has some mechanism for consulting its members, and many associations do not even satisfy the latter, weaker condition. Further, in the case of structure exemptions where groups seek an opt out from non-discrimination and equal opportunity legislation, a group’s structure would have to be democratic enough to authorize a collective view even though it contained discrimination and inequalities which the group wished to preserve on the basis of the desired opt out. I just said that the acquiescence view too has more difficulty justifying structure exemptions for a similar sort of reason. Purpose exemptions, where a group enjoys an opt out from being recruited to pursue public purposes, may therefore be more justifiable than structure exemptions. If so, then exemptions for private associations such as Hobby Lobby may be very hard to justify. Even with purpose exemptions, though, we have to explain how a liberal democratic state can authorize groups to meet public purposes in way which are discriminatory or which offend other liberal democratic norms.

• 4 Conclusion

  These brief reflections show that there are formidable theoretical obstacles in the way of justifying group exemptions. If individual exemptions are controversial, then group exemptions are more controversial still. Like Cohen I believe that the Supreme Court decisions in Hobby Lobby and Hosanna-Tabor were wrong on two counts, in their judgements and in the reasoning used to make those judgements. Cohen has powerfully exposed the intellectual errors and practical dangers of that reasoning and offered a framework for a way forward. I have offered a brief sketch of how that framework might be employed to reason about group exemptions. Exposing errors and offering an alternative secular route to group exemptions are both in my view necessary tasks if we do not wish to see the new religious institutionalists win the argument over group exemptions.

Noten

• 1 Burwell v. Hobby Lobby, 573 U.S. (2014).

• 2 Hosanna-Tabor Evangelical Lutheran Church and School v. Equal Employment Opportunity Commission, 565 U.S. (2012).

• 3 Christian Legal Society v. Martinez, 561 U.S. 661 (2010).

• 4 Jean L. Cohen, ‘Freedom of Religion Inc.: Whose Sovereignty?’ Netherlands Journal of Legal Philosophy 44(3) (2015).

• 5 Cohen’s paper notwithstanding, group exemptions have received rather less scholarly attention than their individual counterparts. They are defended in Mark Rosen, ‘Religious Institutions, Liberal States and the Political Architecture of Overlapping Spheres’, University of Illinois Law Review 3 (2014): 737-83; while Richard Shragger and Micah Schwartzman, ‘Against Religious Institutionalism’, Virginia Law Review 99(5) (2013): 917-85 and Andrew Shorten, ‘Are There Rights to Institutional Exemptions?’ Journal of Social Philosophy, 46(2) (2015): 242-63 raise some important critical worries.

• 6 Jean L. Cohen, ‘Political Religion vs Non-establishment: Reflections on 21st-century Political Theology: Part 1’, Philosophy and Social Criticism 39(4-5) (2013): 443-69, at 455.

• 7 Cohen, ‘Political Religion vs Non-establishment’, 456.

• 8 Shorten, ‘Are There Rights to Institutional Exemptions?’

• 9 Shorten also mentions a third, less important category of subsidy exemptions which principally consist of tax exemptions.

• 10 For this view see James Nickel, ‘Who Needs Freedom of Religion?’ University of Colorado Law Review 76 (2005): 941-64.

• 11 This distinction, between what he terms the corporate and the collective conception of group rights, comes from Peter Jones, ‘Group Rights and Group Oppression’, Journal of Political Philosophy 7(4) (1999): 353-77.

• 12 This is Shorten’s view. Shorten only sees agency being authorised democratically, a position I discuss below.