What’s Fairness Got to Do with it? Fair Opportunity, Practice Dependence, and the Right to Freedom of Religion

Abstract

The right to religious liberty as for instance set out in the European Convention of Human Rights protects acts of religious observance. Such protection can clash with other considerations, including laws aimed at protecting other state interests. Religious freedom therefore requires an account of when the right should lead to exemptions from other laws and when the right can legitimately be limited. Alan Patten has proposed a Fair Opportunity view of the normative logic of religious liberty. But Patten’s view faces several problems. The normative work in his view is mainly done by added accounts of reasonable claims and of justifiability. So, the Fair Opportunity view in itself does not provide a normative criterion. Defenses of the Fair Opportunity view must therefore turn on the theoretical preferability of its structural features. But the Fair Opportunity view has the wrong form to capture the right to freedom of religion. The form of the right to freedom of religion is due to how its point is to address how states limit the liberty of citizens. Given a practice dependent approach, which assigns importance to the point and purpose of the right to freedom of religion, Patten’s theory is thus problematic.

Introduction

Freedom of religion is a standard liberal right in constitutions as well as international human rights conventions. Religious freedom is partly uncontroversial, at least among liberals, since few deny the right to freedom of thought, conscience and religious belief. However, freedom of religion also includes a right to manifest one’s religion or belief. Here religious freedom becomes controversial because manifestation of religious belief might clash with laws and policies, including those that protect other liberal values. Freedom of religion might therefore require exemptions from such laws or policies. Most articulations of the right to religious freedom make clear that the right is not absolute but can be limited subject to appropriate conditions. This is a dilemma where one liberal commitment must be balanced against other liberal commitments. The question therefore is how this balancing should be understood and which criterion should guide how the balance should be struck.

To answer these questions, Alan Patten has proposed an account of what he calls “the normative logic of religious liberty” (Patten 2017a, b), which falls between two other types of views. One is the No Burden view, the other is the No Discrimination view. Patten finds both problematic and suggests that a principle of Fair Opportunity provides a more plausible middle ground between them.

While Patten’s account provides a way for liberals to handle the dilemmas of religious freedom, it also raises several new questions. One is whether it provides a plausible normative criterion for striking the balance in the relevant kinds of cases. Another is whether it provides a reasonable reconstruction of how the right to freedom of religion works. Patten’s account can be questioned with respect to both questions.

The first question is whether the fair opportunity view provides a criterion that actually contributes to decide cases where religious freedom has to be balanced against countervailing considerations, including whether the notion of fairness contributes normatively or whether the normative work in Patten’s account is rather done by other notions such as reasonable claims.

In both much national law and in international human rights law, the right to freedom of religion involves assessment of whether rules or laws burdening religious conduct are proportionate means to legitimate ends.¹ The second question is whether a Fair Opportunity view offers the right kind of reconstruction of this legal right. This question assumes that it matters for assessing a philosophical theory about religious liberty how legal rights to religious freedom in fact work, which is a version of the general methodological discussion in political philosophy about so-called practice dependence. If theories should be practice dependent, and if Patten’s theory does not provide an account of religious liberty that fits how legal rights to religious freedom actually work, then this is a problem for the theory.

This paper is structured as follows. First, I sketch the legal right to freedom of religion using the formulation in the European Convention of Human Rights as illustration. I show how such rights lead to the question of exemptions and how a theory of religious liberty is needed to help us understand and provide normative guidance in such cases. Secondly, I present Patten’s mapping of the different positions in debates over how such a theory of religious liberty should look and how he presents his own Fair Opportunity view. I then return to the noted questions we can ask about his proposed account. I focus on two questions. The first question concerns whether the Fair Opportunity view provides the kind of normative criterion needed to guide decisions in cases of religious liberty. The second question concerns whether the Fair Opportunity view provides an account of the right form to address cases involving the right to freedom of religion. In the concluding section, I recapitulate the challenges facing Patten’s Fair Opportunity view and their link to methodological debates about whether theories of human rights such as the right to freedom of religion should be practice dependent.

Religious Freedom and Exemptions

The right to religious freedom figures in most national catalogues of rights as well as in international human rights conventions. The formulation of the right to freedom of religion in the European Convention of Human Rights (ECHR) will serve as an example. ECHR article 9 says that:

Everyone has the right to freedom of thought, conscience and religion; this right includes freedom to change his religion or belief and freedom, either alone or in community with others and in public or private, to manifest his religion or belief, in worship, teaching, practice and observance.

ECHR article 9, Sect. 2, states that religious liberty to manifest one’s religion or belief can be limited, but only if limitations are prescribed by law and “are necessary in a democratic society in the interests of public safety, for the protection of public order, health or morals, or for the protection of the rights and freedoms of others.”

These limitation clauses mean that special conditions must be fulfilled for freedom of religion to be restricted. This can only happen if the aim of the law limiting religious conduct falls in the class of legitimate aims and if the limitation of religious freedom is a proportionate means to this end.

The formulation of the right to freedom of religion in the ECHR applies to the actions of states since it is states (and public authorities in general) who are directly bound by human rights. So, the upshot of ECHR art. 9 is that states cannot limit freedom of religion in a discretionary way but only through law, and that such laws must be proportionate means to one of the legitimate aims.

If a state action does not live up to these conditions, several things can happen. Either a policy or law is a direct rights violation and must be repealed for this reason. Alternatively, if the law limits freedom of religion but the conditions set out in the limitation clauses in Sect. 2 are fulfilled, then there is no rights violation, and the law can be upheld. But there is also a third possibility; if there are good and weighty reasons for the law that live up to the conditions in the limitation clauses, but the law results in disproportionate limits on freedom of religion for a particular group of people, e.g., a religious minority, then the solution might be to uphold the law or policy but to exempt the group from it. This is what Brian Barry called “the rule and exemption approach” (Barry 2001). The reason for adopting a rule and exemption approach might be that exemptions for a smaller group do not undermine the aim of the law and might be necessary for making the law proportionate. Such religious exemptions range from conscientious exemptions for doctors from performing abortions, over exemptions from animal welfare provisions to allow ritual slaughter, to exemptions from dress codes or safety requirements.

One issue often raised in relation to religious exemptions is whether they assume that religion is special.² For the moment, it suffices to note that most rights catalogues and human rights documents in fact operate with religion as an object of rights protection. However, it is also the case that religion is often not considered uniquely special. ECHR art. 9 for instance protects freedom of “thought, conscience and religion.” Therefore, the protection in principle extends to a broader category of which religion is only a part. This does not answer the normative question whether this is a justifiable state of affairs. But for present purposes, this is not the question I want to discuss.

Given that there are legal rights to freedom of religion, which can in some cases lead to religious exemptions, the question is when religious conduct should be protected and, if necessary, exempted from general laws, and when such general laws can legitimately override the concern with freedom of religion and limit religious conduct. Even if we accept the right to freedom of religion, e.g., as formulated in ECHR, art. 9, the answers to these questions are not given. How should a judge for instance weigh the importance of a specific kind of religious conduct against the aims of a law that limits this conduct? What type of weighing is involved? What are the considerations that are weighed against each other? It is such questions that philosophical theories of religious freedom seek to answer. Ideally, a theory of religious freedom should provide a criterion that could be used to determine how such questions should be answered. A theory should at least provide a way of understanding the type of decision and the considerations that are relevant to it.

In the following, I will examine how Alan Patten’s fairness account of the normative logic of religious liberty performs this role and whether it succeeds in doing so.³

A Fairness View as Opposed to a No Burden or No Discrimination View

Alan Patten presents his fairness account as a middle way between two other types of views. These other types of views both represent prominent positions in the debate (both the political and the philosophical ones) and rely on systematically different theoretical understandings of religious freedom. Patten’s middle way is thus both a political middle way, which seeks to avoid what he from a liberal point of view sees as normatively problematic implications of the two other types of views, and a theoretical third way, which offers a different understanding of religious liberty and the types of considerations it involves. One can thus discuss Patten’s view both from a political/normative perspective and from a theoretical perspective. Both perspectives are helpful in order to lay out the dialectical landscape in debates about religious freedom, as presented by Patten, in which he locates his fairness view.

Patten characterizes the No Burden view⁴ as follows:

Because it is a serious setback to a person’s legitimate interests when the law conflicts with her efforts to follow her religious beliefs, the state should not impose such burdens unless it has a very good reason. In the absence of such a reason, a law conflicting with religious conduct should be withdrawn or amended or an exemption should be carved out to remove the conflict. … the core of this principle is the idea that legal burdens on religious conduct should normally be avoided (Patten 2017a, p. 130).

The No Burden view allows claims to religious liberty to be outweighed by other important considerations, so it only involves a pro tanto claim, but one which is assumed to be weighty, so there is a presumption against any law that burdens religious commitments (Patten 2017a, p. 132).

The problem with the No Burden view is that it “Fails to disentangle judgments about how serious a burden is from judgments about whose responsibility it is to prevent that burden” (Patten 2017a, p. 139). In other words, the No Burden view is not responsibility sensitive and fails to hold people responsible for bearing the consequences of their beliefs.⁵ The extent of a burden constituted by limits on a person’s opportunities to engage in religiously motivated conduct is a different question from whether the person can and should be held responsible for bearing this burden. An account that only focuses on the burden thereby sets aside the issue of responsibility. A further problem with the No Burden view is that it allows exemptions from laws dear to liberals. Patten provides some examples of this from recent US court decisions, including the notorious U.S. Supreme Court’s decision in Burwell v. Hobby Lobby (2014), where the Court upheld religious liberty over a provision in the Affordable Care Act mandating that health insurance policies include contraception in their preventive care coverage (Patten 2017a, p. 129).

Patten then characterizes the No Discrimination view⁶ as follows:

the No Discrimination principle equates fair treatment with the absence of both targeting (where the law singles out a particular religion, or religion in general, for unfavorable treatment) and selective accommodation (where the law extends an accommodation to relevantly similar conduct but withholds it from some particular instance of religiously motivated conduct). (Patten 2017a, pp. 130–31).

According to No Discrimination, there is no presumption against a valid and neutral law of general applicability. If all are treated in the same way, there is no violation of religious liberty (Patten 2017a, p. 143).

The problem with the No Discrimination view is that it thereby rules out any exemptions and that it is ad hoc since it does not explain why targeting and selective accommodation are unfair (Patten 2017a, p. 143).

To avoid the problems of both the No Burden and the No Discrimination views, Patten then suggests a third account, namely the Fair Opportunity view:

According to the Fair Opportunity principle, there is a presumption against any law that leaves individuals without a fair opportunity to pursue and fulfill their religious commitments. (Patten 2017a, p. 144).

The core of the fair opportunity view is what Patten calls “the Fair Opportunity for Self-Determination (FOSD) principle”:

Each person should be given the most extensive opportunity to pursue and fulfil her ends that is justifiable given the reasonable claims of others. (Patten 2017a, p. 145, and Patten 2017b, p. 208).

Does the Fair Opportunity View Deliver What We Need?

Because Patten’s view can be seen both as a theoretical account of how we should understand freedom of religion and as a normative position on how claims for exemptions should be settled, we might have different kinds of discussions of views like this.

One kind of discussion is about whether the view gives plausible normative answers. The question would then be whether the Fair Opportunity view yields plausible ways of balancing competing considerations in cases involving claims for exemptions. As noted, Patten has indicated that the Fair Opportunity view will be less accommodating of claims for exemptions than the No Burden view but more accommodating than the No Discrimination view. More specifically, Patten has suggested that the Fair Opportunity view will not countenance exemptions from general laws dear to liberals, such as laws requiring mandatory employer paid health insurance policies to include contraception in their preventive care coverage and laws prohibiting businesses to discriminate against LGBT customers on religious grounds (Patten 2017a, p. 131), while it does not commit liberals to the general rejection of exemptions earlier championed by Barry.

This presentation of the Fair Opportunity view as a moderate middle way between the two other views can be read as an argument that the Fair Opportunity view is preferable on normative grounds. The argument assumes that liberals will both be committed to the noted kinds of general laws, to which they will not accept exemptions, and will support exemptions for religious minorities in other cases. Given these normative assumptions and given that the Fair Opportunity view will yield moderate prescriptions along the lines suggested by Patten, then it follows that liberals should prefer the Fair Opportunity view over the No Burden and No Discrimination views.

The problem regarding the normative discussion of the Fair Opportunity view is that Patten has not yet developed the view enough to say precisely which cases of exemptions it would countenance and which not. The indicated argument for the normative preferability of the Fair Opportunity view for liberals is still just a suggestion. Patten has not yet specified the Fair Opportunity view sufficiently so that it can be applied to a broad range of the kinds of cases where we would want to compare the implications of the different views. So, at present, the normative preferability of the Fair Opportunity view remains a suggestion rather than something that Patten has shown.

While we cannot yet assess the normative plausibility of the Fair Opportunity view, we can have a different type of discussion about it. A theory can both be assessed in terms of its normative implications and in other terms. Here, I will focus on the form and type of account offered, i.e., whether the Fair Opportunity view provides a theory of a kind and structure that fits the right to freedom of religion. Given my initial sketch of issues related to freedom of religion, an account of freedom of religion is needed to provide a criterion that can inform decisions about the extent of freedom of religion, including whether and when to grant exemptions. The theoretical question is whether the Fair Opportunity view offers a plausible account of how we should understand such decisions and the considerations entering into them.

I will present two issues of this kind: First, does the Fair Opportunity view offer a normative criterion at all? Second, does the Fair Opportunity view provide an account of the right form?

Does the Fair Opportunity View Offer a Normative Criterion at All?

The first issue concerns whether Patten’s Fair Opportunity view offers the kind of normative criterion that is needed to provide normative guidance in cases about freedom of religion in general and exemptions in particular. Such cases are difficult precisely because they involve clashes of different considerations. The Fair Opportunity view in particular needs to provide a way to handle countervailing considerations because it eschews the simpler approaches of the No Burden and No Discrimination views. The FOSD principle explicitly states that “Each person should be given the most extensive opportunity to pursue and fulfil her ends that is justifiable given the reasonable claims of others.” So, in the very formulation of the principle, it requires a weighing of competing considerations.

Does the Fair Opportunity view provide a normative criterion that can help us decide such cases involving competing considerations? At the general level, the view says that such cases should be resolved in a way that is fair. This is a different way of answering the question to at least the one provided by the No Burden view.

What does it mean to say that such cases should be resolved in a fair way? Patten unpacks this in the FOSD principle by saying that fairness requires that each person should be given the most extensive opportunity to pursue and fulfil her ends that is justifiable given the reasonable claims of others. This raises the further question about what counts as reasonable claims. Patten explicitly writes that The Fair Opportunity principle is not tied to a specific account of what these reasonable claims are (Patten 2017a, p. 144, p. 145). Patten clearly has views about which claims are reasonable and which are not. For instance, he does not think that people have a reasonable claim to oppress others, to reject having their income taxed, or not to be offended. But his Fair Opportunity view in itself does not take a stand on this issue about reasonable claims.

At this point, it is evident that the Fair Opportunity view as presented by Patten is not a fully specified normative view. This means that the view as presented by Patten cannot be used to decide cases involving conflicting considerations, since it minimally needs to be supplemented with an account of reasonable claims.

Even given that the Fair Opportunity view needs to be supplemented with an account of reasonable claims, as acknowledged by Patten, this is furthermore not enough to turn it into a normative criterion that can be used to decide cases involving conflicting claims. This is so because the fact that an exercise of one person’s self-determination conflicts with another person’s reasonable claims does not in itself tell us whether the first person’s conduct is justifiable or whether it would to the contrary be justifiable to limit that person’s conduct with reference to the reasonable claims of the other person. Patten also acknowledges this when he writes that “A full discussion of FOSD would need to include an account of the conditions under which a limit on a person’s self-determination is justifiable given the reasonable claims of others” and that “FOSD is compatible with a variety of approaches to resolving conflicts.” (Patten 2017a, p. 146) FOSD only tells us that religious liberty cases involve considerations for some people’s self-determination which can conflict with the reasonable claims of others. It neither tells us which claims of others are in fact reasonable, nor how the conflict should be resolved between exercises of self-determination and claims that are reasonable.

Is this a problem? If the reason for having an account of the normative logic of religious liberty was that religious liberty cases involve competing claims and countervailing considerations, then such an account should help us settle how such cases should be decided. We might expect a view offered for this purpose to provide a normative criterion that we can use to adjudicate in such cases. It is of course true that any general principle proposed will be indeterminate to some extent. This also goes for competing theories, such as Cécile Laborde’s proposal that claims for exemptions should be understood in terms of what she calls integrity-protecting commitments (2017). The problem in Patten’s view is not merely this, but that it answers the question about how conflicting claims should be weighed, not by offering a normative criterion, however indeterminate, but by in a sense rephrasing the question in terms of reasonable claims and justifiability, which require full normative theories of their own. If the view offered does not provide such a criterion, then it does not help us in this regard.

If the Fair Opportunity view does not offer a normative criterion, the question is how much normative work it does. If the view is compatible with different accounts of reasonable claims and of the justifiability of religious conduct relative to the reasonable claims of others, then the Fair Opportunity view does not provide answers to the normative questions on its own. The implications of the view will then depend on which accounts of reasonable claims and of justifiability one plugs into it. If the real work is done by these further accounts, then it seems that these are what we should be discussing, not the Fair Opportunity view.

This might be too harsh, however. Even if the Fair Opportunity view must be supplemented, one might argue that the structure of the view already embodies certain normative commitments and rules out others. It is a liberal individualist view in a general sense, since it specifies that the only relevant considerations are person’s self-determination and claims. Even though the formulation itself does not say so, the structure of the view arguably also embodies a logic of individual responsibility.⁷ So, one might say that the view is formal, but that the form is individualist and at least potentially responsibility catering. Even though the view must be filled out with more specific accounts of reasonable claims and of justifiability, it can only be specified in ways compatible with individualism and responsibility sensitivity.

But even if this is true, there is another worry about the fact that the Fair Opportunity view must be filled out with more specific accounts of reasonable claims and of justifiability to yield specific normative implications.⁸ Such specifications of the Fair Opportunity view risk collapsing into a version of either the No Burden view or the No Discrimination view. Roughly, on restrictive accounts of what counts as reasonable claims and on views according to which limitations on religious conduct are rarely justifiable, the Fair Opportunity view will yield the same normative implications as the No Burden view. In this case, the Fair Opportunity view and the No Burden view end up being extensionally indistinguishable. If, on the other hand, one plugs in an account according to which people only have a claim to be treated the same as others and a view according to which limitations on religious conduct can easily be justified, then one gets the same normative implications as No Discrimination view.

Since Patten originally proposed the Fair Opportunity view as an alternative to the No Burden and No Discrimination views, it is problematic if the former can be specified in a way that is extensionally indistinguishable from the views to which the Fair Opportunity view was supposed to provide an alternative.⁹

Patten might respond that it is not all specifications of the Fair Opportunity view that will be extensionally indistinguishable from either the No Burden og the No Discrimination view. Even if some specifications will yield the same normative implications as one of the two other views, there are many possible specifications, several of which will indeed yield distinct normative implications, especially if we plug in more permissive accounts of what counts as the reasonable claims of others.

This is probably true; it is likely that the Fair Opportunity view can be specified in ways that are normatively distinct from both the No Burden and the No Discrimination views. However, even if we grant this, the possibility that some specifications of the Fair Opportunity view have the same normative implications as either of the two other views nevertheless means that the general argument for the Fair Opportunity view in terms of its normative preferability over the two other views fails since the generic features of the Fair Opportunity view do not ensure the normative difference to which this argument appealed.

This in turn means that the argument for the Fair Opportunity principle must be made either at the level of the normative preferability of particular specifications of the view or in terms of the theoretical preferability of it structural features.¹⁰ If one takes the first route, this confirms the criticism above, namely that what really is doing the normative work is the accounts of reasonable claims and of justifiability invoked to specify the view, and that the discussion therefore should focus on these accounts rather than on the Fair Opportunity view. This seems to block any general normative argument for the Fair Opportunity view and to sideline the Fair Opportunity view in discussions of the normative preferability of particular specifications.

To respond to the criticism and at the same time keep the focus on the Fair Opportunity view as such, Patten needs to claim that the structure of the view is what matters. This will then mainly be a claim about the theoretical preferability of the Fair Opportunity view. Since there is a structural difference between the Fair Opportunity view and the other two views, this is not an implausible way to respond.

But if the distinctiveness of the Fair Opportunity view is what it says about the structure of religious liberty claims, which is necessary but not sufficient for providing the needed normative criterion, then the next question is whether it provides an account of the right form. I will now turn to this second issue.

Is the Fair Opportunity View of the Right Form?

The second issue concerns the fact that the Fair Opportunity view is what we might call a horizontal and comparative account. By this I mean that the Fair Opportunity view is concerned with comparing the opportunities and reasonable claims of different persons who are placed on the same level in relation to each other. Patten makes clear that the Fair Opportunity view is a broad category, where the No Discrimination is one possible specification of a fairness-based view (Patten 2017a, p. 130). The comparative nature of such views is most evident in the case of the No Discrimination view, which is a strict egalitarian view according to which people should simply have the same opportunities in the sense that they should be subjected to the same rules and restrictions. No Discrimination means that no one has different opportunities in this simple sense than others. This shows the sense in which the view is comparative, namely that the view requires comparing the opportunities (defined by applicable rules and restrictions) faced by different people. It also shows the sense in which No Discrimination is what I call a horizontal view, namely that it is the same kind of things that are compared, namely opportunities faced by different people, who are placed at the same level in relation to each other.

The Fair Opportunity view is also comparative and horizontal, but in a more complicated way than the No Discrimination view. The FOSD principle says that each person should be given the most extensive opportunity to pursue and fulfil her ends that is justifiable given the reasonable claims of others (Patten 2017a, p. 145). So, the Fair Opportunity view also requires a comparison, but it is not a simple comparison of opportunities. Rather, in cases where one person’s exercise of self-determination can conflict with the reasonable claims of others, Fair Opportunity requires us to assess the opportunities of the one person (or group of persons) relative to the reasonable claims of others. Rather than a simple comparison of different sets of opportunities, this is a case of looking at opportunities, on the one hand, and claims of others, on the other. This is still a kind of comparison since the assessment concerns which extent of opportunities are justifiable relative to the claims of others. One might say that the Fair Opportunity view involves a moralized comparison where one side is concerned with more or less opportunities and the other side is concerned with reasonable claims. The relevant concept of opportunity is accordingly informed by considerations of fairness rather than being given independently of the moral assessment, as is the case in No Discrimination and No Burden views (Lægaard 2022, pp. 14-15).

Exactly how the comparison of opportunities and claims of others is supposed to work is an important question for Patten’s account. However, for present purposes I will rather focus on how the Fair Opportunity view is horizontal in a sense similar to the No Discrimination view. Just as the No Discrimination view is horizontal in the sense that the different opportunities that enter into the comparison are at the same level, this is also the case with the Fair Opportunity view. The opportunities and the reasonable claims that are compared are at the same level in the sense that they are opportunities and claims of persons. The idea of fairness at stake here is concerned with treating persons equally, not in the simplistic sense of the No Discrimination view, but in a more qualified way where their respective opportunities for self-determination and reasonable claims are given equal concern and respect, to use a Dworkinian phrase. Patten sometimes writes that persons have symmetrical claims (Patten 2017b, p. 209), which is another way of saying the same thing that illustrates the comparative and horizontal nature of the view. The whole point of the Fair Opportunity view is that the opportunities and claims of different persons should be treated equally—as being on the same level—in this sense. This is the sense in which the Fair Opportunity view is horizontal.

Why is it interesting that the Fair Opportunity view is comparative and horizontal in these senses? This is because Patten offers the Fair Opportunity view as an account of the normative logic of religious liberty. One of the central mechanisms for regulating religious liberty is the legal right to freedom of religion. So, one way to assess the Fair Opportunity view as an account of religious liberty is to ask whether it provides an account of the right form to capture and guide us in discussions about the legal right to freedom of religion.

Here the comparative and horizontal nature of the Fair Opportunity view becomes relevant because the right to freedom of religion as articulated in legal documents such as the European Convention of Human Rights has a different structure. Rather than comparing the claims of different persons, which are at the same level, the right to religious freedom requires an assessment of whether the extent of the limitation set by some law on specific persons’ opportunity to engage in religious conduct is proportionate relative to the aims supposed to justify the law in question. This is rather a vertical and non-comparative account. Vertical, since the elements in the proportionality assessment are not of the same type and at the same level. One element is the extent of limits on persons’ opportunities. The other element is the state’s aim that is supposed to justify this limitation of liberty. The relation between the two elements in the proportionality assessment is accordingly one between the aim of the state, which is assumed to have legitimate authority over the persons in question, and these persons’ individual liberty.¹¹

One might object that states do not have moral standing in their own right and that the aims of the state must therefore be understood as in turn justified with reference to the interests of persons.¹² This is indeed the case on broadly liberal views of the state (e.g., Buchanan 2004; Altman and Wellman 2009). This might be taken to mean that, even though a state aim figures in the assessment, this aim in turn should be understood as justified with reference to interests of individual persons, and therefore the assessment involved in the right to religious freedom is really a horizontal comparison after all.

One can grant the broadly liberal view of the state that is the premise of this objection without accepting the conclusion that the right to religious freedom involves a horizontal comparison between individual interests. Even if the state is ultimately justified in liberal terms, i.e., with reference to how it protects the rights and interests of individual persons, this does not mean that the proportionality assessment involved in the operation of the right to religious freedom compares (sets of) individual interests. The former concerns the general philosophical justification of the state, the latter concerns the operational principles that can guide judges in making specific decisions. When a judge assesses whether a limitation on religious conduct is justified with reference to, e.g., public order is proportionate, the assessment does not take the form of a comparison between the aggregate individual interests in public order compared to how the limitation affects the interests of the persons whose liberty is limited. Even if the state is ultimately justified with reference to how it protects the rights and interests of individual persons, this ultimate justification cannot be inserted in the proportionality assessment that a judge must make when considering whether the state’s measures are proportionate. This would be a far too complicated calculation for any judge to make. It would also make the mechanism of rights superfluous, since they would then be replaceable with direct comparisons of how state measures affect all persons. So, the proportionality assessment involved in the right to religious freedom does not compare the limitation set by the law on specific persons’ opportunities to engage in religious conduct to the opportunities of other people.¹³

What does this difference in form mean for what we should think about the Fair Opportunity view? There are two possibilities. One is that the structure of legal rights to freedom of religion as implemented, for instance in ECHR, does not matter. The idea would be that philosophical theories are theories about what ought to be, so if actual legal measures do not fit the theory, this just shows that the measures should be reformed to fit the theory. Another possibility, however, is that the theories should—at least to some extent and in some respects—fit the legal measures for which they are supposed to provide a theory. If we need a theory of religious liberty to guide us in deciding how to understand the legal right to freedom of religion, the theory should be designed with this in mind. On such a view, if the legal right to freedom of religion is vertical and non-comparative in form, then the Fair Opportunity view cannot be an account of the normative logic of religious liberty claims, at least not those religious liberty claims governed by (the vertical part of) the right to freedom of religion.

One prominent example of the latter view are so-called practice dependence views as for instance articulated by Charles Beitz in his book The Idea of Human Rights (Beitz 2009). He proposed that, rather than developing a theory of human rights starting out from some idea of “natural” rights, we should think of international human rights as an emergent normative practice of global scope. Beitz’s idea is that we should grasp the practice’s norms and the kinds of justification to which they are open by examining the ways they are appealed to within the practice’s discourse, where human rights are norms suited primarily for the regulation of the conduct of states. Similar views have been proposed by Alan Buchanan, who also directs us to focus on the system of international legal human rights considered as such rather than on individual moral rights (Buchanan 2013), and Andrea Sangiovanni, who defends a general methodological approach under the label “practice dependence,” according to which we should evaluate actual practices, such as the human rights system, in light of the point and purpose of the practice and how the way in which it shapes relations between agents provide them with reasons for complying with it (Sangiovanni 2008, 2016, 2017).

While it is debatable how we should understand practice dependence as a methodological approach, what we are to understand the current “practice” to be in a given context, whether there is a single practice that calls for justification and what difference the appeal to the point and purpose of a practice makes (Lægaard 2019a), and whether such views leave us with sufficient critical distance to assess a given practice (Lægaard 2019b), practice dependence would make a difference for a discussion like the present one. The identified difference in structure between the legal right to freedom of religion and Patten’s theory raises the question whether a normative theory of a given type of phenomenon should reflect the structure of the phenomenon in question; here whether it is the former or the latter that should be revised. For present purposes, I mention practice dependence as a type of view that would imply that the theory should take the structure of the legal right into account. According to practice dependence, the difference in structure between the legal right to freedom of religion and Patten’s Fair Opportunity view suggests that the latter cannot provide a plausible theory of the former.

One possible upshot of this is that the Fair Opportunity view is rather a theory of religious discrimination. Where the right to freedom of religion is given in, for instance, article 9 in the ECHR, religious discrimination is regulated in, for instance, European Council Directive 2000/78/EC, which has subsequently been implemented in national law in the EU member states, according to which “any direct or indirect discrimination based on religion or belief, disability, age or sexual orientation as regards the areas covered by this Directive should be prohibited throughout the Community.” Religious discrimination is, in the nature of the case, comparative. So, it seems that the Fair Opportunity view offers an account of the right form to provide normative guidance in cases of religious discrimination. This is especially the case for cases of indirect religious discrimination, i.e., cases where a neutral provision or rule that does not directly target religious people nevertheless has disparate impact on religious people. In such cases, the question then is whether there are sufficiently good reasons for the provision or rule, in which case it might be justifiable irrespective of its disparate impact. If not, then the provision or rule might be an impermissible form of indirect religious discrimination, even though it does not directly target religious people.¹⁴ Patten’s account seems well suited to capture cases like this, since it precisely involves competing claims at the same level.

So, one possible upshot of my point about the structure of the Fair Opportunity view is that it might be suited as an account for handling cases of religious discrimination but that it is not suited for handling cases involving the right to religious freedom. To this, one might either respond by rejecting the practice dependence view that informs this argument, or one might question whether the difference in structure is sufficient to show that the Fair Opportunity view cannot be a good theory of the right to religious freedom. Perhaps the structural difference is not that important, and the right to freedom of religion should be reformed and articulated more in line with what is suggested by the Fair Opportunity view. The implication of this would be that we should subsume the right to religious freedom under rules about religious discrimination.

However, there are reasons for thinking that the difference in structure is important in a way that does not allow this kind of reform, at least not if we still hold on to some form of practice dependence. Not only is the right to religious freedom vertical and non-comparative, whereas cases of indirect discrimination are horizontal and comparative, the legal implementation of these two normative mechanisms also apply to different actors. The right to religious liberty as articulated in both human rights documents and various constitutions apply to states. It is only states that are bound directly by human rights. This means that the right to freedom of religion addresses cases where a state regulates its citizens’ liberty. This is why the right to freedom of religion has a vertical structure; the question it addresses is not how the opportunities of different people relate to each other, but how a state can justifiably limit the opportunities of its citizens. So, limits on religious freedom are assessed in terms of what the state’s aim is, not in terms of which claims other people have. Prohibitions of discrimination, to the contrary, also apply to private actors such as employers and service providers. They therefore concern cases where different citizens come into conflict, and therefore cases of discrimination often involve competing claims at the same level that can be compared to each other. So, the difference in structure is explained by a difference in what practice dependence theory terms the “point and purpose” of the measures. If we respond to my point by saying that the two mechanisms should simply be merged into one, then we ignore these differences in purpose and the practice dependence view that theories should take the point and purpose of measures into account.

Given that there is this difference in form and that we need different mechanisms for regulating different relations that fit each of these forms, my initial worry about how the Fair Opportunity view does not have a form suited for capturing the issues involved in the right to religious freedom seems to stand.

Conclusion

I have argued that Patten’s Fair Opportunity view faces several problems. Even though the view is not yet fully worked out so that we can discuss whether it yields plausible normative recommendations in particular cases, one might worry that what would do the normative work in such specifications is the added accounts of reasonable claims and of justifiability so that the Fair Opportunity view itself does not provide much in terms of a normative criterion. This worry is further exacerbated by the observation that some specifications of the Fair Opportunity view can be extensionally indistinguishable from the No Burden or the No Discrimination views, to which the Fair Opportunity view was supposed to provide a normative alternative. These worries push any defense of the Fair Opportunity view in the direction of a focus on the preferability of its structural features as an account of the normative logic of religious liberty. However, precisely as a primarily structural view, the Fair Opportunity view seems to have the wrong form to capture the central mechanism for regulating religious liberty, namely the right to freedom of religion. While one response to this might simply be “so much worse for the right to freedom of religion,” I have provided reasons for resisting the assimilation of the right to freedom of religion to the horizontal and comparative form of the Fair Opportunity view. If the point of the right to freedom of religion is to address how states limit the liberty of citizens, then there are good reasons why the right to freedom of religion has this vertical and non-comparative form, which the Fair Opportunity view does not. These reasons are especially weighty if one adopts a practice dependent approach, which assigns some importance to how the right to freedom of religion actually works and to the fact that it is designed to serve a particular purpose, namely, to regulate the actions of states relative to their citizens. If one rejects this appeal to the purpose of the human rights system or to the claim that these rights should be able to regulate this specific relation, then it is less obvious that the fact that the Fair Opportunity view has a different form than the right to freedom of religion is a problem. Patten’s theory thus provides an example of a case where methodological views like practice dependence makes a difference and raises the question whether he intends his account to address the practice of human rights or to be in principle independent of it.