Introduction
Political inclusion is essential to the democratic ideal of self-government. However, in debates about the proper place of religion in democratic societies, a key issue of contention is whether democracy and secularism are necessarily connected. Fears of such a connection lead some critics of liberalism to the conclusion that liberal democratic institutions are ultimately incompatible with religious forms of life.Footnote 1 Needless to say, if there is no hope that secular and religious citizens can take ownership of and identify with democratic institutions in equal measure, then the future of democracy within pluralist societies is seriously threatened.
These fears commonly arise in debates about the criterion of democratic legitimacy that deliberative democrats endorse, according to which citizens ought to justify the imposition of coercive policies on one another with reasons that everyone can reasonably accept.Footnote 2 Since religious reasons are not generally acceptable to secular citizens and citizens of different faiths as a legitimate basis for coercively backed laws, endorsing this criterion is often understood to entail the claim that religious reasons ought to be excluded from processes of political justification (e.g., see Audi Reference Audi2000). This view suggests that commitment to democracy is ideally suited to secular citizens, but only suitable for those religious citizens who are willing and able to ignore their religious beliefs when forming their political convictions. Since this sort of requirement will impact religious citizens in exclusionary ways, so the argument goes, the secular state may need to find compensatory accommodations for religious citizens whose idiosyncratic religious beliefs and practices cannot be easily reconciled with, translated, or otherwise integrated into a secular outlook. Religious citizens may be tolerated, perhaps even accommodated, but not politically included as equals.
Understandably, critics of this view argue that singling out religion for exclusion from political justification is unfair to religious citizens and incompatible with the democratic ideal of treating all citizens as free and equal. In their opinion, giving equal consideration to everyone’s views is the only way to grant equal treatment to all citizens. This, in turn, requires the inclusion of religious reasons on equal footing with secular reasons in political deliberation (Wolterstorff Reference Wolterstorff and Weithman1997, 176–177). This proposal, however, amounts to relinquishing the criterion of democratic legitimacy that deliberative democrats endorse. According to this deep pluralist model, a majority of religious citizens would be licensed in basing their political decisions on exclusively religious reasons. They would thereby be able to impose coercive policies on other citizens without any obligation to give them reasons that they too can reasonably accept.
The difficulties involved on both sides of the debate have been spelled out in recent years. I will not rehearse this long-standing debate here. What I would like to focus on instead is the pernicious effects of framing the debate on public reason and political justification in terms of the distinction between religion and secularism. This conceptual framework suggests a direct connection between a political commitment to democracy, on the one hand, and the secular or religious identity of citizens, on the other. This seems problematic in several ways. First of all, the values of freedom and equality that are essential to the democratic ideal of self-government are neither religious nor secular values. They are political values that can be endorsed (or rejected) from within many different comprehensive doctrines, whether secular or religious. Second, foregrounding the religious identity of citizens over and above their political identity and suggesting a link between religious identity and special strains of commitment to democratic principles is not a very promising strategy for securing political inclusion in pluralist democratic societies. Moreover, from a political perspective, the suggestion is quite implausible. Antidemocratic extremist groups in current democratic societies attract secular citizens just as much as religious ones. Certainly, a citizen’s secular identity is not a reliable indicator of holding democratic political views.Footnote 3 But if the relevant predictor of commitment to democratic principles is the political identity of citizens and not their religious or secular identity, it seems both misguided (as well as counterproductive for the purposes of strengthening such commitment) to strongly link an account of democratic legitimacy to the religious or secular identity of citizens (Kim Reference Kim2008).Footnote 4
In what follows, I will question the need—not to mention the wisdom—of taking the category of religion as central for an account of democratic legitimacy. In my view, placing central importance on the religious–secular distinction in an account of democratic legitimacy is both unhelpful and misleading.Footnote 5 Although religious reasons may be taken to be a paradigmatic case of reasons that are not generally acceptable to secular citizens and citizens of different faiths, it does not follow that secular reasons can be seen as any more or less generally acceptable than religious reasons simply by virtue of being secular.Footnote 6 Nonreligious reasons that are based on conflicting comprehensive doctrines and conceptions of the good cannot be expected to be generally acceptable to all citizens as a legitimate basis for coercion, whether or not they are secular. Similarly, although a theocratic state might be a paradigmatic example of a political system that is incompatible with constitutional democracy, this does not mean that a secular state can be automatically considered any more or less democratic than a theocracy simply in virtue of being secular. North Korea is a secular state but certainly not a democracy. But if the religious–secular distinction does not capture the decisive features that distinguish democratic from nondemocratic states or that differentiate legitimate from illegitimate reasons for justifying state coercion within democratic societies, then placing central importance on such a distinction in an account of political justification and democratic legitimacy is misleading at best. It seems better to rely on conceptual categories that exhibit the required explanatory properties. Rawls follows this strategy in his account of public reason, which makes no use of the religious–secular distinction at all.Footnote 7 However, although the Rawlsian account is at the center of current debates on political justification and the liberal criterion of democratic legitimacy, the debate largely continues to be framed in terms of the religious–secular distinction.Footnote 8
In order to motivate the abandonment of this conceptual framework, in the second section, I show that approaches to political justification that rely on the religious–secular distinction all share the same fundamental weakness. Although the views defended by each side of the debate are diametrically opposed, their different accounts of political justification exhibit similar democratic deficits. In particular, none of these views can explain how citizens in pluralist societies, whether religious or secular, can all take ownership of and identify with the institutions of constitutional democracy in equal measure. These approaches thus fail to vindicate the democratic ideal of self-government that motivates deliberative democracy’s criterion of democratic legitimacy in the first place. Taking this democratic concern as a guide, in the third section, I defend a conception of public reason that does not rely on the religious–secular distinction and that, in my view, does not produce a similar democratic deficit. This account provides a differentiated answer to the question of whether religion is special.Footnote 9 Religion is not politically special. Religious beliefs and reasons do not have to be uniquely singled out for exclusion in the context of justifying political decision-making in democratic societies. However, as I will briefly show in the concluding section, this is compatible with recognizing religion as socially special, although not uniquely so (4).
Political justification and the religious–secular distinction: exclusion, inclusion, and translation models
According to the criterion of legitimacy that deliberative democrats endorse, citizens owe one another justifications based on reasons that everyone can reasonably accept for the coercive policies with which they must comply.Footnote 10 Only in this way can citizens see themselves not simply as subject to the law but also as authors of the law, as the democratic ideal of self-government requires. To the extent that citizens can mutually justify the political coercion they exercise over one another, they can see themselves as co-legislators or political equals. From the perspective of the democratic ideal of self-government, the essential contribution of public deliberation to democratic legitimacy is that it enables citizens to endorse the laws and policies to which they are subject as their own. In the absence of a commitment to mutual justification, citizens would see themselves as subject to sheer coercion by others and become alienated from the political system. As Audi succinctly puts it, “this kind of basis of coercion breeds alienation” (Audi Reference Audi2000, 67). Thus, the democratic ideal of ensuring that citizens can take ownership of and identify with the political decisions to which they are subject provides us with an important benchmark for assessing the extent to which different approaches to public reason and political justification actually live up to such an ideal.
The exclusion model
Audi’s account of democratic legitimacy offers a clear example of an exclusionary approach that relies on the religious–secular distinction. His account is based on a “principle of secular justifications” that he explains as follows: “One has a prima facie obligation not to advocate or support any law or public policy that restricts human conduct unless one has, and is willing to offer, adequate secular reasons for this advocacy or support (say, for one’s vote)” (Audi Reference Audi2000, 86). To this principle, Audi adds the “principle of secular motivation,” which demands that the secular reasons citizens adduce to justify the policies they favor be strong enough to motivate their endorsement of such policies. According to this view, religious reasons can be included in political debate in the public sphere, but only if they are accompanied by corroborating secular reasons, which are the only reasons that count for the purposes of justifying coercive policies in the legislative process. This is why the obligations in question are stronger for state officials than for private citizens (Audi Reference Audi2000, 92).
Audi also points out that these obligations of citizenship are moral and not legal obligations, since any legal enforcement of such obligations would directly violate important features of constitutional democracies such as freedom of speech or the secret ballot (Audi Reference Audi2000, 86). Moreover, he also indicates that they are “prima facie” obligations and therefore might be overridden by conflicting considerations. The idea that “every ought implies a can” is relevant in this context, so Audi concedes that if a person cannot find an adequate secular reason for the policy she favors and has strong religious convictions in support of such a policy, she is not only not legally but also not morally required to either vote against her conscience or abstain. Addressing the objection that his principles might exclude many religious citizens from democratic participation, Audi cites the example of a religious citizen who is a pacifist for exclusively religious reasons and faces a vote on whether her country should go to war. He explains that in scenarios where a religious citizen cannot find adequate secular reasons for the policy she favors, “given the importance of her religious convictions to her, and given her intellectual and psychological capacities, voting no may be the rational thing to do. In that case, she is not ‘required’ by any moral principle I endorse to vote yes or abstain” (Audi Reference Audi2000, 95).
Seen in this light, Audi’s account of the proper behavior of citizens who engage in political advocacy and voting is not as exclusionary or constrained as advertised. In fact, his account seems to side with aspirational models of political justification. These models endorse the regulative ideal of sincerely trying to offer reasons that other citizens may reasonably accept, but contend that, since there is no guarantee that such efforts may succeed, the only acceptable alternative open to citizens in situations of failure is to vote based on whatever considerations they think are right, which might be exclusively religious. When attempts at finding public justifications fail, there is no other alternative but to fall back on the legitimacy of a purely procedural solution such as majority rule.
To the extent that Audi’s account of the obligations of citizenship does not require citizens to either vote against their conscience or abstain when their reasons are exclusively religious, it fails to uphold the liberal criterion of democratic legitimacy. In such scenarios, his account fails to offer a solution to the democratic concern that motivated the account in the first place, namely, that citizens within the losing minority “would feel alienated if coerced by the majority vote” based on exclusively religious reasons (Audi Reference Audi2000, 87). This is not to say that his account of the moral obligations of citizenship is wrong. It does seem reasonable to think that citizens in a democracy should not face the choice of either voting against their conscience or giving up their right to vote through abstention. The problem is that this account fails to explain how liberal democracies can nonetheless live up to the ideal of self-government. It remains unclear how the obligations of democratic citizenship are supposed to enable a form of political decision-making that all citizens can own and identify with, so that citizens in the minority can also see themselves as political equals instead of being simply coerced by the majority.
The secular translation model
Faced with the dilemma of either giving up the criterion of democratic legitimacy or jeopardizing the political integration of religious citizens, defenders of the secular translation model, such as Habermas, propose a middle path that imposes fewer exclusionary conditions on political justification (see Habermas Reference Habermas2006 and also Laborde Reference Laborde and Rivers2013). Habermas defends the exclusion model at the institutional level of parliaments, courts, ministries, and administrations, that is, in what he calls the formal public sphere. Accordingly, state officials must exercise restraint and appeal solely to secular reasons to justify their political decisions. Yet he proposes to eliminate the requirement that secular reasons be provided in political debates in the informal public sphere. Religious citizens who participate in political advocacy within the informal public sphere can offer exclusively religious reasons in support of the policies they favor in the hope that they may be successfully translated into secular reasons.Footnote 11 They only have to accept the “institutional translation proviso,” according to which only secular reasons count in justifying political decisions beyond the threshold of state institutions. Given that religious citizens “may only express themselves in a religious idiom under the condition that they recognize the institutional translation proviso, they can, trusting that their fellow citizens will cooperate for accomplishing a translation, grasp themselves as participants in the legislative process, although only secular reasons count therein” (Habermas Reference Habermas2006, 10). The idea behind this proposal is that the state must provide secular reasons for the coercive policies it imposes on all citizens to meet the liberal criterion of democratic legitimacy. But, so the argument goes, there is no reason to transform the strict institutional demands of the secular state into obligations of citizenship, which would impede the political participation of religious citizens who are neither officials nor candidates for office. As Cécile Laborde succinctly puts it, “the state must be secular so that ordinary citizens do not, themselves, have to be secular” (Laborde Reference Laborde and Rivers2013, 20). To the contrary, they are enabled to exercise their democratic rights, including rights of free speech, conscience, and free exercise of religion.
By eliminating any constraints on the kind of justifications that ordinary citizens can offer in their political advocacy, the secular translation model might seem to address the democratic concerns raised by the exclusion model. However, this model generates democratic deficits of its own. The model eliminates justificatory constraints in the informal public sphere but at the price of disconnecting the process of opinion- and will-formation in which citizens participate from the outcomes of the legislative (and judicial) processes to which citizens are, in fact, subject. It is easy to imagine scenarios in which political debates within the informal public sphere lead a majority of citizens to reject some coercive policies based on exclusively religious reasons, whereas the available secular reasons lead state officials to enact or uphold these policies. By driving a wedge between the reasons and justifications backing state policies and those that ordinary citizens actually endorse, the secular translation model undermines the democratic ideal that motivated the criterion of democratic legitimacy in the first place.
As already mentioned, the point of requiring citizens to give reasons that everyone can reasonably accept for the coercive power that they exercise over one another is to ensure that citizens can identify with the policies to which they are subject and accept them as their own, rather than being simply coerced into compliance. The more officials are permitted (even required) to make their political decisions based on reasons that ordinary citizens do not share (or may even directly oppose), the more alienated the latter would be from the laws they have to comply with. Over time, the disconnect between processes of opinion- and will-formation in the informal and the formal public sphere would lead ordinary citizens to be estranged from the policies to which they are subject. The democratic ideal of self-government hangs in the balance. If democratic legitimacy requires that the laws to which citizens are subject track their interests, beliefs, and ways of reasoning, an account of public reason needs to explain how justificatory processes in the public sphere can help shape and transform public opinion such that ordinary citizens can continue to see the laws that demand their compliance as reasonably acceptable and meriting endorsement. The problem with the secular translation model is not so much that it fails to explain how citizens can get traction within one another’s views so that a joint, considered public opinion may emerge. The problem is that it does not even try.
Indeed, the secular translation model works in the exact opposite direction. It lets ordinary citizens keep their religious reasons and political opinions intact, so long as they are willing to give up the democratic expectation that the laws to which the secular state subjects them will track their reasons and opinions whenever translation fails. Obviously, asking ordinary citizens to blindly defer to state officials’ justifications for political decisions, even if they disagree with those decisions, would prevent citizens from taking ownership of and identifying with the laws and policies to which they are subject. Seen from this angle, the “institutional translation proviso” seems inimical to the democratic ideal. It amounts to relegating ordinary citizens to the passive role of mere subjects of the law, whenever the religious reasons for the policies they favor fail to be translated into secular ones. In those situations, it seems that they are asked to trade their civic rights to freedom of speech, conscience, and free exercise of religion for their political rights to shape the laws to which they are subject as political equals.
This immediately raises the question of what happens when citizens insist on exercising all their democratic rights, not just civic rights of free speech or conscience, but their political right to vote. Indeed, the existing defenses of the secular translation model all contain a notorious lacuna: in situations where ordinary citizens have exclusively religious reasons for the policies they favor, such a model offers no clarification for how they are supposed to cast their votes on substantive policies (e.g., in a referendum on same-sex marriage). It seems that the secular translation model can work as intended only if, as elite conceptions of democracy recommend, citizens are only allowed to select their representatives but never to directly vote for any substantive policies (see e.g. Schumpeter Reference Schumpeter1942, 263 ff.). However, as soon as popular initiatives and referenda become a possibility, as it is the case in many democratic societies, the question arises as to what ordinary citizens are supposed to do when they exercise the office of citizenship as co-legislators.
Although Habermas does not explicitly answer this question, in the context of criticizing Audi’s principle of secular motivation, he suggests that religious citizens should be able to vote their conscience (see Habermas Reference Habermas2006, 9). However, if this is so, then the secular translation model fails to explain how the criterion of democratic legitimacy can be met in situations where a majority of citizens vote based on exclusively religious reasons and thereby impose coercive policies on other citizens—apparently without any obligation to give them reasons that they can reasonably accept. Given this possibility, the secular translation model seems to collapse into the inclusion model it aims to reject. The other horn of the dilemma is hardly more attractive. If, in order to meet the legitimacy condition, citizens are required either to vote for policies they oppose or abstain, the secular translation model collapses into the exclusion model and thus fails to offer any answer to the objections it explicitly aims to avoid. For the purposes of exercising their democratic right to vote, religious citizens—pace Laborde—do have to become secular after all.
The inclusion model
In light of the difficulties encountered by different attempts to exclude religious reasons from political justification, defenders of the inclusion model, such as Wolterstorff contend that any differential treatment of religious reasons within political justification is unfair to religious citizens and is therefore incompatible with the core values of liberal democracy (see Eberle Reference Eberle2009; Eberle and Cuneo Reference Eberle, Cuneo, Zalta and Nodelmann2015; Wolterstorff Reference Wolterstorff and Weithman1997, Reference Wolterstorff2012). Against the claim that public reasons should take priority over religious reasons when justifying coercive policies, they contend that giving equal consideration to everyone’s views is the only way to grant equal treatment to all citizens. This, in turn, requires the inclusion of religious reasons on equal footing with secular reasons in political deliberation. If giving priority to some type of substantive reasons over others in making political decisions cannot be justified in a way that all citizens can accept, then the only remaining alternative is to fall back on a purely procedural solution such as majority rule. Moreover, since all existing democracies endorse secret ballots and freedom of speech, the norms embodied in actual democratic practices suggest that there is nothing wrong with letting citizens vote on the basis of whatever reasons they see fit. As defenders of the inclusion model contend, the fact that the public reason conception seems unable to account for the legitimacy of these key institutional features of liberal democracies is yet another factor that counts against the plausibility of such a conception.Footnote 12 If this is the case, then citizens seriously committed to the legitimacy of liberal democracy do not have to subscribe to the priority of public reasons.
I totally agree with the institutional perspective that underlies this criticism. Framing the debate on the proper conception of political justification exclusively in terms of the ethics of democratic citizenship and the moral duty of civility is quite misleading. The exclusively ethical perspective suggests that the debate turns on whether or not citizens should follow some ideal moral norms and principles when engaging in political activities, whereas the fundamental question is whether or not citizens can, upon reflection, endorse the ideal norms and principles actually embodied in the democratic institutions and practices in which they participate. However, I disagree with the claim that the priority of public reason is in tension with the institutions of constitutional democracy. In my view, the priority of public reasons is a necessary component of any plausible account of the legitimacy of the institutions of constitutional democracy. Defenders of those institutions may disagree with specific interpretations of the priority of public reasons but, whichever version they favor, they cannot dispense with the priority altogether.
However, in order to avoid the difficulties of the exclusion and translation models analyzed in the prior section, the alternative conception of the priority of public reasons I defend varies in two significant ways. First, in contrast to those models, my conception follows Rawls’s characterization of the content of public reasons as “properly political” reasons. It therefore makes no use of the religious–secular distinction. Second, and here my approach differs from Rawls’s, my account of the priority of public reasons draws from an institutional and not merely an ethical perspective. It does not simply rely on citizens’ moral duty to respect the priority of public reasons, let alone any legal duty to do so. Above all, it relies on citizens’ legal rights to contest coercive policies that they think violate the priority of public reasons by making use of the institutions of judicial review.
Political justification beyond the religious–secular distinction: the prioritizing model
The public reason conception I defend is based on a specific interpretation of the three distinctive claims that characterize such conceptions, namely, that (1) there is a set of reasons that are generally acceptable to all democratic citizens, that (2) these reasons are independent from religious or otherwise comprehensive doctrines, and that (3) they ought to have priority in determining coercive policies.Footnote 13 A defense of the first claim requires identifying reasons and arguments of a certain kind that all democratic citizens, whether religious or secular, can reasonably accept, ought to have priority for justifying coercive policies. However, I find that characterizing public reasons in terms of special epistemic properties such as being “secular,” “accessible,” “intelligible,” “shareable,” and so on is highly misleading. Instead, my proposal follows Rawls in identifying public reasons as “properly political” reasons. These are reasons based on those political values and ideals that are the very condition of possibility for a democracy: the ideal of treating citizens as free and equal, and of society as a fair scheme of cooperation. These values and ideals find expression in the constitutional principles to which citizens are bound in liberal democracies. As Rawls’s characterization makes clear, freedom and equality, fairness, social cooperation, and the like are neither secular nor religious ideals or values. They are democratic political values that can be endorsed from within many different comprehensive doctrines, whether secular or religious. In that sense, they are independent of any particular comprehensive doctrine. These democratic values and ideals embedded in the institutions of constitutional democracies provide a reservoir of generally acceptable reasons from which all citizens can draw to publicly justify the coercive policies they endorse to their fellow citizens (see Rawls Reference Rawls1993, 212–254, Reference Rawls and Freeman1999).
An advantage of the political interpretation of the content of public reasons is that it does not face the kind of skeptical doubts that plague epistemic interpretations.Footnote 14 Since democratic citizens are precisely the citizens committed to the values and principles of constitutional democracies, it is platitudinous to claim that they share the reasons that articulate those values or that they find them generally acceptable. The standard objection is not that this set of reasons does not exist, but rather that the set is too thin to provide a sufficient basis for determining which coercive policies are justified. However, in contrast to Rawls, my proposal is not committed to the “completeness of public reason.”Footnote 15 The claim that public reasons take priority for the purposes of justifying coercive policies is not equivalent to the claim that public reasons alone must be sufficient to provide such justification or that they must be the only reasons that citizens can legitimately appeal to for that purpose. Perhaps the best way to explain the difference is by focusing on the second claim mentioned above, namely, that public reasons are independent from religious (or otherwise comprehensive) doctrines.
This claim is usually cashed out in terms of “neutrality” and, as such, it has been the target of the most vigorous criticisms of the public reason view.Footnote 16 However, it is important to see why this is so. If, following Rawls, one endorses the completeness of public reason, namely, the view that there is a set of reasons shared by all democratic citizens that are sufficient to determine all or nearly all policies that touch upon constitutional essentials and matters of basic justice, then the claim that this set of reasons is independent from all religious or otherwise comprehensive conceptions of the good becomes quite problematic. For it suggests that one could determine the policies that ought to be enforced without any consideration whatsoever as to why they are good. That can’t be right. However, notice that what creates the problem is the assumption of “sufficiency” and not the assumption of “independence.” The problem is not that public reasons are indistinguishable from reasons that are religious or otherwise comprehensive, but rather that the latter cannot be excluded from the set of reasons sufficient to determine the policies that ought to be enforced. Without the assumption of sufficiency, however, all that is needed to justify the claim that public reasons are independent from other types of reasons is the capacity to intuitively distinguish them for the purposes at hand.
My interpretation of the independence claim is based on the intuitive contrast between, on the one hand, reasons and arguments that aim to show whether or not some specific policy is good, desirable, beneficial, valuable, and the like and, on the other, reasons and arguments that aim to show whether or not the policy in question is compatible with the equal protection of the fundamental rights of all citizens. This contrast can be understood as a specific case of a more general distinction between the rationale that motivates a practice and its justification. This is a familiar contrast. The reason why people marry, travel, or go to the movies is that they find these practices good, valuable, desirable, or whatever the case may be. However, this does not yet tell us whether or under which conditions these practices are justified. For present purposes, we can interpret the contrast in terms of Rawls’s catchy characterization of the difference between the right and the good: “the right draws the limit; the good shows the point” (Rawls Reference Rawls and Herman2000, p. 231).
Notice that this way of understanding the logical independence between both types of reasons does not involve any problematic assumption of neutrality. Indeed, if we interpret the claim of independence in this way, it becomes clear that arguments and reasons geared to show the point or rationale of a given practice cannot be “neutral” or independent of conceptions of the good, be they religious or secular, since they aim to show why the practice in question is good (i.e. valuable, important, beneficial, etc.). It seems clear that a crucial element of advocating for the adoption of a specific policy is to offer arguments and reasons that purport to show why the practices the policy regulates are good, beneficial, worth protecting, or whatever the case may be.
However, it seems equally clear that offering these kinds of arguments or reasons may not be enough to justify the adoption of the policy in question. For its justification may also depend on other kinds of considerations or constraints, for example, whether it is compatible with other practices, whether its benefits and burdens can be fairly distributed, whether it would excessively constrain important rights and freedoms, whether it would have discriminatory effects, and so forth. This indicates a sense in which the latter considerations—public reasons—may have constraining priority over the former—nonpublic reasons—without in any way annulling their relevance and import.
Take the example of same-sex marriage. LGBTQ citizens want to be able to marry because of the value of marriage, that is, because they find the institution good, beneficial, desirable, or whatever the case may be. Certainly, no one wants to marry for the sake of freedom and equality. However, this does not mean that equal treatment or protection of freedom are not important considerations, perhaps even decisive ones, for justifying whether same-sex marriage should be permitted or its ban overruled as unconstitutional. This intuitive distinction indicates how the priority of public reasons can be defended without the additional burden of a commitment to neutrality. In contrast to proposals that either exclude religious or otherwise comprehensive views from public debate or that include them without any restrictions, my proposal articulates a policy of mutual accountability that imposes the same deliberative rights and obligations upon all democratic citizens.Footnote 17 This proposal recognizes the right of all democratic citizens to adopt their own cognitive stance, whether religious or secular, in public political debates without giving up on the democratic obligation to justify the coercive policies with which all citizens must comply by providing reasons that are acceptable to everyone.
According to the mutual accountability proviso I defend, citizens who participate in political advocacy can appeal to whatever reasons they wish in support of the policies they favor, provided they are also prepared to show—against objections—that these policies are compatible with the democratic commitment to treat all citizens as free and equal, and can therefore be reasonably accepted by everyone. To fulfill this democratic obligation, citizens must be willing to engage in an argument aimed at showing how their favored policies are compatible with the equal protection of the fundamental rights and freedoms of all citizens, and they must be willing to accept the outcome of that argument as decisive in settling the question of whether these policies can be legitimately enforced. Objections to the compatibility of such policies with the equal protection of the fundamental rights and freedoms of all citizens must be (1) properly addressed in public debate and (2) defeated with compelling arguments before citizens’ support (or vote) for their enforcement can be considered legitimate.
It is in virtue of this democratic obligation that public reasons have a constraining priority. They are the only reasons toward which no one can remain indifferent in their political advocacy. Whereas public reasons need not be the source from which a rationale in support of coercive policies must be crafted, they are the kind of reasons that cannot be ignored, disregarded, or simply overridden once citizens bring them into public deliberation. They are the reasons that must be addressed and properly scrutinized in public debate if they are offered up as objections to the coercive policies under discussion. Since citizens of a constitutional democracy are committed to the equal protection of all citizens’ fundamental rights and freedoms, it is perfectly appropriate for them to call each other to account regarding the kind of reasons that they are considering or ignoring while advocating for the policies they favor, as this allows them to establish whether or not these reasons are compatible with maintaining that commitment. Granted, the shared commitment does not suffice to guarantee agreement.Footnote 18 But it does give rise to forms of argumentative entanglement that enable members of a political community to transform public opinion over time by their continuous efforts to enlist the force of the better argument to their cause and change each other’s minds and hearts. Given citizens’ shared commitment, they can legitimately ask that convincing reasons and considerations be provided in public debate to show that the policies under dispute are in fact compatible with the equal protection of all citizens’ basic rights and freedoms, and that this consideration be given priority in determining whether the policies are legitimate, regardless of other considerations that might also speak in their favor.
So far, I have followed the standard approach to public reason by presenting the duty of mutual accountability from the ethical perspective of the obligations of citizenship. However, as I mentioned before, this duty should not be understood as simply expressing a regulative moral ideal that citizens may or may not be required to comply with when advocating and voting for specific policies. To the contrary, as I will show in what follows, the actual significance of the duty of mutual accountability is that it makes explicit ideal norms and principles that are implicit within practices and institutions of constitutional democracies—such as judicial review and citizens’ rights to legal contestation. These institutional devices provide legal support for transforming what would otherwise amount to mere moral aspirations into effective constraints upon the kind of public deliberation and legislation that characterizes constitutional democracies and distinguishes them from nondemocratic forms of political organization. Indeed, if constitutional democracies had no institutional means to secure the priority of public reasons, that is, if securing that priority amounted to nothing more than a (defeasible) moral obligation or aspiration, then the equal protection of all citizens’ fundamental rights and freedoms, especially those of dissenting minorities, would be quickly eroded by the legislation of transient majorities, as is often observed in countries with mere “manifesto” constitutions that are not backed up by effective institutions for their actual protection.
An institutional approach to democratic legitimacy
Adopting an institutional perspective for interpreting the mutual accountability proviso that I propose helps solve two puzzles that confront standard defenses of public reason. First, as we saw in the prior section, the exclusion and translation models offered no explanation for how the liberal criterion of democratic legitimacy can be upheld if citizens are morally permitted to vote their conscience on substantive issues, even if their reasons are exclusively religious or otherwise comprehensive. My interpretation of the duty of mutual accountability also acknowledges that citizens are morally permitted to vote their conscience. However, in doing so, the liberal criterion of democratic legitimacy is not undermined because, according to my approach—and in keeping with the actual institutions of constitutional democracies—ensuring that the legitimacy condition is met does not simply depend on citizens’ (defeasible) moral duty to respect the priority of public reason. Above all, it depends on their legal right to contest political decisions whenever they think the priority of public reason has been violated. Therefore, the fact that a majority of citizens may vote for coercive policies based on exclusively religious reasons does not mean that they can impose them on the dissenting minority without giving them public reasons. Citizens can legally challenge such policies if they think that they violate the priority of public reasons (e.g., violate any of their constitutional rights and freedoms). By prompting the judicial review of their constitutionality, they can ensure, on their own initiative, that justifications based on public reasons are provided, which can then be scrutinized and, if need be, challenged.
Adopting an institutional perspective is also crucial for addressing another difficulty posed by critics of public reason, namely, that by imposing substantive constraints on citizens’ political deliberation, defenders of public reason cannot account for key institutional features of constitutional democracies such as freedom of speech and secret ballots. Now, it is true that deep pluralist conceptions of democracy can account for the fact that the secret ballot allows citizens to vote on the basis of whatever reasons they see fit. However, in constitutional democracies, this is not the whole story. What also needs to be accounted for is the significant fact that such decisions may be overruled if they are deemed to be unconstitutional.Footnote 19 That is, defenders of the inclusion model need to account for the fact that constitutional democracies impose a constraint upon how insensitive to reasons political decisions taken by secret ballot and majority rule can be. However, since this is a substantive constraint, it cannot be explained by resort to procedural fairness. Whereas secret ballot and majority rule can meet the fairness criterion of giving equal treatment to everyone’s views, constitutional review cannot get off the ground on the basis of such a criterion.
Here it is important to note the crucial difference between voting rights and rights to legal contestation. Whereas the secret ballot enables citizens to exercise their right to vote on the basis of whatever reasons and considerations they see fit, this is not the case regarding citizens’ right to legal contestation. In order to trigger judicial review of coercive policies, citizens must engage in public deliberation. They must argue their case in court on the basis of reasons, considerations, and arguments that are suitable to demonstrate their unconstitutionality. Given its aim, this process simply cannot give equal consideration to everyone’s views. To the contrary, it must identify, evaluate, and reject precisely those views and reasons that support laws and policies that are unconstitutional (e.g. that are incompatible with the equal protection of the fundamental rights and freedoms of all citizens). No matter what specific institutional form this review process might take in different democratic societies, it is of necessity a process that is sensitive to substantive considerations about appropriate standards, reasons, and arguments.Footnote 20
In order for the process of constitutional review to be properly triggered, it must first be discerned and plausibly argued that the policy in question touches upon some constitutional essentials such as, for example, the protection of citizens’ fundamental rights, so that the revision of an otherwise legitimate majority decision (be it by citizens’ referenda or by the legislature) can be deemed appropriate in the first place. Once this has been determined, it must then be established whether the policy in question is incompatible with the equal protection of some fundamental rights and freedoms of citizens and, if so, why, to what extent, and so forth. Now, whatever the standards, criteria, reasons, and arguments may be that are needed and appropriate for such determination in each case, it is clear that they take priority over whatever other considerations may have determined the outcome of the voting process, since the former can legitimately overrule the latter. Citizens may disagree on the standards, reasons, and considerations appropriate for constitutional review and those appropriate for the voting process, but they cannot disagree about the priority of the former over the latter if they accept that majoritarian decisions may be legitimately overturned if deemed unconstitutional. Citizens’ endorsement of constitutional democracy is tantamount to their endorsement of that priority.
Now, since defenders of the inclusion model endorse constitutional democracy, they are committed to the view that “the state is to protect a schedule of basic rights and liberties enjoyed by all its citizens” (Eberle and Cuneo Reference Eberle, Cuneo, Zalta and Nodelmann2015, sec. 7). This indicates that their account of the proper behavior of citizens who engage in political advocacy and voting cannot be as unconstrained as advertised. As Wolterstorff points out, there is an important proviso: citizens should exercise their political voice on the basis of whatever reasons they wish, provided their actions fall within the boundaries of the constitution (Wolterstorff Reference Wolterstorff and Weithman1997, 180). However, once this crucial proviso is added, a tension between the key commitments of the inclusion model surfaces: on the one hand, a commitment to the equal protection of the basic rights and freedoms of all citizens and, on the other, a commitment to the equal consideration of all points of view that grounds the rejection of the priority of public reasons. As mentioned before, it is hard to see how the first commitment could find practical or institutional expression without any deviation from the second. If legislation should be subject to constitutionality constraints, if the latter can legitimately overrule the former, then it must be because the reasons and considerations that properly determine whether a piece of legislation is consistent with the equal protection of all citizens’ constitutional rights and freedoms can overrule other reasons and considerations in support of the policy in question, be they religious or otherwise comprehensive. If we adopt this institutional perspective, we can articulate an interpretation of the priority of public reasons and the duty of civility that is less restrictive than the standard view, but that more accurately reflects what is at stake behind the public reason conception of political justification.
Citizens’ right to legal contestation and argumentative entanglement
In constitutional democracies with judicial review, the right to legal contestation guarantees that all citizens can, on their own initiative, open or reopen a deliberative process in which reasons and justifications aimed at showing the constitutionality of a contested policy are made publicly available, such that they can be scrutinized and challenged with counterarguments that might lead public opinion to be transformed and prior decisions to be overturned.Footnote 21 Citizens’ right to question the constitutionality of any policy or statute by initiating legal challenges enables them to structure public debate on the policy in question as a debate about fundamental rights and freedoms and therefore as a debate in which the priority of public reasons (with its corresponding standards of scrutiny) must be respected. They can do so even if such structuring did not seem antecedently plausible to the rest of the citizenry, perhaps because they had framed it in other terms or because they had failed to foresee the impact that the policy would have on the fundamental rights and freedoms of certain citizens. At least by the time a statute or policy passes through the sluices of judicial review, citizens have to address the constitutionality question and take the priority of public reasons seriously, even if most of them had previously framed the debate in other terms or had given priority to other considerations. Obviously, a claim that a contested statute violates a fundamental right may turn out to be mistaken, and litigants may not be able to change a prior decision or public opinion. But, even in such a case, they still have the right to receive explicit, reasoned justifications of why exactly the statute in question does not violate their rights and why it is therefore compatible with treating them as free and equal. For those who continue to disagree, these reasoned public justifications highlight the type of considerations, arguments, and evidence that they would need to more effectively challenge in order to change the hearts and minds of their fellow citizens on the issue in question.
From this perspective, the right to legal contestation guarantees all citizens that their communicative power, their ability to trigger political deliberation on issues of fundamental rights and freedoms, won’t fall below some unacceptable deliberative minimum regardless of how unpopular or idiosyncratic their views may seem to other citizens.Footnote 22 The conception of public justification as mutual accountability that I defend emphasizes the contribution that structuring political debates in accordance with the priority of public reasons (and its corresponding standards of scrutiny) has upon the legitimacy of enforcing contested policies. Instead of leaving dissenting minorities with no other option but to blindly defer to majoritarian decisions based on secret ballots, the right to legal contestation gives rise to forms of argumentative entanglement that enable all members of a political community to gain traction within each other’s views and transform them over time.Footnote 23 Secret ballots might be blind. Public deliberation can’t be.
The priority of public reasons and religious forms of life
Debates in European countries on whether to ban the Islamic headscarf from public places offer a good example. These debates have mainly focused on the meaning of the practice of wearing the Islamic headscarf. On that question, there are deep disagreements. Some see it as a symbol of gender inequality, others as a mark of cultural identity, and still others as having a strongly religious significance. Even those who agree on its religious significance draw very different political conclusions. For some, this religious significance justifies the obligation of Muslim women to wear it, whereas for others, such significance justifies the need to ban it in order to prevent foreign religious values from displacing or threatening the Christian values of European countries. However, since political initiatives to ban the Islamic headscarf from public places became part of the political agenda in most European countries and citizens began to legally contest them in the courts, the focus has shifted from a debate on the cultural and religious meaning of wearing the headscarf to a debate on fundamental rights, equal treatment, and nondiscrimination.
This example is particularly interesting because it casts some doubt on the contention, held by defenders of the inclusion model, that the priority of public reasons is unfair to religious citizens and threatens religious forms of life in democratic societies. This view fails to appreciate that the priority of public reasons over comprehensive reasons offers strong protections to religious forms of life. As mentioned above, many citizens argue in favor of banning the Islamic headscarf from public places by appealing to secular reasons concerning gender equality. Now, it might seem that precisely because their advocacy of a ban draws on the value of equality, it is based on properly political reasons and therefore meets the “priority of public reasons” test. But herein lies an illicit conflation. For such a claim would seem to imply that secular reasons, whatever they may be, that purport to explain why the practice of wearing the Islamic scarf is bad (undesirable, etc.) are at the same time both appropriate and sufficient for the justification of something entirely different, namely, the imposition of coercion upon others who have the right to be treated as equals. The priority of public reasons over comprehensive reasons (whether religious or secular) that is implicit in constitutional review offers an effective protection against this conflation.
Indeed, judicial review of the constitutionality of the ban of Islamic headscarves from public places in European countries is slowly shifting the focus of the debate from the meaning and rationale of the practice of wearing the headscarf to the justification for its prohibition. As an example, the 2015 ruling by Germany’s Highest Court that a ban on teachers wearing headscarves is not compatible with religious freedom and that excepting Christian symbols from the ban constitutes religious discrimination (and is therefore unconstitutional) is helping to structure public political debates in accordance with the priority of public reasons and the duty of mutual accountability.Footnote 24 It highlights the arguments and reasons that need to be addressed and convincingly defeated before enforcement of the ban can be considered legitimate. Here again there are good reasons to think that the extra political power that the right to legal contestation gives litigants has helped them to structure the political debate as a constitutional debate about fundamental rights and freedoms, and that if such extra power were absent the “unfettered” public debate would continue to turn on comprehensive views (be they religious or secular) about which citizens strongly disagree. As a consequence, the comprehensive secular views of the majority about the meaning of the practice of wearing the Islamic headscarf would simply continue to dictate policy in Europe.
Is religion special? A differentiated answer
The institutional account of democratic legitimacy that is based on the prioritizing model makes no use of the secular–religious distinction. This account gives a differentiated answer to the question of whether religion is special. On the one hand, religion is not politically special. As we have seen, within the context of justifying political decision-making in democratic societies, religious beliefs and reasons do not have to be uniquely singled out for exclusion. On the other hand, this insight is compatible with a recognition that religion is socially special, though not uniquely so. Religious beliefs and practices deserve careful treatment because of their special characteristics (needs, goals, threats, vulnerabilities, etc.). But many other practices and institutions, such as science, the arts, the family, education, and the like, also merit such treatment. Since all of them deserve proper treatment based on their special features, the claim that religion is not uniquely special does not rule out that democratic states might offer some type of special treatment for religion (e.g. grant special accommodations or assistance to religious citizens or institutions). What that claim does rule out is the possibility of the state granting assistance or accommodations to religious citizens and institutions while denying those same assistance and accommodations to similarly situated or affected nonreligious citizens and institutions. In that regard, the prioritizing approach shares some similarities with Eisgruber and Sager’s egalitarian approach to religious exemptions (see Eisgruber and Sager Reference Eisgruber and Sager2007). However, in contrast to the latter, the prioritizing approach is not committed to the view that demands for exemptions are only acceptable on egalitarian grounds of preventing discrimination. Let’s see why.
Since the prioritizing model recognizes the important role that ethical values and reasons play in justifying coercive laws as well as in demands for exemptions from such laws, it has no difficulty in recognizing that considerations related to the intrinsic ethical importance of the affected beliefs, interests, or practices provide valid grounds for justifying such demands. These ethical considerations are, however, subject to the additional prioritizing constraint of showing—against any objections—that the demands for religious exemptions in question are indeed compatible with the equal protection of everyone’s fundamental rights and freedoms.Footnote 25 Yet the fact that an appeal to ethical values and grounds is insufficient to justify exemptions on its own in no way implies that it is not necessary for such justification, as Eisgruber and Sager’s egalitarian approach suggests.
This is a fundamental difference between the prioritizing approach and the egalitarian approach to religious exemptions defended by Eisgruber and Sager. On their account, equal treatment is the only valid rationale for justifying exemptions. Thus, “in the absence of a convincing equality-based claim,” churches would have no constitutional entitlement to an exemption from valid laws (Eisgruber and Sager Reference Eisgruber and Sager2007, 13). Such exemptions are “appealing and justifiable when and only when they function as proxies for the requirement of equal treatment” (Eisgruber and Sager Reference Eisgruber and Sager2007, 14). But this can’t be right. Justifying coercive laws and exemptions from such laws are two sides of the same coin. Thus, in the same way that justifying the enforcement of coercive laws requires showing that the practices which the law regulates are valuable, beneficial, worth protecting, or whatever the case may be, justifying the need for exemptions from such laws equally requires showing that the practices which should be exempted from the law are ethically important, beneficial, worth protecting, or whatever the case may be. The first-order ethical justification of why the practices in question deserve to be regulated by law (or exempted from such laws) in the first place cannot be replaced by the second-order constraint of showing that the regulations or exemptions in question are compatible with equal treatment.Footnote 26 Both argumentative steps are necessary and must be accomplished before any coercive laws or exemptions (from otherwise valid laws) can be legitimately implemented.
Although the difference between the two approaches may seem subtle, it has important practical implications. The prioritizing approach offers support for the core idea behind the egalitarian approach, namely, that religious beliefs, practices, and institutions do not deserve a uniquely special treatment. Indeed, since first-order ethical justifications are equally available to secular and religious citizens, and any exemptions granted for proper ethical reasons are subject to the prioritizing constraint of showing that they are indeed compatible with the equal protection of everyone’s fundamental rights and freedoms, the prioritizing approach can give a plausible account of Eisgruber and Sager’s complementary claims, namely that religiously motivated conduct in constitutional democracies neither deserves “special immunity from otherwise valid laws” nor is it subject to any “special disabilities” (Eisgruber and Sager Reference Eisgruber and Sager2007, 5). However, since the prioritizing approach recognizes the important role that ethical values and reasons play in justifying coercive laws and exemptions, it does not support Eisgruber and Sager’s reductionist claim that exemptions are only acceptable on egalitarian grounds of preventing discrimination. According to the prioritizing model, exemptions can be properly justified by appeal to other important considerations. Let me briefly mention one example.
In Liberalism’s Religion, Laborde discusses the case of justifications for exemptions that are based on claiming that the law in question imposes a disproportionate burden on some citizens. This type of consideration is not about equality and discrimination in the relevant sense. To evaluate its merits, one does not have to engage in comparative judgments about other citizens at all. As Laborde plausibly argues, disproportionate burden is a noncomparative principle. Its evaluation requires a “strict balancing test” which weighs up the interests pursued by the law, the severity of the burden, and the costs incurred in alleviating it. She provides the following example: “It would be unfair to compel Orthodox Jews to endure an invasive postmortem autopsy in case of nonsuspicious death, if they consider this a desecration of the body. There seems to be a disproportion between the aims pursued by the law and the burden it inflicts on the claimants” (Laborde Reference Laborde2017, 220). The unfairness of the burden in such cases is not due to unequal treatment or discrimination against religious citizens vis-à-vis other citizens. It may be due to the fact that granting an exemption would not undermine the aims of the law, and the cost of doing so would be negligible or reasonable. Needless to say, once an exemption is granted on such noncomparative grounds, any similarly situated citizens, whether religious or not, would prima facie have a case to be granted similar exemptions based on equal treatment grounds. Nonetheless, the need for noncomparative, first-order considerations to justify exemptions is particularly obvious in all cases where no precedent is yet available for drawing such comparative judgments.Footnote 27 In contrast to Eisgruber and Sager’s egalitarian approach, the prioritizing approach gives a plausible account of such cases while maintaining the constraining priority of equal treatment considerations for justifying religious exemptions.
Cristina Lafont is co-author (with Nadia Urbinati) of The Lottocratic Mentality: Defending Democracy against Lottocracy (OUP, 2024) and author of Democracy without Shortcuts. A Participatory Conception of Deliberative Democracy (OUP, 2020); Global Governance and Human Rights (van Gorcum, 2012); Heidegger, Language, and World-disclosure (CUP, 2000), and The Linguistic Turn in Hermeneutic Philosophy (MIT Press, 1999).
Open access
