A More Liberal Public Reason Liberalism

1 Introduction

Public reason liberals (henceforth, PRLs) frequently hold that publicly justifying coercive laws and policies requires citizens to offer public reasons for these laws and policies (e.g. Rawls 1987, p. 15, 1999, pp. 573–581; also De Marneffe 1994; O’Neill 1997; Quong 2004). The idea is that if the citizens targeted by a given coercive law or policy cannot reasonably accept this law or policy, then from the perspective of public reason liberalism (henceforth, PRL) such law or policy ‘lacks an adequate basis’ (Audi 2000, p. 123; also Rawls 2005, p. xlvi). Two principles of public justification are especially prominent among PRLs. First, we find the principle of secular rationale (henceforth, PSR), which holds that ‘one should not advocate … any law or public policy that restricts human conduct unless one has, and is willing to offer, adequate secular reason for this advocacy’, i.e. secular reason that suffices to justify the proposed law or policy (Audi 1989, p. 279; also Rawls 1993, p. 247). And second, there is the principle of secular motivation (henceforth, PSM), which holds that ‘one should not advocate … any legal or public policy restrictions on human conduct unless one not only has and is willing to offer, but is also motivated by adequate secular reason’ for this advocacy (Audi 1989, p. 284; also Audi 1991, 1993, 2000, ch. 4–5, 2011, ch. 3).^[1]

In recent years, intense debates have taken place regarding PSR and PSM across moral, social and political philosophy (e.g. Billingham 2016, 2018; Enoch 2013, 2015; Gaus 1996, 2011; Hertzberg 2015, 2018; Perry 2001, 2003; Vallier 2012, 2016a; Wall 2002, 2014; Weithman 1991, 2002). These debates have widespread implications for public deliberation in free and democratic societies. For public deliberation in free and democratic societies frequently targets coercive laws and policies (e.g. Ackerman 1994; Habermas 2006; Van Schoelandt 2015). And coercive laws and policies are proposed to regulate a wide range of issues of paramount moral, social and political importance (e.g. freedom of expression, freedom of education, reproductive rights).

In this paper, I join these debates and provide a critical assessment of PSR and PSM. I shall argue for two main claims concerning these two principles of public justification. First, only some qualified versions of PSR may be justifiably regarded as plausible liberal principles of public justification. And second, even if PSR constituted a plausible liberal principle of public justification, PSM does not constitute a plausible liberal principle of public justification and is untenable on multiple grounds. My claim is not merely that PSM is in need of improvement and should be modified in specific respects. Rather, my main thesis is that the public justifiability of coercive laws and policies crucially depends on what justificatory reasons citizens offer for and against these laws and policies rather than on what reasons happen to motivate such citizens. Hence, PSM’s focus on citizens’ motivating (rather than justificatory) reasons fails to track the set of reasons that bear on the public justifiability of coercive laws and policies, and requiring citizens to abide by PSM would threaten to hamper (rather than promote) public deliberation in free and democratic societies.

The contents are organized as follows. In Section 2, I outline PSR and PSM, explicating their interrelations and alleged justifications. In Section 3, I focus on PSM and provide three major reasons to support my thesis that PSM does not constitute a plausible liberal principle of public justification. In Sections 4–7, I buttress my critical assessment by considering and rebutting four influential defences that the proponents of PSR and PSM may put forward to support PSR and PSM, namely the sincerity defence (e.g. Audi 1989, p. 282, 2000, p. 109; also Macedo 1990, p. 293; Rawls 1993, p. 215), the respect defence (e.g. Audi 1989, p. 283, 2000, p. 110; also Cohen 1993, pp. 1541–1542; Larmore 1999, p. 608), the stability defence (e.g. Audi 1989, p. 282, 2011, p. 75; also Hartley and Watson 2020, pp. 891–893; Rawls 2005, pp. 133–172) and the fairness defence (e.g. Audi 1989, p. 294; also Hartley and Watson 2018, ch. 4; Macedo 1990, p. 295). In each of these sections, I first provide a critical assessment of PSM to support my thesis that PRLs should abandon (rather than just modify) PSM. I then distinguish various versions of PSR to demonstrate that only some qualified versions of PSR may be justifiably regarded as plausible liberal principles of public justification (e.g. Sections 4–7 on restricted and unrestricted versions of PSR).

Before proceeding, the following three preliminary remarks are in order. First, my critical assessment of PSR and PSM is internal to PRL and as such differs from former critiques that challenge PSR and PSM from ethical and/or theoretical standpoints external to PRL (e.g. Quinn 1995; Wolterstorff 1997a, on critiques grounded on other strands of liberalism; Neal 2000; Perry 2001, on religiously inspired critiques). In what follows, I focus on Audi’s (rather than other authors’) formulation of PSR and PSM because Audi has offered a commendably clear formulation of these principles and has developed one of the most systematic defences of such principles (e.g. Audi 1989, 1990, 1991, 1993, 2000, ch. 4–5, 2011, ch. 3). However, I shall expand on different versions of PSR and PSM whenever the differences between such versions directly bear on my evaluation (e.g. Sections 4–7). In this perspective, my critical assessment of PSR and PSM can be seen as a test case that sheds light on at least two general issues that are widely debated among PRLs, namely what role justificatory reasons and motivating reasons should respectively play in the public justification of coercive laws and policies and how other often-made distinctions between different categories of reasons (e.g. Sections 2–7 on religious, anti-religious and religiously neutral reasons) bear on the ongoing debate about the relative merits of specific principles of public justification.

Second, I shall rely on a broadly Rawlsian characterization of reasonable citizens as citizens who are ‘ready to propose [and abide by] fair terms of cooperation … given the assurance that others will likewise do so’ and accept ‘the burdens of judgment’, i.e. the many ‘hazards involved in the correct (and conscientious) exercise of our powers of reason and judgment in the ordinary course of political life’ (Rawls 2005, p. 49 and p. 56; also Brower 1994; Habermas 1995; Mang 2017; Wenar 1995, for discussion). This notion of reasonableness includes both ethical and epistemic elements. Some authors hold that to qualify as reasonable, citizens need only be ready to propose and abide by fair terms of cooperation (e.g. Nussbaum 2011). Yet, citizens who are reasonable only in this ethical sense may offer reasons that systematically fail to be consistent and responsive to evidence. And it is dubious that reasons that systematically fail to be consistent and responsive to evidence can contribute to publicly justifying coercive laws and policies. Because of this, most PRLs concur that ‘a plausible standard of [reasonableness] must include at least some epistemic elements’ (Wall 2014, p. 475; also Enoch 2017; Raz 1990; Vallier 2020a; Wolterstorff 2012, part 1, on different positions concerning the reasoning skills and the amount of information that reasonable citizens can be plausibly assumed to possess).

And third, PRLs advocate dissimilar accounts of what conditions reasons have to satisfy to qualify as public and, as such, be able to contribute to publicly justifying coercive laws and policies. The following tripartition is especially prominent in the PRL literature. Shareability accounts (e.g. Hartley and Watson 2009; Macedo 2000) hold that a reason for a given coercive law or policy can qualify as public if all the citizens targeted by this law or policy regard such reason as adequate based on shared evaluative standards (e.g. basic rules of logic and inference). For their part, accessibility accounts (e.g. Audi 2011, ch. 3; Badano and Bonotti 2020) hold that a reason for a given coercive law or policy can qualify as public if all the citizens targeted by this law or policy can regard such reason as adequate based on shared evaluative standards even if some citizens deny that the reason in question is adequate based on those standards. Still differently, intelligibility accounts (e.g. Gaus 2011, ch. 13–14; Vallier 2014, ch. 4) hold that a non-shareable and non-accessible reason for a given coercive law or policy can qualify as public provided that all the citizens targeted by this law or policy can regard such reason as adequate based on the (possibly varying) evaluative standards of the citizens who offer such reason. In recent years, intense debates have taken place concerning the relative merits of shareability, accessibility and intelligibility accounts (e.g. Boettcher 2015; Enoch 2005; Gaus and Van Schoelandt 2017; March and Steinmetz 2018). My evaluation of PSR and PSM does not directly rest on the merits of these accounts. However, I shall juxtapose distinct accounts when leading PRLs’ claims concerning PSR and PSM rest on the merits of such accounts (e.g. footnote no. 7).

2 PSR and PSM

PSR holds that ‘one should not advocate … any law or public policy that restricts human conduct unless one has, and is willing to offer, adequate secular reason for this advocacy’, i.e. secular reason that suffices to justify the proposed law or policy (Audi 1989, p. 279; also Rawls 1993, p. 247). For its part, PSM holds that ‘one should not advocate … any legal or public policy restrictions on human conduct unless one not only has and is willing to offer, but is also motivated by adequate secular reason’ for this advocacy (Audi 1989, p. 284). The idea is that ‘in addition to having and being willing to give’ adequate secular reasons for the proposed coercive laws and policies, citizens must ‘be (sufficiently) motivated by at least [some] such reason’ (Audi 1989, p. 293; also Audi 1991, 1993, 2000, ch. 4–5, 2011, ch. 3). The following remarks regarding PSR and PSM bear on the evaluation of these two principles as liberal principles of public justification.

First, PSR and PSM ‘express (prima facie) moral obligations’ (Audi 1989, p. 262, italics added; also Audi 1991, p. 75) which apply to both private citizens and public officials qua citizens of free and democratic societies (Audi 1989, p. 293; also Audi 1993, p. 701). PSR and PSM are principles of conscience rather than candidates for legislation (e.g. Audi 1991, p. 66). However, they ‘are not merely counsels of prudence’ (Quinn 1995, p. 36), and significantly constrain what kinds of reasons citizens can offer to publicly justify coercive laws and policies (e.g. Audi 1993, p. 678, on how PSR and PSM constrain ‘the explicit use of, or tacit reliance on, religious considerations’ in public deliberation; also Audi 1989, p. 279, holding that ‘citizens in a free and democratic society are obligated … not to make decisions, as citizens, in support of laws or policies that restrict … liberty, unless they have a sufficient secular basis for so deciding’). In particular, both PSR and PSM presuppose that religious reasons alone cannot publicly justify any coercive laws and policies (e.g. Audi 1993, p. 691; also Anderson 2006, p. 139; Beckwith 2002, p. 111). The idea is that religious reasons ‘are altogether legitimate in the order of discovery [where one chooses] what public issues to concentrate on [but] in the order of justification [one must] have, and be sufficiently motivated by, adequate secular reasons’ (Audi 1989, p. 293; also Audi 2011, p. 104). In this respect, both PSR and PSM are more restrictive than the position advocated by Rawls, who – after holding that ‘comprehensive views are never to be introduced into public reason’ unless citizens introduce such views ‘in ways that strengthen the idea of public reason’ (Rawls 1993, p. 247) – famously allowed that citizens may introduce their comprehensive views ‘at any time provided that in due course [they offer] public reasons [that are] sufficient to support whatever the comprehensive [views] are introduced to support’ (Rawls 2005, p. xlix; also Audi 2011, ch. 3; Cooke 2000; Thrasher and Vallier 2015, for discussion).^[2]

Second, PSM is significantly more demanding than PSR. For PSR does not per se impose any motivational requirement and only requires that citizens have and are willing to offer adequate secular reasons, irrespective of whether these reasons are actually the reasons that motivate the citizens who offer them (e.g. Audi 1989, p. 279; also Bardon 2018, p. 662). Conversely, PSM implies that if a citizen is not sufficiently motivated by adequate secular reasons for a coercive law or policy, then the citizen should refrain from advocating this law or policy even if she has and is willing to offer adequate secular justificatory reasons for such law or policy (e.g. Audi 1993, p. 692; also Anderson 2006, p. 140). Hence, one may fail to satisfy PSM even in cases where she satisfies PSR. To be sure, PSM allows that citizens may have religious motivating reasons for the coercive laws and policies they advocate (e.g. Audi 1993, p. 694, 2011, p. 71). Yet, when it comes to publicly justifying coercive laws and policies, PSM requires that at least some adequate secular reasons suffice to motivate citizens to advocate such laws and policies. The idea is that ‘if the spirit of separation of church and state is to be adequately reflected … we need something stronger than [PSR]’ (Audi 1989, p. 284) and that if a citizen advocates a coercive law or policy ‘wholly from a religious motive’, then the citizen fails to satisfy the relevant principles of public justification even if her advocacy ‘is rationalizable by appeal to secular reasons’ (Audi 1991, p. 70; also Audi 1989, p. 287, on various cases where ‘it makes an immense difference whether I adhere to [PSR] or [PSM]’).^[3]

Third, the reasons that citizens offer in public deliberation can be helpfully classified in terms of the following three exclusive categories (e.g. Audi 1989, p. 279, 1993, p. 692, 2011, pp. 67–78; also Anderson 2006, p. 146; Greenawalt 1995, p. 63; McConnell 1999, p. 641; Strike 1990, pp. 479–480). A reason counts as religious if and only if its normative force rests on the endorsement of theological/metaphysical claims concerning the existence (or the will) of God, the interpretations of putatively sacred texts or the pronouncements of spiritual authorities. A reason counts as anti-religious if and only if its normative force rests on the denial of theological/metaphysical claims concerning the existence (or the will) of God, the interpretations of putatively sacred texts or the pronouncements of spiritual authorities. A reason counts as religiously neutral if and only if it is neither religious nor anti-religious, i.e. its normative force rests on neither the endorsement nor the denial of theological/metaphysical claims concerning the existence (or the will) of God, the interpretations of putatively sacred texts or the pronouncements of spiritual authorities. Below I use the expression ‘non-religious reasons’ as a convenient shorthand to encompass both religiously neutral and anti-religious reasons. However, I shall differentiate between religiously neutral and anti-religious reasons whenever the differences between such reasons bear on my evaluation of PSR and PSM (e.g. Sections 4–7).^[4]

Finally, two importantly different interpretations of the expression ‘secular reasons’ figuring in PSR and PSM can be distinguished in the PRL literature. On the one hand, ‘secular reasons’ may be interpreted narrowly so as to designate only what I call religiously neutral reasons (e.g. Audi 1993, p. 692, 2011, p. 77, contrasting ‘anti-religious’ reasons, which ‘directly concern religion’, and ‘secular’ reasons, which are ‘evidentially independent of religion’). On the other hand, ‘secular reasons’ may be interpreted broadly so as to designate all non-religious reasons, i.e. both what I call religiously neutral reasons and what I call anti-religious reasons (e.g. Audi 1989, p. 279, 1993, p. 692, taking PSR and PSM to exclude only religious reasons from the set of public reasons). I am not concerned here with settling the exegetical issue of which of these two interpretations of ‘secular reasons’ is ultimately favoured by Audi. However, below I shall repeatedly contrast those two interpretations of ‘secular reasons’ in demarcating which versions of PSR may be justifiably regarded as plausible liberal principles of public justification (e.g. Sections 4–7 on restricted and unrestricted versions of PSR).

3 Against PSM

In this section, I provisionally assume – for the sake of argument – that PSR constitutes a plausible liberal principle of public justification and provide a critical assessment of PSM. I shall argue that PSM does not constitute a plausible liberal principle of public justification and is untenable on multiple grounds. More specifically, in points 3.1–3.3 below I articulate and defend three major criticisms of PSM as a liberal principle of public justification. First, PSM’s focus on citizens’ motivating (rather than justificatory) reasons fails to track the set of reasons that bear on the public justifiability of coercive laws and policies. Second, requiring citizens to abide by PSM would impose epistemically and evidentially overdemanding requirements on citizens, which threaten to hamper (rather than promote) public deliberation in free and democratic societies. And third, the proponents of PSM have hitherto failed to substantiate PSM’s assumption that the divide between non-religious and religious reasons reliably tracks the divide between reasons that respectively can and cannot contribute to publicly justifying coercive laws and policies.^[5]

4 Sincerity Defence

The sincerity defence aims to support PSM as a liberal principle of public justification on the alleged ground that allowing citizens to advocate coercive laws and policies on the basis of non-motivating reasons would encourage insincerity and manipulation on the part of citizens. The idea is that a citizen’s advocating coercive laws and policies on the basis of reasons by which she is not motivated implies some form of insincerity or manipulation on the part of this citizen because the citizen uses reasons ‘as psychological levers … on a basis that does not carry [her] own conviction’ (Audi 1989, p. 282; also Audi 2000, p. 109). This use of reasons, in turn, is morally problematic because ‘it smack[s] of manipulation to give reasons that do not move me, in order to get others to do what I want’ (Audi 1989, p. 282; also Whitfield 2020, p. 12). Hence, the defence goes, sincerity-related considerations vindicate PSM as a plausible liberal principle of public justification.

These considerations nicely fit leading PRLs’ view that citizens should offer in public deliberation only those reasons that they sincerely take to be adequate (e.g. Rawls 1993, p. 215, warning against ‘hypocritical … public discourse [where] citizens talk before one another one way and vote another’). Even so, it is dubious that sincerity-related considerations vindicate PSM as a plausible liberal principle of public justification. For the mere fact that one advocates some coercive law or policy on the basis of non-motivating reasons by no means excludes that one may sincerely take these reasons to justify such law or policy. In fact, one may sincerely believe that the reasons she offers in public deliberation suffice to justify a coercive law or policy in cases where she is not motivated by any of those reasons (e.g. Lister 2017; Schwartzman 2012). To illustrate this, consider the following modified version of Section 3’s case of a citizen who is motivated to advocate a minor legal restriction on abortion by some religious reasons, but also offers secular reasons that she sincerely takes to publicly justify such restriction. Suppose that the citizen advocates the legal restriction on the basis of these secular justificatory reasons, publicly acknowledges that her religious motivating reasons differ from the secular justificatory reasons she offers, and explicitly grants that if her fellow citizens show that the secular justificatory reasons she offers fail to justify the legal restriction she advocates, she will either offer additional secular justificatory reasons for this restriction or stop advocating such restriction. No insincerity or manipulation is implied by the citizen’s advocating coercive laws and policies on the basis of reasons by which she is not motivated.

Now, suppose – for the sake of argument – that a citizen’s advocating coercive laws and policies on the basis of reasons by which she is not motivated implied some form of insincerity or manipulation on the part of this citizen. Even under this supposition, the implied insincerity or manipulation would negatively bear on the moral evaluation of the citizen’s character, but would not directly bear on the issue whether the reasons offered by such citizen can contribute to publicly justifying coercive laws and policies. For on PRL, the public justifiability of coercive laws and policies depends on whether citizens offer adequate secular justificatory reasons for such laws and policies. And to meet this justificatory requirement, citizens need not disclose their own motivating reasons, let alone state all such reasons sincerely (e.g. Reidy 2000, pp. 57–63; Schwartzman 2011, p. 377). In this respect, it would be of limited import to reiterate that, according to leading PRLs, citizens should offer in public deliberation only those reasons that they sincerely believe to be adequate (e.g. Rawls 1993, p. 215). For PRL does not require that any of these reasons be motivating (e.g. Reidy 2000, p. 61; Schwartzman 2011, p. 390). In fact, leading PRLs emphasize that citizens’ use of non-motivating reasons may yield valuable epistemic and social benefits to public deliberation (e.g. Schwartzman 2012; also think of how citizens’ reliance on devil’s advocacy can strengthen the epistemic merits of public justification). These considerations, in turn, suggest that requiring citizens to abide by PSM would threaten to hamper (rather than promote) public deliberation and that PSM’s requirement that citizens have and are willing to offer adequate secular motivating reasons for the coercive laws and policies they advocate is unjustifiably restrictive by PRL’s own lights.

A proponent of the sincerity defence may object that allowing citizens to advocate coercive laws and policies on the basis of non-motivating reasons without requiring these citizens to publicly acknowledge that the reasons they offer are non-motivating would encourage insincerity and manipulation on the part of such citizens (e.g. Audi 1991, p. 73) and that this ‘lack of candor, should it become publicly known, would undermine trust’ among citizens (Macedo 1990, p. 293, italics added). The idea is that citizens should abide by PSM*, a weakened version of PSM which allows that citizens may legitimately advocate coercive laws and policies on the basis of non-motivating reasons only provided that, when they do so, they publicly acknowledge that the reasons they offer are non-motivating. This objection correctly notes that in many cases, it may be praiseworthy or desirable of citizens to publicly acknowledge whether or not the reasons they offer in public deliberation are motivating. Still, there are at least two reasons to doubt that the objection vindicates PSM* as a plausible liberal principle of public justification. First, in public deliberation, publicly known ‘lack of candor’ about what reasons happen to motivate citizens does not generally ‘undermine trust’ among citizens to a point that hampers public deliberation. For many citizens have (and are believed to have) limited epistemic and evidential access to their own motivating reasons (Section 3). And many citizens face (and are believed to face) powerful incentives (e.g. conflicts of interest, social and political pressures) not to fully disclose their own motivating reasons in public deliberation (e.g. Grant 1997, ch. 2; Kang 2003). As a result, few citizens expect their fellow citizens to reliably disclose whether the justificatory reasons they offer in public deliberation are motivating (e.g. Dowding and Van Hees 2007; Fumagalli 2020b). And second, even focusing on those cases where publicly known ‘lack of candor’ about what reasons happen to motivate citizens undermines trust among citizens, requiring citizens who offer non-motivating reasons to publicly acknowledge that the reasons they offer are non-motivating will likely incentivize citizens who are unable to offer adequate secular motivating reasons to provide inaccurate characterizations of their own motivating reasons. This, together with the difficulties involved in assessing the accuracy of citizens’ characterizations of their own motivating reasons, may lead citizens ‘to discount arguments presented in public discourse’ (Schwartzman 2011, p. 392), thereby hampering (rather than promoting) sincere and non-manipulative public deliberation (e.g. Greenawalt 1995, pp. 163–164; also Section 6).

A proponent of the sincerity defence may grant that a citizen’s advocating coercive laws and policies on the basis of non-motivating reasons neither implies nor encourages insincerity or manipulation on the part of this citizen. Still, she may object that unless one imposes stringent constraints on citizens’ use of religious reasons, coercive laws and policies could be advocated on the basis of ethically and/or epistemically questionable reasons such as religious bigotry and intolerance (e.g. Habermas 2008, ch. 5; Schwartzman 2004; also Audi 1989, p. 293, on some religious citizens’ ‘unwarrantedly strong tendency to judge public policy issues to be irresoluble on secular grounds’). Hence, the objection goes, citizens should abide by PSR, which allows that citizens may legitimately advocate coercive laws and policies on the basis of non-motivating reasons, but requires that citizens refrain from advocating coercive laws and policies unless they have and are willing to offer adequate secular reasons for such laws and policies. This objection correctly notes that citizens’ reliance on ethically and/or epistemically questionable reasons – besides being prima facie objectionable – may hamper public deliberation by hindering citizens’ ability (or willingness) to identify adequate secular justificatory reasons for or against various coercive laws and policies. Even so, the objection does not vindicate all versions of PSR as plausible liberal principles of public justification. In particular, the objection could support at most what I call restricted PSR – i.e. a version of PSR where the expression ‘secular reasons’ is interpreted narrowly to designate only religiously neutral reasons – rather than what I call unrestricted PSR – i.e. a version of PSR where the expression ‘secular reasons’ is interpreted broadly to designate all non-religious reasons. For on the neutrality principles endorsed by most PRLs, ‘the state should neither favor nor disfavor religion (or the religious) as such, that is, give positive or negative preference to institutions or persons [or reasons] simply because they are religious’ (Audi 2000, p. 33; also Barry 1995a; Rawls 1993). And whereas restricted PSR’s symmetrical treatment of religious and anti-religious reasons satisfies PRLs’ neutrality principles, unrestricted PSR’s selective exclusion of religious (as opposed to all non-religious) reasons from the set of public reasons unfairly privileges anti-religious reasons over religious reasons and violates PRLs’ neutrality principles (e.g. D’Agostino 1996; Greenawalt 1995, ch. 5; also Sections 5–7 for discussion).^[8]

5 Respect Defence

The respect defence aims to support PSM as a liberal principle of public justification on the alleged ground that allowing citizens to advocate coercive laws and policies on the basis of non-motivating reasons would leave many of their fellow citizens ‘pacified … rather than respected as a partner in a shared civil life’ (Audi 1989, p. 283, italics added). The idea is that to respect other citizens as free and equal ‘is to address them as fellow deliberators and co-legislators by seeking justifications [they can] reasonably accept’ (Boettcher 2007, p. 231; also Hartley and Watson 2018, p. 7; Larmore 1999, p. 608) and that ‘there is a certain lack of respect implied in seeking … agreement to a policy by offering reasons by which one is not [motivated]’ (Audi 1989, p. 283; also Audi 2000, p. 110). Hence, the defence goes, respect-related considerations vindicate PSM as a plausible liberal principle of public justification.^[9]

These considerations nicely fit leading PRLs’ view that advocating coercive laws and policies on the basis of reasons that the citizens targeted by these laws and policies cannot reasonably accept implies lack of respect for these citizens (e.g. Larmore 2008, p. 143; also Boettcher 2012). Still, it is dubious that respect-related considerations vindicate PSM as a plausible liberal principle of public justification. For the mere fact that a citizen advocates coercive laws and policies on the basis of reasons by which she is not motivated does not per se imply any lack of respect for her fellow citizens. In fact, a citizen may express respect (rather than disrespect) for her fellow citizens by offering non-motivating reasons in public deliberation. In particular, citizens frequently offer non-motivating reasons in public deliberation because they think that offering their own motivating reasons would lead to unnecessary conflict (e.g. Rawls 1999, p. 591, on cases where citizens ‘reason from … other people’s [comprehensive doctrines] to show them that … they can still endorse a reasonable political conception of justice’). To be sure, one may identify various cases where citizens offering non-motivating reasons in public deliberation betray disrespectful (or otherwise questionable) attitudes toward their fellow citizens (e.g. Section 4 on manipulative attitudes). Yet, the mere fact that one offers non-motivating reasons in public deliberation falls short of implying that she disrespects her fellow citizens (e.g. Schwartzman 2011, pp. 376–377; also Shiffrin 1999). That is to say, there is nothing inherently disrespectful in offering to one’s fellow citizens justificatory reasons by which one is not motivated, and one may respect one’s fellow citizens in public deliberation without having to abide by PSM.^[10]

A proponent of the respect defence may grant that advocating coercive laws and policies on the basis of non-motivating reasons does not per se imply any lack of respect for one’s fellow citizens. However, she may object that advocating coercive laws and policies on the basis of religious reasons does imply lack of respect for one’s fellow citizens (e.g. Audi 2011, ch. 3, on various religious citizens’ tendency to impose their own religious views on their fellow citizens). Hence, the objection goes, citizens should abide by PSR, which allows that citizens may legitimately advocate coercive laws and policies on the basis of non-motivating reasons, but requires that citizens refrain from advocating coercive laws and policies unless they have and are willing to offer adequate secular reasons for such laws and policies. This objection correctly notes that on PRL, it is ‘unreasonable to impose political power on others in the name of values that they [can] reasonably reject’ (Cohen 1993, p. 1539). However, there are at least two reasons to doubt that the objection vindicates all versions of PSR as plausible liberal principles of public justification. First, advocating coercive laws and policies on the basis of religious reasons does not generally imply any lack of respect for one’s fellow citizens (e.g. Button 2005; Zerilli 2012, on religiously inspired calls for coercive laws and policies that safeguard liberal values, including respect for others as free and equal citizens in public deliberation). And second, selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons for the sole fact that they are religious would tend to foster disrespect toward religious citizens by leading their fellow citizens to treat religious reasons as generally inferior to (or otherwise less valuable than) non-religious reasons and to subject religious citizens to expressive subordination by publicly regarding them as ‘less worthy than other citizens’ (Nussbaum 2011, p. 22; also Alexander 1993, p. 774; Anderson 2006, p. 151).

A proponent of the respect defence may object that selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons is justified because many citizens may feel disrespected when their fellow citizens advocate coercive laws and policies on the basis of religious reasons and may resent having to comply with such coercive laws and policies (e.g. Audi 2000, p. 165; also Audi 2011, p. 76, holding that ‘citizens may properly resent coercion based essentially on someone else’s religious convictions’). This objection correctly notes that the mere fact that some citizens care deeply about the opportunity to offer religious reasons in public deliberation does not per se indicate that citizens can legitimately advocate coercive laws and policies on the basis of such reasons (e.g. Macedo 2000). Still, there are at least two reasons to doubt that the objection vindicates all versions of PSR as plausible liberal principles of public justification. First, on PRL, the public justifiability of coercive laws and policies does not directly rest on the issue whether advocating such laws and policies causes feelings of disrespect and resentment among the targeted citizens. In particular, if religious (or non-religious) citizens offer adequate justificatory reasons for the coercive laws and policies they advocate, then the mere fact that non-religious (or religious) citizens may feel disrespected and may resent having to comply with these laws and policies falls short of excluding that religious (or non-religious) citizens can publicly justify such laws and policies. And second, even assuming that the public justifiability of coercive laws and policies directly rested on the issue whether advocating such laws and policies causes feelings of disrespect and resentment among the targeted citizens, many citizens may feel disrespected when their fellow citizens advocate coercive laws and policies on the basis of non-religious reasons and may resent having to comply with such coercive laws and policies (e.g. Stout 2004, ch. 3; Vallier 2020b). And it is dubious that, in pluralistic societies, coercive laws and policies advocated on the basis of religious reasons generally cause more feelings of disrespect and resentment than coercive laws and policies advocated on the basis of non-religious reasons (e.g. Billingham 2017a; Swan and Vallier 2012).^[11]

6 Stability Defence

The stability defence aims to support PSM as a liberal principle of public justification on the alleged ground that allowing citizens to advocate coercive laws and policies on the basis of non-motivating reasons would hamper the stability of public justification (e.g. Rawls 2005, pp. 133–172; also Audi 1989, p. 282, holding that ‘if I persuade someone only by adducing secular reasons that do not move me, I tend to produce only a fortuitous and unstable agreement’). The idea is that political agreements based on non-motivating reasons tend to be inherently unstable and that citizens should abide by PSM if they are to achieve a stable public justification where ‘everyone accepts and knows that the others accept the same principles of justice’ (Rawls 1971, p. 454; also Audi 2011, p. 75, claiming that ‘no democracy can be expected to flourish unless [citizens] constrain their own attempts to restrict liberty by adhering to [PSM]’). Hence, the defence goes, stability-related considerations vindicate PSM as a plausible liberal principle of public justification.^[12]

These considerations nicely fit leading PRLs’ emphasis on the stability of public justification (e.g. Hertzberg 2018; Weithman 2009). However, there are at least two reasons to doubt that stability-related considerations vindicate PSM as a plausible liberal principle of public justification. First, the stability of public justification primarily depends on the adequacy of the justificatory reasons that citizens offer for and against the proposed coercive laws and policies rather than on whether such reasons happen to be motivating for these citizens (e.g. Barry 1995b; Thrasher and Vallier 2015). And second, public justification based on non-motivating reasons is not generally less stable than otherwise identical public justification based on motivating reasons (e.g. conflicts between non-motivating reasons are often easier to overcome or alleviate than conflicts between motivating reasons). In fact, even focusing on those cases where public justification based on non-motivating reasons is less stable than otherwise identical public justification based on motivating reasons, citizens may greatly contribute to the stability of public justification by offering justificatory reasons that – while being non-motivating for them – can be reasonably accepted by their fellow citizens (e.g. Grant 1997, ch. 2). And this contribution, in turn, may offset the loss of stability putatively caused by citizens offering non-motivating reasons (e.g. Dowding and Van Hees 2007; Kang 2003).

A proponent of the stability defence may grant that advocating coercive laws and policies on the basis of non-motivating reasons does not generally hamper the stability of public justification. Still, she may object that allowing citizens to advocate coercive laws and policies on the basis of religious reasons severely hampers the stability of public justification. The idea is that citizens who offer religious reasons in public deliberation tend to be ‘intransigent’ (Audi 1989, p. 282) and ‘treat the public square as a battleground rather than as a forum for debate’ (Marshall 1993, p. 859; also Hartley and Watson 2009). Hence, the objection goes, citizens should abide by PSR, which allows that citizens may legitimately advocate coercive laws and policies on the basis of non-motivating reasons, but requires that citizens refrain from advocating coercive laws and policies unless they have and are willing to offer adequate secular reasons for such laws and policies. This objection correctly notes that several citizens who offer religious reasons in public deliberation tend to be ‘intransigent’ and ‘treat the public square as a battleground’ (e.g. Audi 2000, ch. 4; Swaine 2003). Still, there are at least two reasons to doubt that the objection vindicates all versions of PSR as plausible liberal principles of public justification. First, many anti-religious citizens tend to be ‘intransigent’ and ‘treat the public square as a battleground’ (e.g. Galston 1995; Vallier 2016b, on various instances of anti-religious citizens’ opposition to conscientious objection in free and democratic societies). Hence, if the alleged fact that citizens who offer religious reasons in public deliberation tend to be ‘intransigent’ and ‘treat the public square as a battleground’ licenses excluding many religious reasons from the set of public reasons, then many anti-religious reasons should also be excluded from the set of public reasons. And second, religious citizens do not generally tend to be ‘intransigent’ and ‘treat the public square as a battleground’ (e.g. Cooke 2017; Habermas 2006; McConnell 1999). In fact, many religious citizens endorse (and actively promote) liberal values, ranging from freedom of expression to the separation between church and state (e.g. Boettcher 2005; Hertzberg 2019, ch. 1; Weithman 2002, ch. 2, on religious citizens’ contributions to social reform and civil right movements).

A proponent of the stability defence may grant that allowing citizens to advocate coercive laws and policies on the basis of religious reasons does not always hamper the stability of public justification. Still, she may object that public justification generally tends to be less stable when citizens are allowed to advocate coercive laws and policies on the basis of religious reasons (e.g. Hartley and Watson 2020, pp. 891–893; also Audi 2000, p. 103, claiming that precluding citizens from advocating coercive laws and policies on the basis of religious reasons will ‘help prevent religious warfare and civil strife’). However, public justification does not generally tend to be less stable when citizens are allowed to advocate coercive laws and policies on the basis of religious reasons (e.g. Perry 1991; Weithman 2015; also McConnell 2007; Vallier 2020b, on cases where public justification is more stable when citizens are allowed to advocate coercive laws and policies on the basis of some religious reasons). In fact, selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons – as per unrestricted PSR – would often hamper (rather than promote) the stability of public justification. For unrestricted PSR’s asymmetrical treatment of religious and anti-religious reasons would tend to foster polarization and unjustifiably marginalize or alienate many religious citizens who would otherwise support PRL (e.g. Peñalver 2007; Sunstein 2002). Moreover, even focusing on those cases where public justification tends to be less stable when citizens are allowed to advocate coercive laws and policies on the basis of religious reasons, citizens offering religious reasons for such laws and policies may yield valuable epistemic and social benefits to public deliberation (e.g. Shiffrin 1999; Waldron 1993, on the epistemic and social benefits yielded by free and open dialogue between religious and non-religious citizens about the merits of coercive laws and policies). And these benefits, in turn, may offset the loss of stability that is occasionally caused by citizens offering religious reasons in public deliberation (e.g. Billingham 2020; Hollenbach 2003, ch. 6–9).^[13]

7 Fairness Defence

The fairness defence aims to support PSR as a liberal principle of public justification on the alleged ground that allowing citizens to advocate coercive laws and policies on the basis of religious reasons would be unfair to their fellow citizens (e.g. Hartley and Watson 2009, 2018, ch. 4; also Audi 1989, p. 294, claiming that allowing citizens to advocate coercive laws and policies on the basis of religious reasons ‘would give [these reasons] a role not appropriate [to them]’). The idea is that excluding religious reasons from the set of public reasons may be rather demanding for religious citizens, but is not unfairly demanding for religious citizens. For public deliberation ‘is not a process … whose purpose is to give expression to … religious views’, and ‘if no public reasons can be found that are sufficient to support a political claim, then that claim should be abandoned’ (Schwartzman 2011, p. 394; also Macedo 1990, p. 295). Hence, the defence goes, fairness-related considerations vindicate PSR as a plausible liberal principle of public justification.^[14]

These considerations nicely fit leading PRLs’ emphasis on the importance that fairness-related considerations have within PRL (e.g. Lister 2018). However, it is dubious that fairness-related considerations vindicate all versions of PSR as plausible liberal principles of public justification. For on PRL, whether or not the reasons offered in public deliberation are religious does not directly determine whether these reasons can qualify as public (Section 3). This, in turn, challenges the proponents of PSR to provide a detailed and plausible specification of what difference between religious and non-religious reasons would make it justified to selectively exclude the former (but not the latter) from the set of public reasons. Regrettably, the proponents of PSR have hitherto failed to meet this justificatory challenge (Sections 4–6). In this context, selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons would impose unfair justificatory burdens on religious citizens. In particular, it would make it unfairly difficult for religious citizens (as opposed to non-religious citizens) to support coercive laws and policies (e.g. McCarthy 1994; Wolterstorff 1997b). As a result, the arguments of religious citizens would often come across as less convincing than the arguments of non-religious citizens because of factors that do not directly pertain to these arguments’ potential to contribute to public justification (e.g. Eberle 1999; Talisse 2015). This, in turn, would lead to both political unfairness – whereby religious citizens are unfairly discriminated in public deliberation (e.g. Alexander 1993, p. 774; Vallier 2014, pp. 59–64) – and epistemic unfairness – whereby the credibility of religious arguments in public deliberation is reduced because of factors that do not directly pertain to these arguments’ potential to contribute to public justification (e.g. Carter 1994, ch. 11; Peñalver 2007, p. 535).

A proponent of the fairness defence may object that selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons is justified because religious reasons fail to be shareable by or accessible to all reasonable citizens (e.g. Badano and Bonotti 2020, p. 63; Macedo 1990, p. 295). This objection correctly notes that several religious reasons fail to satisfy shareability and accessibility requirements (e.g. Greenawalt 1995, ch. 8; Lister 2011, on sectarian and authoritarian religious reasons). Still, there are at least three reasons to doubt the justifiability of selectively excluding religious (as opposed to non-religious) reasons from the set of public reasons because of their putative non-shareable or non-accessible character. First, the objection presupposes shareability/accessibility accounts of public justification, and so does not per se provide independent reasons to selectively exclude religious (as opposed to non-religious) reasons to those citizens who endorse different accounts of public justification (e.g. Section 1 on intelligibility accounts; also Cohen 1993; Rawls 2005, pp. 35–40, on how PRL’s idea of reasonable pluralism presupposes that citizens frequently have conflicting religious and non-religious reasons for the coercive laws and policies they advocate). Second, even assuming shareability/accessibility accounts of public justification, many non-religious reasons ‘are on a par with religious [reasons] in not being shared by all citizens’ (Quinn 1995, p. 40; also Anderson 2006; Fumagalli 2018; Gaus and Vallier 2009, on conflicts between reasons grounded on utilitarian, Kantian and virtue ethical doctrines). Hence, if the mere fact that many religious reasons fail to satisfy shareability requirements licenses excluding these reasons from the set of public reasons, then many non-religious reasons should also be excluded from such set (e.g. Sandel 1994; Vallier 2014). And third, one may find a number of accessible religious reasons in public deliberation (e.g. McConnell 1999; Pope 2005, on the natural law tradition in Christian theology). And many such reasons are specific enough to contribute to publicly justifying at least some coercive laws and policies (e.g. Cooke 2017; Thunder 2006; also Audi 1993; Billingham 2017b, on the possibility to achieve theo-ethical equilibrium between one’s religious and non-religious reasons for or against coercive laws and policies).

A proponent of the fairness defence may grant that many religious and anti-religious reasons fare similarly with regard to shareability and accessibility requirements. Still, she may object that religiously neutral reasons typically fare better than both religious and anti-religious reasons with regard to such requirements (e.g. Audi 2011, p. 101, claiming that ‘the “do-unto-others” rule [is] central in any plausible secular ethics’). Hence, the objection goes, only religiously neutral reasons should be included in the set of public reasons. The idea is that citizens should abide by restricted PSR, which requires that citizens refrain from advocating coercive laws and policies in public deliberation unless they have and are willing to offer adequate religiously neutral reasons for such laws and policies. Now, restricted PSR is a more plausible liberal principle of public justification than unrestricted PSR since it corrects for the unfairness inherent in unrestricted PSR’s asymmetrical treatment of religious and anti-religious reasons (Sections 4–6), while retaining leading PRLs’ view that many religious reasons should be excluded from the set of public reasons (e.g. Rawls 1993, pp. 62–63 and pp. 150–154). This, however, does not per se imply that restricted PSR is, in general, the most plausible liberal principle of public justification. In fact, there are various reasons to doubt that restricted PSR is, in general, the most plausible liberal principle of public justification.

To illustrate this, consider a more permissive liberal principle of public justification which requires that citizens have, and be willing to offer, adequate justificatory reasons for the coercive laws or policies they advocate, but does not directly exclude any reason from the set of public reasons for the sole fact that this reason is religious, religiously neutral, or anti-religious. There are at least three reasons to doubt that restricted PSR is, in general, a more plausible liberal principle of public justification than this permissive liberal principle. First, on most accounts of public reason (e.g. Section 1 on accessibility and shareability accounts), the set of non-public reasons includes many religiously neutral reasons besides religious and anti-religious reasons (e.g. think of appeals to noumenal selves and transcendental freedom in Kantian theorizing; also March 2013; Rawls 1995, for other examples). Second, citizens offering religious and anti-religious reasons in public deliberation may yield valuable epistemic and social benefits to public deliberation which cannot be yielded by citizens offering religiously neutral reasons (e.g. Hertzberg 2015, 2019, ch. 2, on how citizens offering religious reasons in public deliberation may generate ‘chains of persuasion’ that inform public deliberation; also Sections 4–6 on how citizens offering religious reasons may strengthen the epistemic merits of public justification). And third, in pluralistic societies, public deliberation grounded solely on religiously neutral reasons often fails to yield determinate verdicts about the public justifiability of coercive laws and policies (e.g. Horton 2003; March 2012; Quinn 1995). This does not exclude that, in some situations, citizens may be able to overcome ‘a stand-off between different political conceptions [by voting] according to their complete ordering of political values’ (Rawls 1999, p. 605; also Reidy 2000; Schwartzman 2004; Williams 2000, for discussion). Still, it heavily constrains restricted PSR’s potential to yield determinate verdicts about the public justifiability of coercive laws and policies. That is to say, restricted PSR is a more plausible liberal principle of public justification than unrestricted PSR, but in many cases even restricted PSR is less plausible than a more permissive liberal principle of public justification which does not directly exclude any reason from the set of public reasons for the sole fact that this reason is religious, religiously neutral, or anti-religious.

8 Conclusion

In recent years, leading PRLs have argued that publicly justifying coercive laws and policies requires that citizens offer both adequate secular justificatory reasons and adequate secular motivating reasons for these laws and policies. In this paper, I provided a critical assessment of these two requirements and argued for two main claims concerning such requirements. First, only some qualified versions of the requirement that citizens offer adequate secular justificatory reasons for coercive laws and policies may be justifiably regarded as plausible liberal principles of public justification. And second, the requirement that citizens offer adequate secular motivating reasons for coercive laws and policies is untenable on multiple grounds. PRLs should focus on assessing the justificatory reasons offered for and against coercive laws and policies rather than requiring citizens to offer adequate secular motivating reasons for such laws and policies.