Public Reason and Public Health: Can Anti-smoking Policies Be Justified According to a Public Reason Account of Justification?

Abstract

Public reason demands that policies are justified to all reasonable citizens. Public health aims at protecting or improving aggregated health outcomes. Since health is not an uncontroversial value, an insurmountable chasm between public reason and public health seems to preclude any viable synthesis between the two outlooks. For any given public health policy, some reasonable citizen seems to have a reason to support ‘no policy’ over ‘some policy’, meaning that the policy cannot be justified to all. The paper first spells out what exactly this conflict is about. Then, using smoking as a case, the paper outlines a model of reconciliation between public reason and public health that should give us some optimism if we want to have public health policies that are compatible with treating citizens as free and equal in the public reason sense.

Public Justification is not simply valid reasoning, but arguments addressed to others; it proceeds correctly from premises we accept and think others could reasonably accept to conclusions we think they could also reasonably accept. – Rawls, ‘The Idea of Public Reason Revisited’, p. 786

Introduction

Public reason liberalism (from now: PR) shows how we may reconcile the fact of deep yet reasonable disagreement with equal respect for all reasonable citizens: only policies/rules that are not rejected by any reasonable citizens are justifiable. The aim of public health programmes is to improve or protect population health. A potential conflict is obvious: there is deep disagreement concerning the value of health, tobacco regulation and many other staple issues of public health programmes. Does PR preclude public health interventions, or is some form of synthesis between the two possible? Using smoking regulation as a case, I shall attempt to build a modest case for a positive answer. If the case is successful, I hope to contribute by placing PR into the core of public health ethics research.

The paper proceeds as follows. First, some central PR themes are presented. I will then introduce the case of smoking policies. Then, I discuss the possible conflict between PR and public health. In the central constructive part of the paper, the following is addressed: there are reasons to believe that even quite demanding PR-accounts can be reconciled with some of the goals of standard public health interventions. Various aspects of rules regulating smoking are used to exemplify the resources of PR vis-à-vis public health, and it is shown that some rules regulating smoking can be justified to all reasonable citizens, even if they disagree profoundly about values and rankings concerning health, liberty and similar central issues.

PR and Public Health

PR takes its outset in the fact of (reasonable) pluralism, and asks ‘How is it possible that there may exist over time a stable and just society of free and equal citizens profoundly divided by reasonable though incompatible religious, philosophical, and moral doctrines?’ (Rawls, 1993: xviii). Note that pluralism is not an obstacle to be removed, nor necessarily the upshot of bad reasoning, bias or egotism, but a welcomed result of persons employing their agency and deliberation in fairly free societies (Rawls, 1993: xxivf). Moreover, all recognizable PR-accounts agree that persons, as a starting point at least, are to be seen and treated as free and equal. Hence, one cannot simply, without a justification addressed to another taking her seriously as a free and equal person, impose one’s (purported) moral authority on another free and equal, but disagreeing person (see Gaus, 2011: 14ff; Quong, 2017). In a phrase, free and equal ‘…moral persons are all equally authoritative interpreters of the demands that morality places on one’ (Gaus, 2011: 37).

Most PR accounts of justification do not appeal to the truth of any specific first-order moral doctrine: ‘I propose that in public reason comprehensive doctrines of truth or right be replaced by an idea of the politically reasonable addressed to citizens as citizens’ (Rawls, 1997: 776).

Why this move? Different accounts of PR put forward different rationales, but it seems fair to say that there are two main groups of ideas here, and that many, if not all, of the influential accounts of PR draw from both: epistemic rationales and normative rationales.

Epistemic rationales focus on the fallibility or boundedness of our normative judgements: given ‘the burdens of judgment’ (Rawls, 1993: 54–58)—the various sources of reasonable disagreement, such as conflicting evidence, disagreement about methods of enquiry, values and their relative rankings—we should come to see ourselves as fallible or limited epistemic agents. Rather than insisting on the epistemic superiority of one’s moral beliefs, one should adopt a method of deliberation suited to fallible and ‘bounded’ moral deliberators.¹

Normative rationales focus on our status as free and equal moral agents and citizens: how can we respect each other as free and equal citizens, bound by the same laws and living in the same institutional setting, if those laws and institutions are not justified to each citizen in light of the reasons and values held by, compatible with, or at least not downright rejected by, each citizen? Or, at least, that we justify laws, rules, etc. to each citizen as if that citizen is at least minimally committed to finding mutually acceptable/non-rejectable, reciprocal rules to govern our social life.

Accordingly, at the heart of the PR project lies a commitment to reciprocity.² The deliberators are presumed to have a sincere desire to find mutually acceptable rules and do not engage in purely strategic deliberation. Agents that do not care about ‘justifiability to all’ fail as relevant deliberators. Correspondingly, all PR accounts operate with some level of idealization of the relevant deliberators. Idealization reflects the rationales mentioned above: deliberators may be idealized epistemically, normatively or both. Idealization implies that certain aspects of our belief sets/normative beliefs are non-public, and hence not relevant material for PR.³ On a minimal account, non-public reasons can, in very rough terms, be described as those incompatible with the ideal viewing citizens as free and equal (e.g. racists beliefs), or those that elude any plausible reconstruction as reasons that all reasonable agents could come to grasp (given idealization), e.g. ‘mystical insights’.

Rawls’ account may be termed a strongly idealized version of PR. The ‘deliberator’ assumed is both epistemically and normatively quite specific and ‘far removed’ from any actual citizen (think of the deliberating parties in the Original Position). Strong idealization invites the objection that the very problem that PR sets out to solve—the problem of legitimacy under conditions of reasonable pluralism—is simply whisked away.⁴ As a counterpoint to strong idealization, Gaus’ model is instructive:

Gaus attempted to have a more ‘realistic’ idealization than Rawls, meaning that the deliberators (‘Members of the Public’ in Gaus’ terms) ‘…are not so idealized that their reasoning is inaccessible to their real-world counterparts’.⁵ However, the reasoning of Members is not identical to the real citizens; the Members ‘…hold the beliefs that their real-world counterparts would be justified in holding after engaging in a respectable amount of good reasoning…’ (Gaus, 2011: 250). Moreover, Gaus stresses that even though a justified system of social-moral rules must serve the interests of each person, we cannot expect morality to be costless. Hence, Members cannot simply reject a specific rule because it does not promote their interests.

It may be queried how citizens assumed to disagree profoundly on matters of value may ever come to agree any rule. Solving this problem is part of the raison d'être of idealization: certain forms of epistemic restraints and normative commitments are assumed when we are looking for the kinds of rules persons motivated to find such rules in the light of the values of PR itself; norms governing our political or social life as free and equal citizens. Moreover, we can presuppose that all reasonable agents want some system of rules governing our lives in order to make it possible to harvest the fruits of peaceful social interaction and coordination (see Gaus, 2015: 132).

Hopefully, the reader now has a workable idea of the aspirations of PR. I will briefly comment on some forms of public health aims or areas that I believe can be squared relatively easy with PR. It is plausible to assume that there are valid PRs for supporting consumer choice in the form of information (e.g. labels warning about the health risks associated with smoking) and that certain forms of protection of third parties against harm could be buttressed by PR (see below). Considerably more problematic are broadly paternalist interventions, and policies that involve difficult conflicts between concerns, e.g. conflicts of liberty vs welfare, and the like. These will be the primary focus for the remainder of the paper.⁶

Case: Smoking Policy

To recapitulate: PR builds on the idea that ‘…[e]ach of us is free in the sense of not being naturally subject to any other person’s moral or political authority, and we are equally situated with respect to this freedom from the natural authority of others’ (Quong, 2017).

Is PR compatible with public health interventions? In order to show the prima facie clash between the two, in this section, I will construct a (very) minimal PR-model of justification. Most PR-models I am aware of would be sympathetic to this construction (at least those what follow Gaus) but none I know of would proceed with such minimal premises. Additional premises (or more substantial interpretations of them) will be introduced below in my attempt to reconciliate PR and public health.

An abstract but illuminating outset is to remind ourselves about what problem is PR trying to solve.⁷ PR tries to provide a way out of the following trilemma:

1. When individuals employ their private reason,⁸ they will end up in deep disagreement, including disagreement over moral matters

2. No one has natural moral authority over others

3. We need moral rules in order to harvest the fruits of social cooperation

The challenge is obvious: we disagree over morality, no one has the moral authority to decide what right morality is, yet we need moral rules. Authoritarian attempts to cut through the trilemma are available, but none that engages seriously with (2). Instrumentalist (e.g. Gauthier) or anarchist solutions are on offer (i.e. it is possible to deny (3) by denying that we need ‘real’ moral rules, or rules at all), but none of them seem persuasive, or so I shall maintain. Is a liberal way out—one that accepts all three of the above, yet treats persons as free and equal, possible?

Focusing on the implications of (2) in the above, a minimal (Gaussian) model of justification could be constructed as follows:

• An assumption of moral equality: (competent) adults are assumed, ceteris paribus, at least, to be equal in the capacity and right to interpret the demands of interpersonal morality (see Gaus, 2011: 37)

• (From a): An assumption of blameless liberty: ‘…blameless liberty is the default: a person must have reason to endorse the authority of a moral rule. There is thus an asymmetry between Alf, who claims to have moral authority over Betty, and Betty who denies the rule. Alf’s claim is justified only if Betty has reasons to endorse the authority; Betty’s claim is justified simply if there are no reasons’.⁹

• (From b): An assumption of a veto-procedure as a test for the validity of interpersonal moral claims/rules: For any given proposal of an interpersonally moral rule, such a rule is only interpersonally justifiable if no involved parties (or ‘Members’) prefers ‘no rule’ in the relevant context to the proposed rule (Gaus, 2011: 263ff).

• An assumption of minimal reciprocity: only rules that are sincerely meant to be followed jointly are eligible.¹⁰

Imagine PR-deliberators discussing regulation concerning smoking. Amy is a smoker who does not want to quit. Ben is an anti-smoker; he does not smoke and he does not want other parties, including himself, to be harmed by second-hand smoke. Ben may also want to ban or severely restrict smoking for paternalist reasons. We need not concern ourselves with factual disagreement; both may agree on basic facts about tobacco consumption such as the statistical harm to the consumer, the addictiveness of nicotine and the probability of harmful second-hand smoke.¹¹

Amy and Ben could disagree over a huge number of other issues though, from an outright ban on tobacco to how to handle smoking in specific social situations (should a primary school teacher be allowed to smoke in classroom? In the teachers’ lounge? At school premises outdoors?). If we are to take serious PR’s ambition to respect persons as ‘free and equal interpreters of morality’, and Amy, at first blush at least, finds no restrictions on smoking acceptable, how can we proceed? I will here first focus on whether the state is permitted to interfere with smoking for the good of smokers.

Ben may put forward the factually correct claims that smoking is statistically harmful; that Amy has no way of knowing whether various smoking harms befall her, and hence, it would, health-wise, be in the best interest of Amy to quit smoking, if need be assisted by various state interventions (price regulations, bans on smoking indoors, even in one’s own home, etc.). Ben may also be assumed to adhere to premise (4) in the above: He could very well be sincere in his proposal, even if he was a smoker himself, and accept the restrictions for his own good. However, Amy may reply that she is well aware of the risks; that she prefers the pleasure involved even in the face of the risks, so while Ben’s reasoning may be right (at least probabilistically) vis-à-vis her health, it does not address her as an adult being and her interests more generally speaking. Interventions would just make it harder for her to pursue her preferences. Assume that Amy is reasonable—no serious conceptualization of PR would infer from her preferences here to any sort of unreasonableness—and hence, she should not be forced to submit to paternalistic interventions that second-guess her best interest. Granted, a policy making a preference harder to pursue is not a sufficient reason according to PR to make such a policy illegitimate. However, given Amy sincerely believes that the benefits of smoking outweighs the risks of harms, the paternalistic rationale does not address her as an equal interpreter of morality. Accordingly, Ben, even if he believes there are robust reasons to refrain from smoking, should abstain from imposing moral rules on Amy that she has reasons to reject since we should ‘refrain from making moral demands of other people that they themselves do not have reason to endorse’ (Quong, 2014: 549, referring to Gaus, 2011: 19).

Since the ideal of PR is to forward only rules that are not rejected by any reasonable citizens (cf. premise 3 in the above), then given an assumption that coercion must be justified, PR cannot condone paternalistic public health interventions of the kind proposed here. It would be illegitimate to force restrictions upon Amy, even if they are justified if only probabilistic health outcomes are taken into consideration.

Appeals to ‘the true interest’ of Amy risks being (unjustifiably) paternalistic: Even if can be argued that Amy would never have started smoking but for peer pressure and clever manipulative marketing, Amy—the real Amy as she is—still wants to smoke. In short, paternalistic considerations are out of bound for PR, at least ceteris paribus and when dealing with competent adults.

Protection against harm to third parties is different: Ben could argue that even if Amy should be allowed to smoke, other parties should be protected from potentially harmful smoke. This could form the basis for some restrictive policies. No assumptions about Amy’s health or her ‘true interests’ are involved, so such policies cannot not be ruled out for that reason. In short, Ben’s argument is not prima facie unreasonable and must be taken seriously (see below on autarchy). On the other hand, it seems not unreasonable if Amy protests against restrictions on her smoking that severely limits her opportunities to do so; in short, she can put forward claims that are also compatible with PR—e.g. a concern for her liberty and preferences, and the availability of less restrictive alternatives—that should also be taken into consideration (this point will be elaborated below).¹² What Amy cannot reasonably do is to simply ignore Ben’s claim (i.e. the fact that Amy may not be concerned at all with Ben’s health is not a reasonable claim, or at least, it seems to go against the endeavour to find mutually binding rules for the regulation of our social interactions). However, before more premises are added, we cannot proceed. Ben’s claim is not unreasonable and is not in conflict with any of the premises in the above. Yet, Amy’s denial of any restrictive rule is also not unreasonable.

This highlights the conflict between PR and (at least many forms of) public health rationales: according to PR, we must, as far as possible, take citizens and their preferences as they are. When, for some rule or policy, there is no compromise acceptable to all involved parties available, we must assume that no justifiable rule or policy is available. Amy cannot be deemed ‘unreasonable’ simply because, say, her evaluation of the justifiable weighing of just protection of others vs her own opportunities to smoke is tilted towards the latter. Ambitious PR does not only admit reasonable disagreement about the good; it must admit reasonable disagreements over matters of justice or ‘what we owe to each other’, including disagreement over what sorts of limits to our actions justice requires (see, e.g. Vallier, 2019: 5f).

It seems, then, that only very limited smoking policies could be justified according to PR: information campaigns, and (maybe) certain sorts of protection against harm to third parties, as long as rules do not severely infringe the liberty of smokers (we will return to this important point below).

However, that is not the full story. Before proceeding, though, we need to get to grips with some fundamentals of public health (ethics):

The Classic Concerns of Public Health, and PR

Public Health is undergirded by either of two normative concerns¹³: the promotion and protection of human wellbeing through promotion and protection of population health, and social justice concerns about the health status of underprivileged groups, or, more generally, egalitarian concerns (Faden and Shebaya, 2015).

Public health ethics can seem to be dominated by the first-order normative analyses of either particular interventions or attempts to give a broader justification of particular principles meant to govern such interventions. In that sense, public health may of course be buttressed by a number of concerns as they are expressed in ethics proper, from utilitarianism to virtue ethics.¹⁴ PR does not claim that we should not base public health interventions on values. It does, however, claim that a justifiable rule based on these values cannot be rejected by any reasonable citizen. This does not (necessarily) mean that all reasonable citizens must agree on the underlying values. Reasonable citizens can converge on specific rules—i.e. endorse a given rule—even if they do not endorse the same underlying values (or share a particular interpretation of the values (see Vallier, 2011)).

To take a toy example, basing a particular intervention exclusively on a specific normative theory, say, act-utilitarianism, is moralizing and authoritarian vis-à-vis those citizens who do not share a utilitarian outlook. Moralizing denies equal standing—denies that some person is an equal interpreter of the demands of morality.

Now, few public health ethicists would endorse strict enforcement of a specific normative theory upon unwilling citizens. Rather, public health ethicists weigh concerns and principles from different ethical outlooks when they bring forward analyses and/or propose policies. From a PR perspective, the problem is, or may be, that we cannot assume that all reasonable citizens agree to the weighing, e.g. between concerns for equality, general aggregated welfare and liberty (see below).

Moreover, there is in the PR literature considerable scepticism about attempts to justify rules or principles by the way of appealing to conceptions of the good (or ‘the good life’).¹⁵ What exactly counts as a conception of the good is subject to controversy,¹⁶ but Sher gives us the following: ‘In general… such conceptions are taken to include religious doctrines, ideals of character or virtue, aesthetic and cultural values, and norms of sexual behaviour…’ (Sher, 1997: 38).

To cut a long story short, direct appeals to conceptions of the good are out of bounds—ceteris paribus—for PR. Epistemically, it may be argued that such conceptions tend to be ‘opaque’ or ‘inaccessible’ (Bird, 1996: 67ff). Normatively, the PR claim is that a conception of the good cannot form the basis for agreement because (or insofar) such a conception is not shared by all. This reflects traditional liberal scepticism about the legitimacy of building policies on, e.g. specific religious or communitarian ideals (for a comprehensive discussion of this, see Mulhall and Swift, 1996).

It may seem more plausible that separate values could form the basis of justification. After all, who would deny that, e.g. liberty, or wellbeing, is valuable? However, disagreement about conceptions of the good also concern rankings of values. This is shown in the seminal work on values by Rokeach. Gaus comments: ‘According to Milton Rokeach, Americans largely agree in affirming a set of thirty-sex values; what they differ on is “the way they organize them to form value hierarchies or priorities”’.¹⁷ A paternalistically inclined consequentialist and a liberty-loving motorcyclist may agree that health is a value but disagree about the relative position of health in a ranking of values. Of course, citizens need not agree on a complete ranking of values; what is needed is that no reasonable citizen’s weighing of values is completely overridden by some other citizen’s weighing. The consequentialist may emphasize health while the motorcyclist ranks liberty over health; so some proposals of policies that may be endorsed by the consequentialist may not be endorsed by the motorcyclist. Simply assuming the priority of health over liberty is, for a given policy proposal, incompatible with PR.

Consider again smoking policy. Could a restrictive smoking policy be supported by PR? For every policy—I follow Gaus, 2011 here—the ‘Members of the Public’ has, in nutshell, to choose between Policy × (where × is some concrete suggestion) and ‘no policy at all in this matter’ (this reflects the veto-procedure mentioned in the above). The aim is not to find the uniquely justified policy, but whether there is some set of policies that all prefer over no policy at all. This set would then be the eligible set of policies—policies that could be justified, even if there is no agreement among Members as to the ranking of these policies. (This is confusing to some ethicists who believe that the ‘point’ of ethics is to find the best solution. PR, at least in the Gaussian variant, aims rather to find solutions to social problems that all can live with.)¹⁸

As concerns smoking policy, it is prima facie likely that some Members prefer ‘no policy’ to any policy suggestion. Almost any conceivable policy may be said to interfere negatively somehow in the sphere of the smoking Member, and seems therefore rejectable, at least given the premises so far. Hence, the eligible set of smoking policy is empty—or so it may seem. Having no smoking policy seems the only legitimate ‘policy’.¹⁹ If so, we are undeniably at an impasse. This is unsatisfying indeed, because Ben’s claim that his health is also a relevant concern—some protections against second-hand smoke should be justifiable, especially if one stresses (as I believe one should) the spirit of reciprocity underlying the PR-project.

In the light of this, the question arises: is there more conceptual room to reconcile public health and PR? In the following, I shall attempt to show that there is.

Public Health and PR: An Attempt of Reconciliation

I now attempt to outline the resources within PR thinking that may be mustered in order to accommodate some of the aspirations of public health—interventions and—thinking.

Let me briefly elaborate on why I choose the (presumably) less well-known PR framework provided/inspired by Gaus rather than Rawls. In certain regards, Gaus’ PR is more ambitious than Rawls’. Gaus attempted to address citizens ‘as they are’ as much as possible. Recall that the Members ‘…are not so idealized that their reasoning is inaccessible to their real-world counterparts’ (Gaus, 2011: 276). Moreover, controversies over justice are on a par with controversies about the good; we cannot assume as uncontroversial a substantial core of the issues that sets citizens apart. Gaus’ project engages with the real depth and width of diversity in contemporary liberal societies. Note that I am not taking a stand here on whether Gaus’ project is ‘better’ than Rawls’. If we can find convincing reasons to accept a reconciliation between Gaus’ account of PR and certain public health interventions, then we can assume that this is true for PR accounts that are less demanding (i.e. less inclusive of our disagreements).

I believe that the conjunction of some concerns that, according to Gaus, should be taken into account in the construction of The Members can give us some leverage in the attempt to reconcile public health and PR. These are autarchy; reciprocity and a willingness to compromise; and role reversal/reversibility.

Autarchy

Gaus defended a conception of liberty and agency inspired by Stanley Benn. One key term is autarchy, which can be defined as the normal condition of being able to choose at all; those who lack autarchy to some extent are to that extent less than fully persons. Autarchy should not be confused with autonomy, the latter being a much thicker and demanding notion, and hence an obvious subject of reasonable disagreement. Whereas autonomy can be rejected as justificatory material, some concern for autarchy could not. Addressing us as free and equal persons implies basic respect for us as intelligible, though far from perfect, choosers.

Here, I diverge from, or at least expand somewhat on, Gaus’ account and use of autarchy: minimal, but sufficient respect for persons as choosers involve a minimum of respect for their bodily integrity, and also proper concern for the common good,²⁰ including, among other things, a concern for person’s having ‘control over basic resources needed to live a life’.²¹

From this, it can be deduced that basic safety is one concern for the Members. Of course, different Members will value safety differently, but some sort of concern for basic safety in areas that all of us will be involved with—say, workplace and food safety—seems to me to follow seamlessly from Gaus’ account. Some of the goods protected here are as close to ‘real’ public goods as it gets. Clean drinking water, functioning sewage systems… all the things that pertain to the ‘sanitary revolution’ are genuine public goods. As these are public health goods, there seems to be genuine opportunity to make public health and PR compatible on these basic matters, at least when policies are moderate.

Autarchy, then, is one component in the conjunction. Minimal respect for other citizens entails a respect for their agency, which entails a modicum of concern for their bodily integrity, including their health.

Reciprocity and Willingness to Compromise

In the presentation of PR in the above, it was stressed that reciprocity and a willingness to compromise is assumed of the deliberators in PR. This is built into both ‘reasonability’ and the ‘idealization’ of agents, and can be seen in many corners of the PR project: Rawls’ concept of civility, the idea of reversibility (see below) and the very idea that we come together as free and equal persons in order to find common binding rules that respect us as free and equal persons. All this relies on reciprocity and some amount of willingness to compromise. What does this entail?

Let’s revisit smoking policy. Combining the concern for autarchy and a willingness to compromise, one may arrive at the following: Amy is an avid smoker. Ben is a stern anti-smoker. Amy’s preferred policy may be ‘no policy’, while Ben’s may be an all-out prohibition. If prohibition and ‘no policy’ were the only alternatives, then Amy reasonably could veto Ben; she is not addressed as a free and equal citizen by prohibition. However, imagine that among the set of proposed policies are various compromises that seek both to protect non-smokers and give smokers opportunities to continue smoking. Amy may admit that smoking is probabilistically bad for health, that tobacco is addictive to most people, and so on. Still she wants to smoke. But given weak idealization in the light of reciprocity and a willingness to compromise, Amy may plausibly be construed to accept as eligible moderate policies that (i) still gave her access to tobacco, (ii) restricted smoking to, e.g. private homes and dedicated areas and bars/restaurants that accepted smoking and (iii) information campaigns so that new/potential smokers were as informed as is possible (see the discussion below).²² Ben, also being reasonable, may come to realize that (i) some people rank smoking over the probabilistic dangers of doing so, (ii) people derive pleasure from smoking, and so on. Ben, again in the light of weak idealization in the form of reciprocity/compromise, may come to accept the sorts of policies that Amy accepts, even if his privately preferred policy is all-out prohibition.²³ So, the eligible set is not empty: some smoking-related policies may be justifiable. Again, there seems to be some room for conciliation between public health and PR.

This also holds true for the ‘harm to third parties’-conundrum mentioned in the above: Amy should recognize that some persons not unreasonably want to be protected against second hand smoke, and accept some limitations (e.g. designated smoking areas). A plausible construal of Amy as a Member should incorporate both the concern for autarchy (i.e. a recognition of some concern for the health of others) and a willingness to compromise. However, it would against the spirit of compromise if these limitations are too restrictive. We allow for all sorts of probabilistic dangers to other parties (e.g. transportation) because severe restrictions on the activities citizens actually want to engage in compromise their liberty.

No easy formula to weighing up concerns for protection of third parties against concerns for liberty (and/or for aggregated welfare) is forthcoming. But even those who believe we should emphasize protection when such weighing is to be done must admit that it is at least not straightforwardly unreasonable that we allow cars, air travel, power tools, alcohol and very many other things even if they involve potential for harm to third parties. Of course, it is also not unreasonable when we set limits to our exercise of liberty in relation to these products and activities: we do not allow driving under the influence; there are safety-related standards for vehicles that must be met, and so on. On many issues, we are able to find ways of protecting third parties that do not severely restrict the activity involved. Why should it be different for smoking?

Where, then, to draw the exact limit on acceptable restrictions on where smoking is allowed? I don’t think that PR in the abstract alone can answer that question—actual political communities must be engaged—but it may be assumed that Amy will find some restrictions acceptable, while Ben do not have any reasons not to prefer those restrictions (over no restrictions at all) even if they do not go as far as he would prefer. The eligible set need not be empty.

Note, however, that reciprocity does not entail that paternalistic policies could be accepted. Amy can reasonably reject all paternalistic policies, as they do not address her as a competent chooser. In this sense, paternalism entails a rejection of PR, at least when paternalism is directed at reasonable citizens (see Quong, 2011: 100ff). I would add here that I believe that PR is somewhat more accommodating of ‘soft paternalism’ as concerns children. (I understand here soft paternalism as paternalism directed at less-than-fully competent agents.) Surely, reasonable citizens could converge on some policies aiming to protect and inform children.

Role Reversal/Reversibility

Gaus proposed a requirement on eligible proposals that models some features of standard universalizability, namely ‘universability as reversibility’ (Gaus, 2011: 299ff). Central here is the following: ‘Genuine moral proposals must be universalizable in this crucial sense: a person’s advocacy must not depend on her knowledge that she will only occupy specific roles or positions’ (Gaus, 2011: 300). This notion is perhaps somewhat underdeveloped, but here is my take: PR is not instrumentalism: it is not a watered down version of, say, Gauthier’s contractarianism. Members are not out to find rules that maximizes their utility, nor do they seek to avoid any sort of sacrifice.²⁴ Reversibility secures sincere reciprocity: Amy would walk a mile in Ben’s shoes, and vice versa. Hence, proposals that are ‘rigged’ in the sense that one would only support them given one’s actual or foreseeable role or position in the social world, are not eligible proposals.²⁵

Reversibility helps us deal with some more complicated choice situations. Let’s look at smoking policy again, this time introducing Carla, the tobacco producer. On the face of it, Carla may see any sort of restriction on smoking as a threat to her good, and Gaus is adamant that we have no reason to internalize a rule that ‘…is the enemy of one’s good’ (Gaus, 2011: 302). Hence, it may seem that Carla has a reason to rank ‘no policy’ over any smoking policy suggestion. However, it also seems that Carla respects neither reversibility nor reciprocity if she simply blocks any smoking policy. Policies that has little or no impact on existing customers do not ‘threaten her good’ here and now, and while policies that lead to a gradual reduction of smokers may be taken to reduce profitability, it seems only reasonable for Carla to protest to this long term reduction if she knows she will occupy a specific role (tobacco producer) and if we had some sort of right to perpetual profitability of our current business endeavours, surely an implausible suggestion. It may be protested that if Carla’s future good is threatened (long-term profitability), then she has reasonable ground for rejection. I admit this is tricky, but there is, I think, a case to be made against the idea that reasonable citizens should have their specific current good protected in perpetuity, as it were. In the 1980s, a new technology, desk top publishing, gradually reduced the demand for professional typesetters. While regrettable seen from the point of view of the individual typesetter, and while such large-scale changes may call for various sorts of compensation, offers of re-training etc., the fact that ‘one’s good’ is not profitable in perpetuity does not in itself seem to delegitimize the introduction of new technology and changes in demand and supply ratios. Hence, Clara do not have decisive reasons to reject all smoking policies. Reversibility help us towards some reconciliation as well.

For those attracted to a very strict interpretation of PR, my choice of case may seem particularly vulnerable to the objection that Gaus himself seems to deny the possibility of having legitimate rules restricting smoking (see Gaus, 2011: 536f). What Gaus argued is roughly as follows: we may posit that there is some sort of agreement on the disvalue of smoking (the risk imposed on our health). However, we cannot infer from this agreement on (dis-)value to the agreement on policies, because while we may agree that there is badness involved in smoking, we do not agree on the weighing against its positive value (the pleasure derived from smoking). Gaus concluded: ‘Thus, although the badness of ill health caused by smoking can…help justify a law, that its badness outweighs the goodness of the pleasure of smoking cannot; and without that, no state policies discouraging smoking will be justified’ (Gaus, 2011: 537).

I believe that either Gaus was incautious in his wording, or that he was overstating his point. Showing that the badness of an activity outweighs its goodness is not the only way to show that a policy of regulation is legitimate. Other points of interpretation arise here, such as whether the line of reasoning just presented only pertains to policies that ‘discourages’ smoking, e.g. anti-smoking campaigns, and whether all restrictions of smoking (e.g. designated smoking areas) should count a ‘discouraging’ in any unpalatable way. If he should be read in that way, it is an overstatement, because surely one could appeal to other considerations apart from the overall badness of smoking, such as fairness, non-interference (vis-à-vis some non-smokers preference for not being close to tobacco fumes), and to the fact that some variants of non-intrusive and non-punitive policies such as designated smoking areas may be near-pareto improvements (making some non-smokers better off while being, at most, a trivial nuisance for smokers),

Naturally, this leaves many details left to be unpacked, not all of which can be addressed. As hopefully illustrative issues, let us focus on smoking in private and information campaigns²⁶ in order to illuminate some of the most pertinent details.

Smoking in Private

At first blush, one likely candidate for a rule that Members could converge on is letting people decide for themselves whether or not they smoke in their homes. This raises the issue of smoking in apartments and similar housings, where it may be argued that second-hand smoking is ‘intrusive’. How is a reasonable compromise to be found? In short, I think regulation should allow for, but not demand, that some apartment complexes are reserved for non-smokers while remaining silent on smoking policies (in private) for others. The market would then regulate via supply and demand. This raises the following issue:

What about those who want to, but cannot afford (or cannot find) an apartment with rules that fit their preferred smoking policy? Presumably, the worry is that some low-income citizens who want to live in a non-smoking condo or apartment cannot afford this. However, note that the opposite is also the case: low-income citizens who want to be able to smoke. Gaus proposed that we have a weak duty of assistance to support citizens’ autarchy or agency, but it is doubtful that autarchy is by definition undermined if you have to, say live in an apartment complex where smoking is allowed for some time until you can find another place to live,²⁷ and it is equally doubtful that your agency is undermined if you for some time has to live in an apartment complex where you cannot smoke indoors. If living in either a complex where smoking is allowed or where it is forbidden ranks very highly in one’s preferences, then a market regulated system gives one adequate possibilities to satisfy that preference, at least under normal circumstances.²⁸

A related worry concerns children in homes where adults smoke. Children—not being full Members—raise special considerations that are not well addressed in PR literature as far as I know. The question is whether the state has any standing to interfere in families where the adults smoke in order to protect the children. The answer may seem obvious: if the state legitimately can protect children (who cannot adequately choose for themselves) from the harmful effects of second-hand smoke in general,²⁹ then obviously it can legitimately protect children against the harmful effects of their parents’ second hand smoke. However, PR builds on the idea that we are equal interpreters of morality; and it would be absurd to say that all parents who smoke in the vicinity of their children are doing something morally wrong seen by their own lights. The smoking parent may weigh up the probabilistic dangers to the child against the pleasure derived from smoking and end up preferring to smoke and be near their child³⁰; hence, such a parent may reasonably reject a policy outlawing smoking in the vicinity of their own children. However, I think there are reasons to expect convergence on other, less draconic rules that are relevant here. Health personnel should have a right to inform the parents about the dangers of second-hand smoke, for instance. Some policies with indirect consequences may also be justifiable, such as provision for new buildings that there are well-ventilated spaces for smokers, or similar indirect protections.

Information Campaigns and Labelling

At first blush, it seems easy to accommodate various forms of information campaigns informing the public about the possible health implications of smoking, including labelling tobacco products. The dissemination of information is necessary for qualified consumer choice, so in itself it is hard to see why any reasonable Member could voice a relevant complaint against such measures. However, one should move carefully here concerning the kind or way of information.³¹

Information can be given in manipulative or intrusive forms that some Members may reject. Interrupting a TV-transmission of an important sports event to tell viewers about the dangers of smoking seems unjustifiably intrusive; telling or implying that smokers are ‘losers’ seems unjustifiably manipulative. This leaves a range of other forms of information that are more serious eligible candidates. One issue here is the use of ‘scary’ graphic images of smoker’s lungs, etc. Are they unjustifiably manipulative? It is beyond the scope of this paper to develop a complete theory of manipulation, but some thoughts: it would be easy to reject the use of such images if they did not contain a grain of truth. The reason why they pose an interesting question is that it may be manipulative because they overstate the risks of smoking. After all, cars do not come wrapped in posters depicting victims of car accidents, even if car accidents are bound to happen. However, a stronger PR case for graphic images may be built on the fact that tobacco is a different product from most because it is addictive. If you develop a severe shellfish allergy, it is rather easy to stop that consumption, whereas it is much harder to give up tobacco. This may imply that it is harder to reasonably reject warnings about some possible implications of tobacco consumption using graphic images to make sure a consumer accepts the risks: when going down that road (tobacco consumption) be sure to take into consideration the possible adverse effects. While I believe more thought, and data, such as the actual psychological effects of such images, should be put into this, for as long as consumers are not precluded from buying tobacco products, I see no principled or strong presumption against using explicit graphic warnings on tobacco products; it may be one of the rules that Amy, in the spirit of compromise, could accept. The same goes, perhaps somewhat less easily, for Carla. Carla could accept a policy where graphic warnings are mandatory if she can then continue producing and selling tobacco, at least given the proviso that the rules are the same for anyone on the marketplace.³²

A related issue is the flip side of labelling and graphic warnings is what PR implies as concerns tobacco advertizing. Those familiar with the debate about commercial free speech knows that it is a tangled matter. In a PR perspective, the question becomes whether some Members may reasonably reject various sorts of regulations of advertizing for tobacco products. One mode of justification (for regulation) can be ruled out from the outset, the paternalistic one, though it applies to protection of children (no tobacco advertisement on Nickelodeon). However, other forms of regulation aimed at adults that Carla may reasonably be held to not reject would seem to be derivations of protection of third parties, and that again seems to be rather contrived: after all, we do not regulate car commercials or air travel in order to protect third parties from harm.

However, Carla may also recognize that other parties want to protect their children (or perhaps themselves³³) against becoming smokers. Since Carla can also reasonably be assumed to admit that tobacco is a special product due to its potential for risks and addiction, and since we can posit that Carla has some interest in justifying her position in the light of reversibility, it may be the case that we could assume Carla to accept at least some forms of further regulation of advertisement. For instance, one could appeal to a principle discerning between social arenas that are highly visible to all (including children) and hard to avoid (e.g. important public spaces, sports events…), and those that are less visible and easier to avoid (e.g. newspapers, bars, magazines, websites for adults, etc.) and restrict tobacco advertizing to the latter. This does seem to me to be an unreasonable compromise in the light of the values of PR.

In conclusion, I believe that there are solid reasons for being optimistic about some common ground between Amy, Ben and Carla as concern certain smoking policies. Hard paternalist policies are out of bounds. But Amy, being reasonable, may be assumed (i) to have at least some concern for the health of others based on the idea of autarchy (legitimizing some forms of protections of non-smokers and those who cannot consent, e.g. children), (ii) to be willing to restrict her smoking to designated areas, and maybe also (iii) accept a non-punitive regime of taxation raised to inform consumers about the harmful effects of smoking. Ben may be assumed to be willing to compromise as concerns his anti-smoking sentiments given Amy’s accept of certain restrictions. The most complicated case is probably Carla; but if it is reasonable to say that we do not have a right to ‘perpetual profitability’, then her case for rejecting all restrictions of smoking is severely weakened.

Conclusion

At first glance, the projects of public health and PR collide. PR demands that policies are justified to citizens in the light of their own values, and we disagree deeply about the role and relative values of health, liberty, etc. Nevertheless, given some plausible level of idealization, and armed with the assumptions respect for autarchy; reciprocity and a moderate willingness to compromise; and reversibility, I have provided an outline for a reconciliation attempt.

Note that I intend all of these assumptions to work in tandem; it is not as if the concern for autarchy, or the willingness to compromise, is intended to be sufficient by themselves to show that some rule or policy relevant for public health is justifiable. It is when these are taken as joint points of relevance for the construction of PR(-ing) that we can arrive at the conclusions drawn. For instance, one citizen may reason that smoking should not be inferred with at all because non-smokers could avoid the dangers of smoking by withdrawing themselves totally from smokers; however, that seems to disregard both a willingness to compromise and reversibility. In short, I believe the case for reconciliation is most plausible when all of the components are taken into consideration.

So, PR is compatible with some public health interventions, including regulation of smoking if they are (i) informational and in non-manipulative ways supports consumer choice, or (ii) the rationale is protection of third parties, and the intervention does not severely restrict the opportunity for smokers to smoke. I cannot see how hard paternalistic interventions could be supported by PR.

If PR is compatible with public health interventions, we have gained an important insight: it is possible to pursue at least some important public health goals in ways that are justified to all reasonable citizens, even if we disagree deeply on issues of morality and value. We can have non-authoritarian public health policies. Hopefully, others will join in exploring PR as a resource for public health.

Conflict of Interest

None declared.

Endnotes

Acknowledgements

The author would like to thank participants at the Copenhagen University Practical Philosophy Research Group seminars and the participants at the Nordic Network for Bioethics workshop in Aarhus, Denmark, 2019, for their comments and suggestions. I would also like to thank two anonymous reviewers for their detailed and insightful comments.

References