Is Anti-Paternalism Enough?

Abstract

The article addresses the question of how sufficiency for the imprudent may be ensured. Imprudent conduct includes both spectacular acts such as jumping from heights into water, and everyday acts and omissions such as neglecting to fasten one’s seat-belt. We argue that to avoid thoroughly bad or insufficient situations, one must pay attention to an important and often neglected concern of avoiding insouciance. The latter tells us not to stand idly by when people are about to act in ways that may irrevocably land them in insufficient situations. The policies and actions needed to avoid insouciance are most plausibly justified on paternalistic grounds; in fact, they cannot be justified nonpaternalistically. Although controversial, we argue that paternalistic interventions are often justifiable in the service of preserving sufficiency.

Keywords

sufficiency insouciance paternalism harshness

Introduction

Sometimes we act imprudently. For example, we smoke, we jump from heights into water, or we use sunbeds (Andersen and Midtgaard, 2016; Grill and Voigt, 2016; Snelling, 2014). Although such acts are typically readily avoidable to us, it is still regrettable if the risk involved materializes, leaving us in a thoroughly bad or insufficient situation. A decent society is in part characterized by having institutions and policies in place that cater to sufficiency. But how can sufficiency for the imprudent be secured? There is wide agreement that sufficiency in some sense should be ensured even for the imprudent.1 Yet scrutinizing the “how” question, we will argue, gives rise to interesting and somewhat surprising conclusions. We argue specifically that elementary aspects of sufficiency can only be ensured by the adoption of paternalistic measures. When thinking about concerns of sufficiency, people have tended to be preoccupied with the importance of avoiding harshness. This involves being prepared to offer assistance to people when they stand in need of help in emergencies or when they experience serious disadvantages pertaining to their past imprudent conduct. We are not out to deny the importance of avoiding harshness given a concern with sufficiency. Our main argumentative aim in this article is to defend the claim that often this is not enough. One needs to pay attention to a further important concern, namely the concern to avoid what we shall call insouciance. That sufficiency can only be achieved by introducing this further concern—a concern that brings along controversial paternalistic elements—is the primary novel claim the article makes. Insouciance, in brief, represents the inaction of others with respect to stages prior to a person ending up in a thoroughly bad situation. Avoiding insouciance is sometimes a prerequisite of avoiding situations of insufficiency because, in some cases, we are unable to restore people to a reasonable sufficiency threshold if we let them proceed without restrictions. If we do so, they may, like Humpty Dumpty, experience harms of such a nature that “all the king’s horses and all the king’s men” cannot put them together again.2 A central line in our argument is that the reasons supporting the avoidance of harshness are at the same time reasons for avoiding insouciance.

The importance of avoiding insouciance and the paternalistic considerations this concern brings along have in various ways been neglected by key traditions in the literature. First, relational egalitarians such as Elizabeth Anderson, while not dismissive of certain minimalistic paternalistic considerations, have still failed to see this concern as one that is intrinsically related to the aim of achieving sufficiency (Anderson, 1999, 2010); Second, anti-paternalists such as Joel Feinberg have, of course, been in outright opposition to the kind of paternalistic considerations integral to the avoidance of insouciance. They have tried instead to justify policies similar to those involved in the avoidance of insouciance on nonpaternalistic grounds. While ingenious, their proposed justifications have arguably been unsuccessful (Bou-Habib, 2006; Feinberg, 1986; Jones, 1985; Midtgaard, 2015). Third, proponents of so-called “sufficiency-constrained luck egalitarianism” (Casal, 2007: 322) embrace a choice-thesis according to which we can achieve sufficiency either by not abandoning people (i.e. avoiding harshness) or by adopting the kind of paternalistic restrictions involved in catering to the avoidance of insouciance (Casal, 2007; Williams, 2006). If we are right, though, achieving sufficiency requires both. Fourth, some libertarians concerned with sufficiency have accepted certain paternalistic considerations (Bou-Habib, 2006: 250). They have, however, taken a too limited view of the kind of measures that may be involved. They have argued, for example, for schemes of compulsory insurance. But this does not ensure sufficiency. It only ensures that sufficient means for providing assistance are available (means provided by the imprudent individuals themselves).3 Alternatively, they have mistakenly (as we shall show below) argued that relevant paternalistic considerations are unable to justify a number of sufficiency-preserving measures (Flanigan, 2016).

Our argument for the importance of the concern to avoid insouciance and for the paternalistic considerations it brings along is structured in the following way. First, we account for the way in which we understand paternalism. Second, we present our argument that avoiding insouciance alongside the avoidance of harshness is crucial to ensuring a threshold of sufficiency, and that the avoidance of insouciance brings to the fore paternalistic considerations. Third, we consider three key challenges to our argument. Fourth, we address the content and value of ensuring sufficiency and examine the relation between harshness, insouciance, and insufficiency. Fifth, we argue that imposing restrictions on some risky and dangerous activities, which is part and parcel of the campaign to avoid insouciance, cannot be justified nonpaternalistically. Finally, we conclude.

Paternalism

In this article, we adopt a conception of hard paternalism that is focused on reasons. That is, it takes paternalism to involve essentially the affirmation of a certain kind of reason for interfering with individual liberty. The latter should here be understood broadly so as to include concerns of autonomy (Hausman and Welch, 2010). The conception of paternalism, we have in mind, is Feinberg’s well-known formulation of “the paternalistic principle” according to which:

It is always a good and relevant (though not necessarily decisive) reason in support of a criminal prohibition that it will prevent harm (physical, psychological, or economic) to the actor himself (Feinberg, 1986: 4).

For this to reflect the aforementioned broader concerns of autonomy, the term “prohibition” in the cited formulation of the principle should be followed by something such as the following: “or any other mean apart from rational persuasion” (Hausman and Welch, 2010: 129; Scoccia, 2008). Such means affect negatively people’s “control over their own evaluation and deliberation” (Hausman and Welch, 2010: 128), their “independence,” or crucial aspects of their autonomy (Dworkin, 1988; Raz, 1986). The reason pointed out by the paternalist principle is taken to obtain independently of the fact that the harm in question flows from the “fully [or ‘sufficiently’ (Feinberg, 1986: 98–142, cf. Arneson, 2005: 265–269)] voluntary choices and undertakings” of “competent adults” (Feinberg, 1986: 12, cf. Voorhoeve, 2008). While we have a reason to respect people’s voluntary choices, we have, according to the paternalist principle, a countervailing reason of the indicated kind to cater to their well-being.

Anti-paternalism, by contrast, consists of the denial that the reason for preventing self-regarding harm is ever relevant.4 Paternalistic reasons, as Feinberg emphasizes, “never have any weight on the scales at all” (Feinberg, 1986: 26). Insofar as the harm in question pertains to choices or undertakings that are fully or sufficiently voluntary, the will of the agent is authoritative. This means that the reason to respect the agent’s will silences or excludes whatever reason we may have to act so as to benefit this person or to cater to his or her good. Daniel Groll speaks in this regard about treating the will of another in self-regarding matters as structurally decisive. When we, by contrast, treat a person’s will as merely substantively decisive, we let the will of the person play a role in deciding what to do, but alongside this reason we always or sometimes allow considerations for his good to weigh in on the scales. In brief, according to the view on the table, we are paternalists. This is, of course, a fortiori true if we think that paternalistic reasons may in fact sometimes outweigh reasons to respect the will of a person in self-regarding matters. These are the conceptions of anti-paternalism and paternalism we rely on in the article.

Opponents of paternalism are keen to avoid improperly overriding the autonomous choices of agents. Many agree that paternalistically justified policies and actions express a problematic attitude of disrespect toward the people interfered with (see e.g. Anderson, 1999: 301–302; Begon, 2016: 364; Flanigan, 2016; Shiffrin, 2000: 207). For the same reasons, avoiding paternalism is often seen, at least, as an important desideratum. Meeting this desideratum, however, we argue below, has implications that are unappealing. Specifically, it would involve intolerable situations of insufficiency.

Avoiding Harshness and Insouciance

We intend to explain the importance of the arguably unrecognized duty to avoid insouciance against the background of an important and much discussed exchange between so-called luck egalitarians and so-called relational egalitarians. We believe that this context is appropriate to bring forth the concern we would like to highlight. It should be clear, though, as indicated in the introduction, that the concern itself has import also for certain libertarian views, for example.

Luck egalitarians believe that “[i]t is in itself bad with regard to inequality if [and, perhaps, only if] some people are worse off than others through no responsibility of their own” (Lippert-Rasmussen, 1999: 478). On at least some plausible ways of construing this view, it is compatible with holding that people do not have a claim to be bailed out of certain thoroughly bad or insufficient situations. This is true if their predicaments are ascribable to their choices or to the way in which they have exercised their responsibility. Relational egalitarians hold that this implication renders the luck egalitarian view an unacceptable view on justice. It is incompatible, they argue, with the most important abstract desideratum that egalitarian views should satisfy, namely that people must be treated with equal concern and respect (Anderson, 1999; Dworkin, 2000). There are some things that we cannot do to people even with their consent, if we are to relate to them as free and equal (Anderson, 1999: 319, 329). This includes not offering them assistance in emergencies. To refrain from doing so would be unduly harsh (Voigt, 2007). What the harshness objection objects to is, as pointed out by Zofia Stemplowska, “not that people can find themselves in thoroughly bad situations but that the fact of responsibility can be used to justify leaving people in such situations” (Stemplowska, 2009: 251).

Much ink has been spilled on this debate. What is important in the context of this article is to note that whether or not luck egalitarians can respond to the harshness criticism in a satisfactory way, there is a further important concern which neither luck egalitarians nor their critics pay special attention to. And this is especially unfortunate in light of what is arguably the important underlying consideration driving the debate. The founding concern is this. It is regrettable if people end up in situations of insufficiency, below a threshold of decency or with some of their basic needs not attended to. And it is so irrespective of such situations being in some sense reflective of people’s choices, that is, even if the situations in question were as such reasonably avoidable to them.

Given this underlying sufficiency-consideration, the following further concern arises. We need to recognize that sometimes sufficiency cannot be secured or reestablished after an accident has taken place. The damage done might to some extent be irrevocable. Take the paradigmatic case of the reckless biker Bert (Fleurbaey, 1995). In the wake of an accident causing serious harm to Bert, it would certainly be harsh not to assist him. This has gained the bulk of attention in the relevant debate. However, it is in a way glaringly obvious that some activities, including riding a bike without a crash helmet, carry the risk of landing those who engage in such activities with grave and partly irrevocable injuries. The latter include some that will transport individuals to insufficient situations on a permanent basis.

We shall later elaborate on various ways in which one might conceive of sufficiency, but to illustrate the basic point consider the following. One may reasonably think that having basic conditions of autonomy satisfied is part and parcel of an acceptable conception of sufficiency (Blake, 2001). If this is the case, then there are certain activities that may jeopardize autonomy or its basic conditions, and may do so in such a way that the requisite capacities cannot be restored after the fact. At the same time, there may be ways in which the society can restrict the activities in question that both infringe people’s liberty or autonomy in a limited way and stand a good chance of preventing the regrettable and irrevocable potential harm involved. Bert’s activity of riding without a crash helmet is a good case in point. It carries the risk of him experiencing a trauma to his head and associated brain damages. The latter may adversely affect a person’s “mental abilities to form intentions of a sufficiently complex kind, and plan their execution” (Raz, 1986: 372). As Feinberg has it, the harm involved is of such a nature that it cannot be avoided “at extortionate cost to our pocketbooks” or by any measure of assistance, it cannot ex post be avoided at all (Feinberg, 1986: 140). At the same time, there seems to be available measures of a kind that could prevent the harm in question at limited costs to the person’s autonomy. As Paula Casal points out, “one may attach considerable value to the freedom to drive … but not to the freedom to drive without seat belts” or, we may add, to the freedom to drive without a crash helmet (Casal, 2007: 322).

In the passage just quoted, Feinberg focuses on irreparable harm to others, predominantly other motorists (so-called “psychic harm”). But we think the point is valid a fortiori when we have the well-being of the motorcyclist in mind. The motorcyclist will have to live on in an insufficiency state, which could easily have been avoided. This brings us to the question of the shape of the duty that we plausibly have if we are concerned with sufficiency and are attentive to the problem of potentially irrevocable situations of insufficiency. The relevant duty is what we have referred to as the duty to avoid insouciance. The content of this, we suggest, is that when we, by infringing a person’s autonomy in limited ways, can prevent her from experiencing an unamendable situation of insufficiency, we have a duty to do so. This duty is ostensively and unashamedly paternalistic. That is, the reason for the proposed intervention is to prevent harm of a sufficiency-threatening kind to the person whose autonomy gets restricted. Given the importance of securing sufficiency, or the badness of insufficiency for the person experiencing it, and the limited badness of the necessary means for catering to sufficiency, there are strong reasons to think that we have a duty of the mentioned kind.

Straightforwardly, it would be a case of insouciance to let Bert proceed without headgear. Insouciance is avoided in similar cases referred to by Gerald Dworkin as “safety cases” in which we, for example, oblige “hunters to wear brightly colored jackets, sailors to wear life-preservers, and drivers to wear seat belts” and do so in order to significantly reduce the risk of immediate and irreparable harm in the event of an emergency (Dworkin, 1988: 125). Other examples are safety regulations concerning the time periods within which people may engage in certain activities and the conditions under which they may do so. For example, Peter de Marneffe mentions a policy of prohibiting hikers from hiking in the mountains after sunset (De Marneffe, 2006: 82). And a park ranger may be authorized to bar hikers from embarking on a hike if the weather conditions are treacherous.5

The duty to avoid insouciance may also be discharged in a different way. For example, we may in various ways try to temper or completely extinguish people’s desire to engage in certain activities that may land them in insufficient situations. Say that we could mold people’s desires in such a way that they do not even desire to drive without seat belts or without a crash helmet. To some extent, existing mandates may have had that effect. Given that thwarting the desire in question once it has developed may be acceptable, it might be the case that it would be acceptable alternatively to act so as to avoid that people develop the preference in the first place (cf. Arneson, 1989: 433–434).

Objections to the Importance of Avoiding Insouciance

The argument in the previous section for the duty to avoid insouciance is subject to various kinds of objections. We consider three of these below. Two of them are derived from Jessica Flanigan’s recent article on the case for seat belt mandates (Flanigan, 2016). The third challenges the symmetry we claim between the duty to avoid harshness and the duty to avoid insouciance (as both being related to sufficiency and as some that we have strong reasons to discharge).

Are Restrictions on the Liberty to Drive without Seat Belts and Helmets Really “Limited”?

In our argument for helmet mandates and similar security mandates, we suggested that people plausibly attach limited value to the liberty to drive without seat belts and helmets. Flanigan challenges this claim: some attach considerable value to these freedoms (Flanigan, 2016: 12). In a review of Brian Barry’s Justice as Impartiality (Barry, 1995), Andrew Reeve made a similar rejoinder to Barry’s case for helmet mandates. Reeve pointed out that riding without a helmet may be pivotal to a certain “born to run” lifestyle (Reeve, 1996). Does this point refute the case for helmet mandates? It does quite clearly imply that helmet mandates are not justified in their application to people with a “born to run” lifestyle. While helmet mandates, if effective, would still offer protection to people with such a lifestyle, the mandates would involve a serious infringement of their liberty or autonomy. Hence, the duty to avoid insouciance would not be triggered with respect to them (this duty, recall, is activated when the infringement required to avoid insufficient situations is limited or trivial). However, the group in question is presumably small, and it is not plausible to hold that a mandate ought only to be enacted in case that it is justified in its application to each and every person. In translating paternalistic principles from the interpersonal setting in which they are often developed, we need what Douglas Husak calls mediating maxims; and the “absolutist” maxim just mentioned is not a plausible one (Husak, 2003: 400). On a more plausible consequentialist maxim, it is a reasonable assessment that refraining from enacting a helmet mandate would have the regrettable consequence of allowing a number of costly situations of insufficiency. In the standard case, these situations could have been avoided at limited costs to people’s liberty or autonomy, and they would have been thus prevented had a mandate been in place.

The argument for helmet mandates based on the duty to avoid insouciance presumes a subjective understanding of the limited nature of the restrictions on liberty or autonomy involved. That is, it focuses on the value people attach to the liberty to drive without a helmet. The argument may alternatively appeal to an objective understanding: the liberty to drive without a helmet is not a liberty to which one ought to attach much value. We believe that such a rendering of the argument constitutes a plausible version of the argument. Flanigan, however, believes that it implies problematic perfectionist claims (Flanigan, 2016: 13). In response, at least some form of perfectionism might in fact not be objectionable from a liberal point of view—and seat belt and helmet mandates and other similar policies may be part of such an acceptable program of perfectionism. In his ingenious scrutiny of the notion of “conceptions of the good life” and the role it plays in leading liberal accounts, Joseph Chan usefully distinguishes between two different accounts of judgments of ways of life, one more ambitious than the other (Chan, 2000). The more ambitious “comprehensive” judgments cover all aspects of life and makes systematic rankings of the values involved. These are very hard to establish, and they are the kind of doctrines that would clearly be objectionable for the state to impose on its citizens. The less ambitious local judgments involve more elementary, easier to establish kind of judgments such as the judgment that “those ways of life which possess a substantial range of goods are more valuable than those which are seriously lacking” (Chan, 2000). Chan refers to the state perfectionism that builds on the less ambitious kind of judgments as “moderate perfectionism.” This kind of perfectionism, he plausibly points out, is a much more acceptable basis upon which a state may in part conduct its policies than is the kind of perfectionism associated with comprehensive doctrines (“extreme perfectionism”). Moderate perfectionism has the state promoting, for example, certain basic human virtues including “reason (especially practical wisdom)” and presumably necessary conditions for this (Chan, 2000: 11, 14).

As an instance of moderate perfectionism, we suggest that a way of life which includes, say, a good deal of risk-taking but where certain safety precautions are observed in order for its practitioners not to face excessive risks of jeopardizing their rational capacities is more worthwhile or valuable than a way of life which includes also a good deal of risk-taking, but where safety precautions are routinely flouted. The state may legitimately adopt measures that either discourage people from developing intentions to live lives of the latter kind or prevent them from undertaking risky activities without basic safety precautions. In brief, a just and decent state should see to it that insouciance is avoided.

The Autonomy-Argument

When introducing our case for the duty to avoid insouciance we made use of an argument from autonomy. To rehearse, one may reasonably think that basic capacities for autonomy are part and parcel of a plausible sufficiency threshold. Some of our activities may jeopardize these capacities and do so in ways such that they cannot be restored. Accordingly, we may have strong reasons (indeed, be duty bound) to impose certain restrictions on the activities in question―restrictions with the aim of avoiding that the activities have autonomy-undermining consequences. In the section on sufficiency below, we shall expand on this autonomy-related argument. It offers, we believe, a strong case for our proposed duty to avoid insouciance.

However, one of Flanigan’s objections to the case for seat belt mandates targets exactly this argument. To be precise, Flanigan does not deny that certain activities may be restricted on the ground that they jeopardize our autonomy. However, the set of activities that can be restricted on these grounds is very limited, and it does not include activities such as riding a motorcycle without a crash helmet as we and others would like to think. Why not? Because in order to qualify as a member of the restricted set, the activity in question should “intrinsically damage a person’s autonomous capacities” (Flanigan, 2016: 14). Ending one’s life may be an example (Flanigan, 2016: 13; Velleman, 1999). Operating a motorcycle without wearing a crash helmet, smoking, not exercising, eating fatty food, and so on are not. Such activities merely involve risking the loss of one’s autonomous capacities (Flanigan, 2016: 14).

What do we say to this objection? We say that it relies on a questionable categorical distinction between activities that “intrinsically damage a person’s autonomous capacities” and risky activities. Once this distinction is dissolved, it is not clear why certain activities that involve significant risk of damaging a person’s autonomous capacities could not qualify as some that we may reasonably restrict in the service of preserving autonomous capacities. The distinction is questionable in that even actions such as ending one’s life are not necessarily foolproof. For example, the gun one is using for the purpose might jam or one might, as, for example, Vronsky does in Tolstoy’s Anna Karenina, miss (Tolstoy, [1877] 1999). In terms of the threat involved to one’s autonomous capacities, this act may not be different from an extremely risky act such as a high-wire walk across Grand Canyon in treacherous conditions. Presumably, there are sound autonomy-catering reasons for imposing restrictions on such acts. If this is so, it is not clear why certain acts or activities which involve less but still considerable risks to one’s autonomous capacities might not qualify for having restrictions imposed on them. There is no clear nonarbitrary cut-off point to be invoked. On the implied rationale, some arguments for restrictions on certain risky activities may come out as compelling (Andersen and Midtgaard, 2016; Grill and Voigt, 2016). To be sure, a number of considerations go into such arguments. And the jury is still out with respect to even obviously risky activities such as smoking. However, the important point here is that autonomy considerations may warrant a critical look on these activities. In some cases, a balanced verdict may show that a given activity ought to be restricted with the purpose of preserving autonomous capacities.

The Disanalogy Objection

The central claim of our argument is that if you care about sufficiency, it is, in addition to avoiding harshness, important to avoid insouciance. The disanalogy objection that we consider now takes issue with this claim. In the case of harshness, the argument points out, the situation is one in which a person is in a very bad situation and where this is (more or less) amendable. In other words, we can lift him out of the bad situation or approximately so, and we may do so in ways that involve little or reasonable costs to us—and, importantly in this connection, without costs or nearly so to the person being offered assistance. The person who needs assistance is simply offered a crucial benefit, a benefit that he or she may decline. But whether or not he or she does so, receiving the offer would certainly seem only to augment his or her autonomy. Indeed, we may be offering the individual assistance of a kind that is necessary for him or her to retain or restore conditions necessary for his or her autonomy. In such circumstances—being in the position to benefit people considerably and to do so at little or reasonable costs to oneself—it would surely be wrong not to offer assistance (cf. Singer, 1972). This makes the harshness objection profound. Furthermore, and important in this context, discharging the duty of avoiding harshness does not seem to involve any kind of paternalism. That is, we would not, it seems, for their own good be imposing restrictions on people’s liberty or affecting them by the use of means other than rational persuasion; apparently, we do not infringe their autonomy at all.

Insouciance, in contrast, or rather the alleged virtue of a noninsouciant society, involves tangible restrictions or means other than rational persuasion being imposed on the risk-incurring or imprudent agents. While such measures may indeed benefit the person in certain ways, perhaps even protect him against suffering instances of insufficiency without recourse, they may involve restrictions that are, all things considered, intolerable. Coercion is always an affront to autonomy, something that deflects the person from directing his or her path in accordance with his or her own will and not with that of another (Blake, 2001; Raz, 1986). Coercing people for their own good evinces a problematic attitude of disrespect on the part of the coercer, implying that the coercer questions the capability of the coercees to run their own lives (Anderson, 1999: 301–302; Begon, 2016: 364; (Flanigan, 2016; Shiffrin, 2000: 207, 231). This, as opposed to the avoidance of harshness, makes the avoidance of insouciance, at least, prima facie morally problematic. Perhaps the risk of ending up in thoroughly bad situations from which one cannot be bailed out is the fair and acceptable price that one would have to pay for enjoying the freedom to indulge without interferences in one’s preferred activities. Accordingly, society should indeed display insouciance to some extent. It is not something society should avoid on par with avoiding harshness.

In response, the claimed disanalogy is overstated. As has been emphasized in discussions surrounding the right to receive aid in emergencies, it is appropriately considered nonwaivable when related to sufficiency concerns. In case that it were waivable, imprudent people might end up in situations of insufficiency where others would not be obliged to come to their rescue (Bou-Habib, 2006: 256–257; Jones, 1985: 145). Denying people the liberty to waive their right to be offered assistance when in an emergency and doing so apparently in their own interests, appears, however, not only to infringe people’s autonomy in nontrivial ways but to do so on paternalistic grounds (Midtgaard, 2015). Indeed, it seems to raise concerns similar to those pointed to above in connection with the duty to avoid insouciance.

However, there is a rejoinder to this point. Although the infringement of autonomy involved even in offering assistance might be nontrivial, it may seem less serious than the infringements involved in avoiding insouciance. The potential recipient may say “Thank you, but no thank you,” and even if this utterance is reflective of some irritation (“Thank you very much, but no thank you”),6 the restriction might seem relatively limited. The force of this rejoinder is, however, questionable. First, some recipients might feel strongly against being offered assistance when in an emergency, for example, finding this humiliating (cf. Jones, 1985: 146). Second, note that on some accounts, the duty to avoid harshness is used as a foundation upon which a collective system of mutual assistance is erected (Anderson, 1999).

The upshot is that the disanalogy claimed by the objection under consideration in this section is questionable. And we would like to add that even if it were possible to show that the duty to avoid harshness merely involved an innocent offer, this would not suffice to challenge the limited nature of the restrictions needed to avoid insouciance. Hence, demonstrating a disanalogy would not call in question the duty to avoid insouciance as an important duty alongside the duty to avoid harshness.

Sufficiency

Sufficiency is, as we claimed above, the underlying concern in debates about the plight of the imprudent, and the concerns to avoid harshness and insouciance are propelled by it. In this section, we elaborate first on the content of the sufficiency concern, second on how egalitarians understand its value, and third on its relation to harshness and insouciance.

Sufficiency in our account requires, first, not leaving people in thoroughly bad situations (i.e. avoiding harshness). Second, at least with regard to those thoroughly bad situations from which people cannot be bailed out, it requires measures ensuring that people do not find themselves in such situations (i.e. avoiding insouciance; cf. Stemplowska, 2009: 251). The phrase “thoroughly bad situations” implies a fairly low threshold, which is intentional. The relevant threshold may be described in various ways, including in terms of basic needs (e.g. Olsaretti, 2004: 154; Segall, 2010: 68, 75–78), basic capabilities required for living a life of human dignity (Fourie, 2017: 21; Nussbaum, 2000), basic capabilities required for being able to function as an equal citizen in a democratic state (Anderson, 1999; Fourie, 2017: 21), basic goods required for human flourishing (Pogge, 2002: 27–51), and basic goods required for living as rationally autonomous agents (Blake, 2001: 266–273; Kymlicka, 1991; Rawls, 1999a; Raz, 1986a). Although there are important differences between these accounts, they converge in part on the kind of rights or goods that they believe should be in place in order for us to say that people are not in a deprived state. They point, we may say, to the importance of people having access to a minimally adequate share of liberties and opportunities and means for making use of such freedoms (Casal, 2007: 323; Pogge, 2002: 49; Rawls, 1999b: 582).

Crucially in the present context, the accounts point plausibly to the importance of people having certain basic rational or deliberative abilities. Michael Blake, for example, following Joseph Raz, emphasizes the importance of people having “appropriate mental abilities,” to wit, “the individual in question must have the abilities to form the complex intentions required of an autonomous planning agent, and must have the forms of rationality sufficient to follow through on what those intentions require” (Blake, 2001: 267).7 Anderson focuses on the importance of people having access to “the basic conditions of human agency.” These include “knowledge of one’s circumstances and options, the ability to deliberate about means and ends, the psychological conditions of autonomy, including the self-confidence to think and judge for oneself, freedom of thought and movement” (Anderson, 1999: 317–318 emphasis added). Now, as Blake points out, there may not be much that political institutions or “political life” can do in terms of providing or securing the mentioned capacities.8 It is important to note, though, that there is at least one important thing that political institutions can do in this regard: they can in various ways try to keep people from acting in ways that carry a significant risk of them losing the rational capacities in question. In brief, institutions may be constructed so that our duties to avoid insouciance or some of these duties are discharged through them. Accordingly, the avoidance of insouciance is integral to a program required to cater to ensuring sufficiency where this is understood in accordance with a recognized and eminently plausible conception of what this amounts to.

The “positive thesis” that it is especially important to ensure that people have enough is widely shared and integral to the doctrine of sufficiency. In Casal’s rendering, it stresses “the importance of people living above a certain threshold, free from deprivation” (Casal, 2007: 298–299; Fourie, 2017: 21; Shields, 2012: 105, 115). Since Anderson’s relational egalitarianism represents “a sufficientarian standard of justice in the metric of capabilities,” her view obviously endorses the positive thesis (Anderson, 2010: 83). According to Anderson, if people lack important capabilities “(due to severe brain damage, for instance),” this will plausibly undermine their potential for relating to others as equals (Anderson, 2010: 84).

The positive thesis is endorsed by many nonsufficientarians as well, but the way in which egalitarians believe that this thesis relates to other principles of egalitarian justice varies (Axelsen and Nielsen, 2015: 407; Temkin, 2003: 65; Shields, 2012: 106). Some argue that ensuring a certain sufficiency threshold should be given lexical priority to other principles. For example, as John Rawls points out:

[T]he first principle [of justice] covering the equal basic rights and liberties may easily be proceeded by a lexically prior principle requiring that citizens’ basic needs be met, at least in so far as their being met is necessary for citizens to understand and to be able fruitfully to exercise those rights and liberties. Certainly any such principle must be assumed in applying the first principle (Rawls, 2005: 7; notes excluded).

In this way, the principles of Rawls’ liberal egalitarianism presuppose an approval of the positive thesis. In the same way, some luck egalitarians have argued that luck egalitarian principles should be “constrained” by sufficiency principles where both types of principles are considered to be concerns of egalitarian justice (Barry, 2006: 100; Casal, 2007: 322; Brown, 2005: 307–308; Knight, 2015: 122–123).9 Others point out that luck egalitarianism is incomplete in the sense that it does not include an account of stakes. That is, it does not specify the pay-offs or costs that should be associated with various types of conduct (Olsaretti, 2009; Stemplowska, 2009). And when we provide an account of stakes, it might be a plausible part of this account that it rules out certain types of absolute disadvantages, that is, that it caters to sufficiency.

That some minimum threshold of sufficiency should be ensured may not, however, be required by the distributive principles of justice that luck egalitarians appeal to. As described previously, on a standard interpretation, luck egalitarians believe that outcomes that reflect people’s voluntary choices or their exercises of responsibility are fair. Among these outcomes may be situations characterized by being insufficient. Only in so far as the outcomes in question are involuntary would luck egalitarians be committed to preventing them. Relating this to a distinction often invoked in the literature on paternalism would at most commit them to so-called soft paternalism. The latter endorses interferences aimed at avoiding involuntary self-regarding harm, and only such interferences. This is arguably autonomy-preserving in that the harmful act in question is not an expression of the person’s will. Hence, many liberals, including, for example, John Stuart Mill, have found it innocuous (Mill, [1859] 2011). Hard paternalism, by contrast, prevents self-regarding harm even if it is perfectly voluntary. This view is much more controversial than soft paternalism because it is prepared to infringe people’s liberty or autonomy. Luck egalitarians, however, as noted, appear to steer clear of this and to be unconcerned with cases of insufficiency due to voluntary choices.

However, as has been pointed out recently by Kristin Voigt, the standard of voluntariness pertaining to the luck egalitarian view is plausibly much more stringent than the standard involved when we draw the distinction between soft and hard paternalism (Voigt, 2015). This means that luck egalitarians would in fact embrace measures averting self-regarding harm of a hard paternalistic kind. The harm in question may, for example, stem from unchosen deficiencies in choice-making capacities. Hence, the luck egalitarian view is not necessarily hostile to the importance of avoiding insouciance emphasized in this article. Still, to be sure, there will be, at least in principle, cases in which the more stringent luck egalitarian standard of voluntariness is met and where agents act in ways that land them in insufficient situations. In such cases, some luck egalitarians are prepared to sacrifice sufficiency (see, for example, Stemplowska, 2009), and we have to admit that our argument would be ineffective against these theorists. They do not have to endorse hard paternalism (or more precisely ultra-hard paternalism) because they are not committed to sufficiency in an uncompromising way.10

How does the concern of ensuring sufficiency relate to the concerns of avoiding harshness and insouciance? Sufficiency, nonharshness, and noninsouciance are obviously closely intertwined notions. Sufficiency is the ultimate concern, whereas nonharshness and noninsouciance are derivative. To illustrate this point, consider Tom, who is well above a plausible understanding of sufficiency. If Tom acts imprudently—say, by participating in high-risk gambling—but there is no risk that these gambles will make Tom fall below the threshold, it would neither be a case of insouciance to let Tom incur the risk nor harsh not to assist him if he loses. For there to be a plausible case of insouciance or harshness, sufficiency must be at stake.

Avowedly Nonpaternalistic Justifications for Restrictions Preventing Insufficiency

Until now, we have argued that to adequately cater to the important concern of ensuring sufficiency, it is sometimes necessary and reasonable to impose certain restrictions on people’s conduct. This point is relevant relative to all of the positions mentioned in the introduction. However, anti-paternalists may have another arrow in the quiver. They may try to justify the kind of restrictions that we have invoked in the service of avoiding insouciance without appeal to paternalistic reasons. However, the prospects for such justifications are, we believe, dim.

Consider a common nonpaternalistic rationale for catering to the needs of the imprudent. If risk-takers get injured, they release a burdensome moral duty of assistance on the part of others. The latter, however, have a right to defend themselves against having to discharge their duty of assistance or against having to do so uncompensated. They may, as forcefully argued by Paul Bou-Habib, do so by compelling risk-takers to take out insurance (Bou-Habib, 2006; Jones, 1985). Compulsory insurance, however, will often fail to secure sufficiency. Although it facilitates imprudents setting aside means for their own rescue, assistance in emergencies, as we have recurrently underlined, often falls short of restoring sufficiency. To preserve sufficiency, society would often need to constrain the sufficiency-endangering activity in various ways.

Could such constrains or safety requirements be justified nonpaternalistically with reference to material or financial costs? Feinberg’s ingenious analysis of this so-called public charge argument for helmet requirements, for example, provides reasons for skepticism in this regard. As mentioned earlier, Feinberg points out that the salient harm involved in the case of the motorcyclist with the “smashed head” is not of a kind that can be solved by paying up. Hence, the public charge argument does not provide reasons for compelling the biker to wear a helmet (Feinberg, 1986: 140–141; see also Dworkin, 1988: 127). In general, it is a highly contingent matter whether a certain self-regarding harmful act generates other-regarding costs that warrant imposing restrictions on the conduct of the imprudent. The match between cases in which proscribing people’s actions is justified nonpaternalistically in this way and the cases in which we find it independently important that measures ensuring sufficiency are in place is, at best, imperfect.

This is not the end of the story, though. Feinberg turns instead to the so-called psychic harm argument. In a nutshell, he suggests that helmet requirements may be justified on the ground that people in proximity to a biker who crashes and crushes his or her head would suffer significant psychic harm. Other interpretations of the psychic harm argument furthermore emphasize the more general costs to society or “society’s moral environment” of witnessing or being aware of situations of insufficiency (Casal, 2007: 322–323). It is doubtful, however, that the psychic harm argument is a plausible or sufficiently weighty ground upon which to impose restrictions on people’s liberties. First, just as in the case of the public charge argument, it is a highly contingent matter whether psychic harm of a kind and magnitude that underwrite restrictions arises relative to sufficiency-threatening acts, which we have independent reasons for believing should be restricted. For example, the population may be of a quite stern and unforgiving nature, believing that a biker who incurs a head injury that could have been avoided had he or she worn a crash helmet “had it coming to him or her” and losing no sleep from being in proximity to a person who suffers such a loss. In addition, allowing this rationale may justify restrictions on conduct that we would be loath to prohibit.11 For example, it might be that some experience severe psychic harm from seeing people in revealing summer clothing, eating ice cream in public, or kissing. Still, this does not seem to warrant imposing restrictions on the conduct in question. And even if, finally, psychic harm may justify some independently plausible measures that ensure sufficiency, the paternalistic argument seems to us to be much more compelling and forceful. While the prevention of psychic harm may suggest a reason for imposing restrictions on certain conduct, averting harm to the agent himself or herself is a much more pressing concern.

Conclusion

In this article, we have emphasized an important dilemma between avoiding paternalism and preserving sufficiency—concerns to which most attach profound value. Specifically, we have argued that the requirements of sufficiency presuppose not only a duty to relieve but also a duty to prevent suffering—both harshness and insouciance must be avoided. Avoiding insouciance demands, however, certain paternalistic actions and restrictions. For the same reason, one cannot avoid paternalism without this involving making huge sacrifices. Of course, we should sometimes stomach such outcomes and break free from the requirements of sufficiency, but certainly not in every case. Indeed, our reasons to avoid insufficiency often outweigh our reasons to respect people’s voluntary self-regarding conduct. There are, in brief, grounds for thinking that avoiding harshness is “not enough,” and that this is a compelling reason to accept particular paternalistic policies and actions. If our argument is correct, those who embrace avoiding harshness to cater to sufficiency but reject the duty to avoid insouciance and its implied paternalism occupy an instable position. In light of the importance of achieving sufficiency, they ought to rethink their opposition to paternalism.

Footnotes

Acknowledgements

For helpful comments to various versions of this article, we are grateful to: the members of the Political Theory Section, Department of Political Science, Aarhus BSS, Aarhus University; participants in a Political Theory workshop, ‘Justice, Freedom, and the Good Life’, 8–9 December 2016, at the same institution; participants in the Political Theory Section of the ECPR General Conference, 7–10 September 2016, in Prague; participants in the Senior Seminar, Philosophy, University of Glasgow, February 2017; three anonymous referees for this journal. We are especially grateful to David V Axelsen, Robert Huseby, Kasper Lippert-Rasmussen, Liam Shields, and Jens D Thaysen for detailed written comments.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Viki M L Pedersen is a PhD student in the Political Theory section at Department of Political Science, Aarhus BSS, Aarhus University. Her project focuses on paternalism, especially its relation to the two concerns of ensuring sufficiency and internalizing the costs of people’s choices.

Soren Flinch Midtgaard is an Associate Professor at the same institution. His work is presently focused on the topics of paternalism and egalitarianism.

References

Andersen

Midtgaard

(2016) Stay Out of the Sunbed! Paternalistic Reasons for Restricting the Use of Sunbeds. Public Health Ethics. Epub ahead of print 1 September. DOI: 10.1093/phe/phw038.

Anderson

(1999) What Is the Point of Equality? Ethics 109 (2): 287–337.

Anderson

(2010) Justifying the Capabilities Approach to Justice. In: Brighouse

Robeyns

I I

(eds) Measuring Justice. Cambridge: Cambridge University Press, pp.81–100.

Arneson

(1989) Paternalism, Utility, and Fairness. Revue internationale de philosophie 43: 409–437.

Arneson

(2005) Joel Feinberg and the Justification of Hard Paternalism. Legal Theory 11: 259–284.

Axelsen

Nielsen

(2015) Sufficiency as Freedom from Duress. Journal of Political Philosophy 23 (4): 406–426.

Barry

(1995) Justice as Impartiality. Oxford: Clarendon Press.

Barry

(2006) Defending Luck Egalitarianism. Journal of Applied Philosophy 23 (1): 89–107.

Begon

(2016) Paternalism. Analysis Reviews 76 (3): 355–373.

10.

Blake

(2001) Distributive Justice, State Coercion, and Autonomy. Philosophy & Public Affairs 30 (3): 257–296.

11.

Bou-Habib

(2006) Compulsory Insurance without Paternalism. Utilitas 18 (3): 243–263.

12.

Brown

(2005) Luck Egalitarianism and Democratic Equality. Ethical Perspectives 12: 293–339.

13.

Casal

(2007) Why Sufficiency Is Not Enough. Ethics 117 (2): 296–326.

14.

Chan

(2000) Legitimacy, Unanimity, and Perfectionism. Philosophy & Public Affairs 29 (1): 5–42.

15.

De Marneffe

(2006) Avoiding Paternalism. Philosophy & Public Affairs 34 (1): 68–94.

16.

Dworkin

(1988) Paternalism—Some Second Thoughts. In: Dworkin

(ed.) The Theory and Practice of Autonomy. Cambridge: Cambridge University Press, pp.121–129.

17.

Dworkin

(2000) Sovereign Virtue: The Theory and Practice of Equality. Cambridge, MA; London: Harvard University Press.

18.

Feinberg

(1986) Harm to. Self. New York: Oxford University Press.

19.

Flanigan

(2016) Seat Belt Mandates and Paternalism. Journal of Moral Philosophy 14: 291–314.

20.

Fleurbaey

(1995) Equal Opportunity or Equal Social Outcome. Economics and Philosophy 11: 25–55.

21.

Fourie

(2017) The Sufficiency View: A Primer. In: Fourie

Rid

(eds) What Is Enough?: Sufficiency, Justice, and Health. Oxford; New York: Oxford University Press, pp.11–29.

22.

Grill

(2015) Antipaternalism as a Filter on Reasons. In: Schramme

(ed.) New Perspectives on Paternalism and Health Care. Bern: Springer International Publishing, pp. 47–63.

23.

Grill

Voigt

(2016) The Case for Banning Cigarettes. Journal of Medical Ethics 42: 293–301.

24.

Groll

(2012) Paternalism, Respect, and the Will. Ethics 122 (4): 692–720.

25.

Hausman

Welch

(2010) Debate: To Nudge or Not to Nudge. Journal of Political Philosophy 18 (1): 123–136.

26.

Husak

(2003) Legal Paternalism. In: LaFollette

(ed.) The Oxford Handbook of Practical Ethics. Oxford; New York: Oxford University Press, pp.387–412.

27.

Jones

(1985) Toleration, Harm and Moral Effect. In: Horton

Mendus

(eds) Aspects of Toleration: Philosophical Studies. London; New York: Methuen, pp.136–157.

28.

Knight

(2015) Abandoning the Abandonment Objection: Luck Egalitarian Arguments for Public Insurance. Res Publica 21: 119–135.

29.

Kymlicka

(1991) Liberalism, Community and Culture. Oxford: Clarendon Press.

30.

Lippert-Rasmussen

(1999) Arneson on Equality of Opportunity for Welfare. Journal of Political Philosophy 7 (4): 478–487.

31.

Midtgaard

(2015) Non-renounceable Rights, Paternalism and Autonomy. Utilitas 27 (3): 347–364.

32.

Midtgaard

(2016) Paternalism. In: Thompson

(ed.) Oxford Research Encyclopedia of Politics. Oxford: Oxford University Press. DOI: 10.1093/acrefore/9780190228637.013.201.

33.

Midtgaard

(2017) Stakes, Choice-Sensitivity and Paternalism. Unpublished paper.

34.

Mill

([1859] 2011) On Liberty. Northampton: White Crane Publishing.

35.

Nussbaum

(2000) Women and Human Development: The Capabilities Approach. Cambridge: Cambridge University Press.

36.

Olsaretti

(2004) Liberty, Desert and the Market: A Philosophical Study. Cambridge: Cambridge University Press.

37.

Olsaretti

(2009) Responsibility and the Consequences of Choice. Proceedings of the Aristotelian Society 109: 165–188.

38.

Pedersen

(2017) On Central Non-Paternalistic Arguments. Unpublished paper.

39.

Pogge

(2002) World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms. Cambridge: Polity Press.

40.

Rawls

(1999a) A Theory of Justice (Revised edition). Oxford: Oxford University Press.

41.

Rawls

(1999b) John Rawls: Collected Papers (ed. S Freeman). Cambridge, MA; London: Harvard University Press.

42.

Rawls

(2005) Political Liberalism (Expanded edition). New York: Columbia University Press.

43.

Raz

(1986) The Morality of Freedom. Oxford: Oxford University Press.

44.

Reeve

(1996) Impartiality between What? Lifestyles, Conceptions of the Good, and Harm. Political Studies 44 (2): 314–317.

45.

Scoccia

(2008) In Defense of Hard Paternalism. Law and Philosophy 27 (4): 351–381.

46.

Segall

(2010) Health, Luck, and Justice. Princeton, NJ; Oxford: Princeton University Press.

47.

Shields

(2012) The Prospects for Sufficientarianism. Utilitas 24 (11): 101–117.

48.

Shiffrin

(2000) Paternalism, Unconscionability Doctrine, and Accommodation. Philosophy & Public Affairs 29 (3): 205–250.

49.

Singer

(1972) Famine, Affluence, and Morality. Philosophy & Public Affairs 1 (1): 229–243.

50.

Snelling

(2014) What Wrong with Tombstoning and What Does This Tell Us about Responsibility for Health? Public Health Ethics 7 (2): 144–157.

51.

Stemplowska

(2009) Making Justice Sensitive to Responsibility. Political Studies 57: 237–259.

52.

Temkin

(2003) Equality, Priority or What? Economics and Philosophy 19: 61–87.

53.

Tolstoy

([1877] 1999) Anna Karenina. London: Wordsworth Classics.

54.

Velleman

(1999) A Right of Self-Termination? Ethics 109 (3): 606–628.

55.

Voigt

(2007) The Harshness Objection: Is Luck Egalitarianism Too Harsh on the Victims of Option Luck? Ethical Theory and Moral Practice 10: 389–407.

56.

Voigt

(2015) Paternalism and Equality. In: Schramme

(ed.) New Perspectives on Paternalism and Health Care. Bern: Springer International Publishing, pp.87–100.

57.

Voorhoeve

(2008) Scanlon on Substantive Responsibility. Journal of Political Philosophy 16 (2): 184–200.

58.

Williams

(2006) Liberty, Equality, and Property. In: Dryzek

Honig

Phillips

(eds) The Oxford Handbook of Political Theory. Oxford; New York: Oxford University Press, pp.488–506.