The Presumption of Duties to Oneself

Abstract

1. Introduction

Dale Dorsey (2016) argued that the project of identifying a distinctive or defining feature of morality should not be prior to substantive first-order moral inquiry. Since any candidate for being such a feature would play the role of the gatekeeper into the moral domain, it must be identified with a high degree of certainty. It must also be sufficiently fine-grained to rule out other normative but nonmoral rules, and sufficiently coarse-grained to allow for variations between different viable moral theories. We cannot hope to identify such a feature without substantive first-order moral inquiry. If Dorsey is right, then the popular belief that moral considerations do not bear on intrapersonal conduct cannot be established without substantive inquiry into the content of morality.

I agree with Dorsey and think that the best case for duties to oneself would rely on substantive first-order inquiry. In this paper, I do the next best thing which is to show that a prima facie plausible first-order principle of moral impartiality provides a strong reason for establishing a presumption in favor of the existence of duties to oneself.¹ A virtue of this approach is that it does not operate within any specific moral framework and should therefore appeal even to those who are disinclined to take the idea of duties to oneself seriously.

The view that moral considerations do not bear on purely self-regarding decisions conflicts with our shared conviction that morality is in some fundamental sense impartial. Moral impartiality means, roughly, that everyone matters equally for moral justification: no one can be simply excluded, and any difference in consideration must be justified by some morally relevant distinction (e.g., Warnock 1971, 149).

Conceptions of moral impartiality differ depending on what is taken to be central to moral justification.² I remain neutral among the competing views. However, I take it to be uncontroversial that morality is not impartial unless it takes both the welfare and the will of individuals to have equal significance for justification. So, whether an action would harm or benefit or whether it is consensual are moral considerations that are relevant not only for other-regarding decisions but for purely self-regarding ones, too; and those moral prohibitions, requirements, and permissions that are plausibly thought of as grounded in the consideration of well-being or autonomy would potentially apply in the context of purely self-regarding behavior. Therefore, the principle of moral impartiality provides a robust reason to establish the presumption in favor of duties to oneself.

It is important to note that the ideal encapsulated in the correct conception of moral impartiality is substantive and its application takes morally relevant distinctions into account. If we have reason to believe in a morally relevant distinction between oneself and others, and if this distinction constitutes a fundamental feature of morality, then the correct conception of moral impartiality will factor it in, just as it factors in other morally relevant distinctions. Thus, it is possible that the conflict between this ideal and the view that moral considerations do not bear on purely self-regarding decisions is merely apparent.³

In what follows, I proceed on the assumption that everyone matters equally for moral justification. First, I claim that the strongest reason to believe in a morally relevant self-other distinction derives from commonsense morality. Commonsense morality evaluates actions differently or asymmetrically merely on the basis of whether they affect the agent or someone else. If this asymmetry cannot be explained otherwise, i.e., if the self-other distinction is necessary and sometimes sufficient to explain our asymmetric intuitions, then we are justified in positing the distinction. I argue that the asymmetry is best explained by the presupposition of self-consent. Second, I consider the objection that the consent-based explanation of the asymmetry renders potential duties to oneself practically inconsequential. I respond that the objection overlooks significant differences between other-regarding and self-regarding contexts. I demonstrate those differences by discussing both individual autonomy as the predominant ground for consent and what autonomous living involves. I argue that there is a disanalogy between respecting the autonomous agency of another competent person and that of oneself. The former typically requires taking another’s will as structurally relevant and morally decisive in our practical reasoning. The latter typically requires guiding our practical reasoning by our values and deep commitments.

If my argument is on target, then the first-order principle of moral impartiality provides a strong reason to establish a presumption in favor of the existence of duties to oneself. This shifts the burden of proof from those who think that the moral considerations bear on purely self-regarding decisions onto those who deny it. While this may not settle the substantive debate (that requires substantive first-order inquiry), it is a significant step towards finding a way to eventually resolve it.

2. From The Self-Other Asymmetry To The Self-Other Distinction

Commonsense morality evaluates actions differently or asymmetrically merely on the basis of whether they affect the agent or someone else. The asymmetry is expressed in widely shared and robust intuitions concerning the permissibility of self-harming and risk-taking actions, as well as about the praiseworthiness of self-sacrifice.

Consider three generic formulations of such asymmetrical evaluations:⁴

1. It would be wrong to allow (risk of) harm to befall someone else, yet we believe that it is permissible to harm oneself.

2. Whenever an agent chooses to sacrifice some benefit to herself or to go through a great deal of trouble for the sake of a small benefit to another person, we consider her action praiseworthy, provided that she was not required to benefit that person.⁵

3. If, while benefiting someone, an agent takes the opportunity to simultaneously advance her self-interest, her doing so takes away from the praiseworthiness of her action.

One might take commonsense morality to indicate that there is a morally relevant distinction between oneself and others (Ross 1939, 72, 74, 272–79; Sidgwick 1981, 431ff; Portmore 2003, 307–12; Hooker 2013, 719). Some then infer that commonsense morality assigns positive moral value to the well-being or happiness of others, but not at all or to a lesser degree to the well-being or happiness of the moral agent herself.⁶ Others put the same point in terms of reasons. Douglas Portmore (2003, 307–308) writes: “The mere fact that performing an act would be in the agent’s self-interest does not constitute a moral reason for doing so. The reason an agent has to further her own self-interest is a nonmoral reason.”^7,8

It is important to keep apart (a) the commonsense morality thesis of the self-other asymmetry and (b) the philosophical thesis of a fundamental, morally relevant self-other distinction. For it is the latter that blocks the idea of duties to oneself while removing the tension between the ideal of moral impartiality and the view that moral considerations do not bear on purely self-regarding decisions. The impartial application of moral norms requires that the agent is guided solely by the distinctions identified as relevant by moral norms (Hooker 2013). If there is a morally relevant fundamental self-other distinction, i.e., that only others’ interests matter morally, then there is no contradiction between the principle of moral impartiality and the view that moral considerations do not bear on purely self-regarding decisions. Importantly, for the self-other distinction to qualify as a fundamental feature of morality, asymmetry must constitute the final court of appeal: it must be the ultimate justification-chain stopper, necessary and sometimes sufficient to explain our asymmetrical moral appraisal of self-regarding and other-regarding actions (Rawls 2005, 112–18).

In what follows, I argue that it is the presupposition of (self-)consent, not the self-other distinction, that best explains the self-other asymmetry of commonsense morality.

2.1 Explaining the asymmetry: consent or the self-other distinction?

One natural thought when comparing analogous self-regarding and other-regarding actions is that the former but not the latter always bring about the state of affairs to which the agent consents. This invites the hypothesis that commonsense morality does not reveal a morally relevant self-other distinction but simply reflects the presupposition of consent in purely self-regarding behavior.

Michael Slote (1985) and, following him, Douglas Portmore (2003) reject this, insisting that the asymmetry makes sense only in light of the self-other distinction. Both offer thought experiments designed to demonstrate overwhelming intuitive support for the self-other distinction as the only plausible explanation of the asymmetry. For convenience, I divide said thought experiments into two categories: (1) those where an action can either bring some benefit (or some loss) to oneself or the same size benefit (or loss) to someone else; and (2) those where an agent incurs significant loss for the sake of a small benefit to someone else.

The first kind of cases supposedly demonstrate that consent or the lack thereof has little to do with our asymmetric intuitions so that the self-other distinction is their only plausible explanation. The second kind of cases are intended to strengthen the case for the distinction by evoking asymmetric intuitions which, according to Slote and Portmore, would make sense only in light of said distinction.

I believe that neither Slote and Portmore make a good case, nor are they correct about the cogency of the consent hypothesis. I argue that the first kind of scenario (paired, equal benefit cases) yields conflicting intuitions between advocates and critics of duties to oneself. As for the second kind of scenario (paired, unequal benefit cases), even if they yield matching intuitions between advocates and critics of duties to oneself, they do not challenge the consent-based explanation. They merely presuppose rather than establish the self-other distinction.

According to Slote, the presence of consent makes no difference, because when others consent to actions which would harm them, it is nonetheless intuitively more wrong for the agent to perform them. As Slote (1985, 20) puts it: “If someone irrationally asks me to harm or kill him, it will presumably be [...] wrong of me to kill him, more wrong at any rate than if I irrationally choose to kill myself [...]” (my emphasis). Similarly, the absence of consent makes no difference because it is worse to negligently and thus unintentionally kill another person than to negligently kill oneself (Slote 1985, 20–21). According to Portmore’s less dramatic scenario, the presence of consent makes no difference because it is intuitively wrong to fail to take a 15-minute drive to a baseball game in order to cheer a depressed friend but not wrong to stay home and sulk when you yourself are depressed while knowing that there is only 15-minute drive away a game that would cheer you up (Portmore 2003, 310).

Contrary to Slote and Portmore, such cases (where the benefit/loss to oneself or to someone else are quantitatively and qualitatively equivalent) are too weak to dislodge the proposed consent-based explanation of the asymmetry for the simple reason that not everyone shares Slote’s and Portmore’s intuitions about these cases—I know I don’t. These cases do not give the opponent any additional reason to accept the claim they are meant to support. Since at least some advocates of duties to oneself would deny the asymmetry of intuitions in these cases, the appeal to intuitions cannot settle the issue.

However, Slote and Portmore take cases where the benefit or loss to the agent and to someone else are nonequivalent to be the most persuasive. Let us, therefore, consider two such thought experiments.

In the first scenario, proposed by Portmore (2003, 311), you hold a winning raffle ticket for a genuine autographed Babe Ruth bat, something you coveted for years and would forever cherish. You could give it away to a stranger who is sad because he didn’t win anything but who otherwise has no interest in baseball. The stranger’s sadness would quickly pass on its own. Portmore judges that if you gave the stranger the winning ticket, you would be acting foolishly but not immorally.⁹ He concludes that commonsense morality assigns no positive value to the agent’s well-being.¹⁰

In the second scenario, proposed by Slote (1985, 21), A chooses to endure a longer period of pain in order to spare B a shorter one, though both A and B agree that it is foolish and B does not consent to A’s choosing a longer period of pain. Since most people would judge it permissible for A to do so, the presupposition of consent makes no difference to our moral appraisal.

It is not clear whether cases such as ‘Winning Raffle Ticket’ and ‘Longer Pain’ are intended to disprove the consent hypothesis or simply to provide direct intuitive support for the self-other distinction.¹¹ The wording of ‘Longer Pain,’ for example, makes it look like it builds a case against consent. Yet the fact of B’s consent or nonconsent to A’s choosing a longer period of pain is irrelevant, since B’s consent can only affect A’s obligations to B, and it’s not clear that A has an obligation to B not to choose this way. As for A’s choosing the course of action that she deems foolish, Slote must mean that in so doing she does not implicitly consent.

However, the belief that a course of action is foolish does not necessarily invalidate consent. I might think it foolish of me to lend my car to a friend who scratches it every time I let him drive. My friend might even think I’m a fool for lending him my car. This does not mean that I lack the consenting mental state when I give him the key. My desire to help my friend become a better driver may be stronger than my attachment to my car. Thus, ‘Longer Pain’ does nothing to disprove the idea that it is the presupposition of consent that makes us judge it permissible for A to choose a longer period of pain.

On a charitable interpretation, we should read both “Winning Raffle Ticket’ and ‘Longer Pain’ as providing direct intuitive support for the self-other distinction. However, this support is effective only after the consent hypothesis is successfully refuted. Otherwise, a critic of the self-other distinction would maintain that the asymmetry of intuitions is best explained by the presupposition of self-consent. If someone, like me, thinks that the consent hypothesis is unrefuted by the equal benefit (or loss) cases, then one cannot credit the unequal benefit cases with their intended effect. Thus, Slote and Portmore seem to just presuppose the self-other distinction rather than to establish it.

We need a better strategy to test the hypothesis. I offer one in the following section.

2.2 Testing the hypothesis

I propose to consider a pair of scenarios that are identical in all but one respect: one includes self-consent, while in the other a competent agent’s consent is somehow invalidated. Furthermore, because commonsense intuitions are so impressed upon us that any revision is set to appear counterintuitive, instead of checking our intuitions about the moral status of the agent’s action, we should consider our intuitions about the patient’s uptake. If, for example, an agent temporarily cannot exercise her capacity for consent, then people might judge it morally problematic at best, and possibly outright wrong, for others to exploit the agent’s sacrifice. If correct, this would offer indirect evidence of the role that the presupposition of consent plays for the commonsense morality thesis of self-other asymmetry.

Consider the twin of ‘Winning Raffle Ticket’: you hold a winning raffle ticket for a genuine autographed Babe Ruth bat, something you coveted for years and would forever cherish. You could give it away to a stranger who is sad but has no interest in baseball. The stranger’s sadness would quickly pass on its own. Imagine further that you get drunk and, as a result, end up offering your winning ticket to the sad stranger. Intuitively, it would be morally problematic for the stranger to take the ticket, for he can no longer presuppose that you truly consent to giving your winning raffle ticket away. If a bystander gives the stranger a dirty look, a response “What?! He gave it to me!” would appear disingenuous.

Equivalent scenarios trigger similar intuitions. It is morally objectionable to knowingly accept a sacrifice from a person who is brainwashed, confused, underage, or misinformed. The presence of conditions that typically render consent invalid also makes the otherwise intuitively permissible uptake of a sacrifice morally objectionable. Moreover, the bigger the stakes, the stronger our intuitions of impropriety. Think of a decision to join a military operation in someone else’s stead, signing away all earthly possessions in favor of somebody else, or engaging in sexual intercourse with a complete stranger. Most people would share the strong intuition that it is morally inappropriate to accept the offer if the beneficiary is aware of or even suspects a condition that invalidates consent.¹²

In view of the above, it is a mistake to interpret the self-other asymmetry as indicative of a morally relevant self-other distinction that, to boot, constitutes a fundamental feature of morality. Instead, asymmetrical moral appraisal is best explained by the presupposition of consent or, more precisely, self-consent. We believe that it is permissible for an agent to allow harm or risk of harm to befall herself or to sacrifice a significant benefit to herself in favor of someone else because we presuppose that in so doing she consents. This is evidenced by the fact that our intuitions switch or waver when there are reasons to doubt that the agent’s actions are truly hers, i.e., that her consent is valid.¹³

This conclusion is significant. It shows that we can accommodate the self-other asymmetry of commonsense morality but deny that this gives us reason to believe that equal moral concern does not extend over the agent’s interests, well-being, or respect for the agent’s will.¹⁴ For instance, the fact that a course of action would meaningfully take away from the agent’s well-being is a moral reason against that course. By implication from the principle of moral impartiality, this reason is as weighty as the reason against a course of action that would lead to an equivalent reduction from any other person’s well-being, other things being equal.

From this it follows that, contrary to a popular belief, moral considerations are relevant for purely self-regarding decisions. At the very least, those moral prohibitions, requirements and permissions that are plausibly thought of as grounded in the consideration of well-being or autonomy would potentially apply in the context of purely self-regarding behavior. Thus, the principle of moral impartiality establishes the presumption in favor of duties to oneself.

A critic might point out that, while the consent-based explanation of the asymmetry establishes the presumption of duties to oneself, it also makes said duties practically inconsequential.¹⁵ For if, say, harming oneself is indeed morally permissible because of an implicit valid consent, this just means that, generally, in moral justification, whenever a competent agent’s will goes against what is best for her, the consideration of the will is morally decisive. So, presumption in favor of duties to oneself notwithstanding, our moral landscape remains ultimately unchanged.

In response, one could point out that the objection mistakenly presumes consent to be trivial. Consent requires the right kind of mental state (e.g., Alexander 2014) which, on my preferred view, is the state of affirmative endorsement—a robust pro-attitude toward the situation to which an agent is consenting (Guerrero 2021). The endorsement is robust whenever the agent’s beliefs, desires, and values are relatively well integrated with one another. Robustness also indicates depth, thereby ruling out pro-attitudes one might have toward an artificially created single option such as ‘Your money or your life.’ (ibid., 353–54). Thus, if an agent voluntarily shoots herself in the leg because it is the only way to escape mandatory conscription to a criminal war her state is waging, it is incorrect to think that she consented to her leg being shot. This response, while promising, has its challenges, chief among them being that our deliberation about the nature of consent is often informed by considerations of policymaking in spheres such as business or medicine which have little to do with the purely self-regarding domain. Arguably, these considerations pull in the direction of a less demanding view of consent which, in turn, trivializes and possibly even discredits the idea of self-consent.¹⁶

In view of this, I am going to offer a different response, one which does not rely on a specific view about consent’s nature. Perhaps the objection rests not on the idea that consent is trivial but on the presumed analogy between the role that the self-regarding will of a competent person plays in the practical reasoning of others and its role in that person’s own practical reasoning. For others, the self-regarding will of a competent person is structurally relevant and indeed morally decisive in determining how they should act toward that person. Simply put, competent people typically have the last word in matters about their lives. And so, a critic might think, if an agent’s self-regarding will is to play the same role in that agent’s deliberation regarding her life, then whatever she chooses to do, her action bears the autonomy-conferred stamp of moral approval.

In the following section I will consider the roles that a competent person’s self-regarding will can play in the deliberation of others and argue that we cannot simply transpose these roles onto the purely self-regarding domain. To determine the role(s) of self-regarding will in an intrapersonal context, I turn to the value of autonomy. I consider what is involved in living autonomously and claim that, in the intrapersonal case, the will is typically a ground for morally decisive reasons whenever it coheres with the agent’s values and deep commitments. Thus, I tackle the objection by appealing to the reason we require consent, namely, respect for autonomous agency. However, since the reason why we require consent and what a valid consent involves are clearly related, the correct account of consent’s nature should be able to accommodate the disanalogy. Thus, one welcome implication of my argument is that it supports the view that consent requires a relative integration of one’s beliefs, desires, and values.

3. Deliberation and the Authority of the Will

It is possible to have both our asymmetric intuitions and the ideal that everyone matters equally for moral justification. To do so, it needs to be the case that (i) the former is to be explained by the presupposition of consent and that (ii) the consideration of a competent agent’s autonomy or will is always morally decisive in the purely self-regarding domain, which is why it trumps the consideration of the agent’s well-being.¹⁷ However, I argue that we should not accept (ii).

3.1 The structural relevance of the will

It is tempting to think that a competent person’s will is always morally decisive in the purely self-regarding domain. To begin with, this idea is motivated by the widely shared liberal conviction that rational beings should be free to live according to their own conceptions of the good and unburdened by interference from others. Furthermore, in our deliberation concerning how to act regarding another competent individual, the will of that individual concerning her purely self-regarding matters is typically morally decisive for us. It settles our deliberation, leaving no room for competing considerations. However, what this liberal conviction recommends when it comes to treating others cannot be simply transposed onto the intrapersonal domain. To see this, we need to consider the role or roles that a competent person’s will plays in our deliberation about how to treat them.

Another person’s will regarding her welfare can play two roles. First, the will is structurally relevant and, indeed, decisive in determining how others are permitted to act toward her. In this role, the will grounds a legitimate demand on others that they do not interfere with a person’s decisions, even if they plausibly judge that it would affect badly that person’s overall well-being. Structurally decisive reasons are insensitive to considerations of well-being. Second, in other-regarding deliberation, the self-regarding will of a competent person is substantively relevant, albeit nondecisive, in determining how to act. In this role, a person’s will is relevant because acting in accordance with it can derivatively or nonderivatively (or both) contribute to that person’s well-being. It can contribute derivatively whenever doing so promotes that person’s overall well-being and nonderivatively in virtue of the fact that autonomy is an important element of the good (Groll 2012).¹⁸

When we deliberate on how to act regarding another person, it is not unusual for us to gauge that person’s will in both above roles. However, in other-regarding deliberation about a competent person’s welfare, her will is typically structurally decisive.¹⁹ This explains why other-regarding reasoning appears to suggest that a competent person’s self-regarding will is always morally decisive. If a competent person’s self-regarding will is structurally decisive for others, why would it not be decisive for her?

The answer to this almost but not quite rhetorical question lies in what it means for a will to be structurally decisive. To respect another competent person by taking her will as the ground of structurally decisive reasons for action requires disregarding our beliefs about what would make her life go best. It is implausible to think, though, that self-respect requires one to disregard the consideration of one’s well-being in purely self-regarding decision-making. It does not make sense to say that one disrespects oneself if one’s purely self-regarding will is settled by one’s personal good. In first-personal deliberation about purely self-regarding matters, the will is typically settled by a consideration of well-being. In view of this, I believe that we need to consider the role of self-regarding will in intrapersonal moral reasoning without simply relying on what we believe we know about it from the inter-personal context.

3.2 Sovereignty, non-alienation, and respect for autonomous agency

We require consent out of respect for the autonomous agency of a competent person. Autonomy requires that people have the last word on matters concerning their lives. We attach value to living autonomously, i.e., to being a self-governing agent or a part author of one’s life; and one cannot be a part author of one’s life if one does not have the last word on at least some matters concerning one’s life, including purely self-regarding decisions and actions.

However, it is possible for one to have sovereignty but still fail to govern one’s life in the required sense. One could, for instance, be a slave to one’s passions, an addict, or suffer from false consciousness. In addition to sovereignty, autonomy seems to require that one’s decisions and actions are guided, in a non-accidental manner, by one’s values and deep commitments. Following Enoch (2021), I will refer to this as non-alienation.

Sovereignty and non-alienation are both valuable and together they capture what is important about the autonomous life of a rational being. Furthermore, they are clearly related. Often, even if not always, by respecting a person’s sovereignty we thereby enable that person to shape her life according to her values and deep commitments. More importantly, non-alienation and sovereignty are related in another, more fundamental way. On an independently plausible view, globally, it matters to us to have the last word on issues concerning our lives because it matters to us to live according to our values and deep commitments. This makes non-alienation more basic and explanatorily prior to sovereignty. To treat sovereignty as more fundamental than non-alienation, on the other hand, would be fetishistic.

This latter view has recently been defended by David Enoch (2021), who claims that while sovereignty can achieve local independence, globally, non-alienation is a more fundamental value.²⁰ Enoch first observes that sovereignty is mainly about the relations between oneself and others. As he puts it: “[...] having the last word on an issue [...] is mostly about others [...] not having the last word on that matter” (2021, 10). This makes sovereignty a less suitable candidate for a final value than non-alienation. Furthermore, considering an imaginary scenario can help us see that there is something fetishistic about taking sovereignty to be fundamental. A young researcher’s family situation requires them to be in a country X, where they try but fail to get a job. They could apply for a position at a university outside of X, but on the off chance that they got it, they correctly believe that they would not accept it as it goes against their commitment to their family. Given this, it would be irrational for them to invest more time and work in more applications to secure a position just to have an option to decline.²¹

The uncertainty makes all the difference. It is reasonable to put effort into opening new options if this can affect what choice one makes in the end. Once the uncertainly is removed, investing resources into creating new options is irrational. This shows that there is something fetishistic about treating sovereignty as fundamental. Globally, choice matters because of the contents of the choice.²²

In view of the above, I propose that respecting oneself as an autonomous agent requires two types of action. First, since autonomy is realized more fully whenever one’s life is in harmony with one’s values and deep commitments and because of one’s free choices, it always makes sense to resist the interference of others.²³ Second, and more importantly, in deliberating what to do, one should take guidance from one’s values and deep commitments. Free but systematically chaotic, irrational, weak-willed, self-destructive, slipshod self-regarding behavior, which shows no concern with the shape or direction of one’s life, fails to manifest the self-respect that is due to one as an autonomous agent (cf. Dillon 2022).

Where does this leave us regarding self-consent and the worry that the potential duties to oneself would be practically uninteresting? The worry dissipates, I think, once we recognize that the value of sovereignty has priority only in the other-regarding domain because it is essentially about not letting others dictate what one should be doing in matters concerning one’s life. Sovereignty is not all that matters for autonomy; it is not even the main thing that matters. Granting that, globally, non-alienation is more basic and explanatorily prior to sovereignty places us in a better position to understand what it means to fail to consent in the self-regarding domain. Acting contrary to one’s values out of fear or social pressure, weak-willed action, and acting while being confused about one’s motives are but a few examples of self-regarding behavior which arguably lacks the consenting mental state.

4. Conclusion

In this paper, I defended the idea that the prima facie plausible first-order principle of moral impartiality establishes the presumption in favor of the existence of duties to oneself. The challenge comes from commonsense morality which judges other-regarding and self-regarding conduct asymmetrically. While some philosophers take the asymmetry to indicate the existence of a morally relevant self-other distinction, I argued to the contrary. The asymmetry of intuitions is best explained by the presupposition of consent; hence, it does not constitute evidence for the self-other distinction.

If one’s will and well-being are equally as morally important as those of others, then, contrary to popular belief, moral considerations are relevant for purely self-regarding decisions. Those moral prohibitions, requirements and permissions that are plausibly thought of as grounded in the consideration of well-being or autonomy would potentially apply in the context of purely self-regarding behavior.

An opponent might object that, even if everyone matters equally for moral justification, this will change very little in our moral landscape. For we consent to, and thereby render morally permissible, everything that we do voluntarily. In response, I claimed that this objection, apart from mistakenly taking consent to be trivial, overlooks significant differences that exist between the other-regarding and self-regarding domains. Granted, in relation to another competent agent, we typically treat her self-regarding will as structurally decisive. However, in self-regarding deliberation, the role of the will as the ground of structurally decisive reasons for action does not make sense since one cannot be required to disregard the consideration of one’s own well-being.

To determine the correct role(s) of self-regarding will in intrapersonal practical reasoning, I turned to the consideration of individual autonomy. Individual autonomy is thought to require not only a life shaped by one’s choices (sovereignty) but also a life shaped by one’s values and deep commitments (non-alienation). In our dealings with others, the value of sovereignty takes pride of place, but non-alienation is globally more fundamental than and explanatorily prior to sovereignty. We show respect for our own autonomous agency not only and not primarily by resisting the interference of others but by minding our values and deep commitments, and by showing concern for the shape and direction our life takes. This suggests that valid self-consent requires nonaccidental coherence between the agent’s self-regarding decisions and actions and her values and deep commitments.

Although more is required to prove it, one possible implication of this paper’s argument is that commonsense morality gets some things wrong. Not all voluntary self-injurious actions are consensual, which opens the possibility that some such action may be morally impermissible. Most importantly, however, if the arguments in this paper are on target, the burden of proof is on those who want to insist that the moral code concerns only how we treat one another.²⁴

REFERENCES

Footnotes