Paul Schofield’s Duty to Self (2021) is an excellent contribution to recent moral philosophy. It is a much-needed addition to a literature that has, up until now, largely ignored the possibility of reflexive moral relationships. Thorough and challenging, the book is an indispensable read for students and scholars with an interest in ethics, metaethics, and political philosophy. In this review, I outline what I perceive to be the book’s main contributions and discuss some areas of concern about Schofield’s innovative framework. It should be noted from the outset that Schofield does not claim to tell us what we owe to ourselves but rather seeks to establish the possibility of moral duties to the self. As he insists: “The project is most concerned with the metaphysics of morals, rather than the content of morality” (p. 18).

In its traditional ‘social’ understanding, the term ‘moral’ is applied to our relationships with others and is only rarely applied to our relationships with ourselves. It is often taken for granted that the normative dimension of the self-relationship, how one should act towards oneself, is rather to be understood in terms of prudence. Schofield questions this assumption by focusing on moral duties to oneself. As we shall see, he is the first to offer a compelling way out of the objection raised by Marcus Singer in the early 1960s, according to which the notion of moral duties to oneself is incoherent. In a nutshell, Singer’s (1959, 1963) argument is that we cannot be bound by a duty if we are at any time free to release ourselves from it. We typically think of duties as binding because the power to release us from them lies with another individual. Since we are always free to release ourselves from what we owe to ourselves, Singer argues, there can be no genuine duties to the self. This is the ‘waivability objection’.

The first chapter, “On the Significance of Duties to the Self,” seeks to elucidate whether duties to the self can be moral obligations. Drawing on Stephen Darwall’s (2006) influential insights, Schofield argues that having a genuine moral duty to ourselves entails that we adopt a second-personal stance towards ourselves. So, the main issue is “whether a person can have duties second-personally directed to herself, of the kind had by a person standing on another’s foot” (p. 26). Singer’s waivability objection is based on the idea that we cannot relate morally to ourselves in this way. While using Singer’s objection as his target, Schofield sets aside the possibility of conceiving our moral relationships to ourselves on a monadic model of morality, like virtue ethics (p. 23).¹ This is where his approach is most ambitious: it aims to understand our moral relationships to ourselves and others on a unified model.

A first way to reply to the waivability objection is introduced in chapter 3 (“Defending Duties to the Self Part 1: Duties Across Time”). Schofield’s idea, roughly, is that we can understand our relationships to ourselves second-personally by making use of the notion of temporal perspective. To illustrate this idea, Schofield discusses the case of a smoker. Let us call him Alan. Why should we think that Alan owes it to himself to quit smoking? Schofield’s answer is that Alan can adopt a second-personal stance towards himself by paying attention to the legitimate demands that might be issued from one of his temporal perspectives. We can think, for instance, of his perspective in 20 years’ time. When so doing, Alan may be confronted with a conflict between his present and his future interests, i.e., the interests of his future perspective. Alan does not want to quit smoking now since he finds it enjoyable and is not suffering (yet) from any side effects. By contrast, judging from his future perspective, he should quit smoking, as continuing to smoke will be detrimental to his health in the long run. As Schofield explains, “these cross-temporal conflicts between various of a person’s interests enable us to tell a story about intrapersonal generation of duties that parallels in its essentials an interpersonal story” (p. 67).

The waivability objection does not threaten duties owed to others because the power to release us from our obligations lies in another’s hand. Understanding duties to ourselves in terms of obligations towards future perspectives allows reconsidering the idea that we can always release ourselves from such duties: we cannot when the power to do so lies with another temporal perspective. A question still remains: Does a person knowing that she will die soon from an incurable disease have no duty to herself? To analyze duties to ourselves as duties over time may not do justice to all these duties.

This limitation is addressed in chapter 4 (“Defending Duties to the Self Part 2: Duties at a Moment”), which seeks to establish the existence of duties to ourselves at a time. When considering duties at a time, Schofield introduces the notion of a practical perspective. The idea that we can occupy different practical perspectives stems from the observation that we all wear different hats: we are philosophers, parents, sportsmen and women, chess players, etc. Sometimes, our duties as philosophers conflict with the duties of another of our practical perspectives; for example, we wonder whether we should go to the chess club or grade philosophy papers. As Schofield puts it, “When an individual addresses another, a person might address herself from the perspective of one of her practical identities, issuing demands that will be received from a perspective associated with a different practical identity” (p. 107). Again, the conflict between different perspectives’ interests helps explain how we can have moral duties to ourselves at a time. We cannot release ourselves from our duty to go to the chess club because the power to do so lies with a different practical perspective, that of a chess player.

Schofield’s proposal is promising, but it also raises some concerns having to do with the notion of a person. Quoting Schofield:

It has been my aim to give an account of duties […] while maintaining that the person herself is the locus of moral value. But one might doubt whether I’ve succeeded in this. Talk of personal identities and of second-personal interactions between them will suggest to some readers a picture on which multiple “selves” constitute a person at a moment. (p. 124)

Indeed, the idea that we adopt several practical and temporal perspectives in our practical deliberations intimates a kind of division that may threaten the unity presupposed by the idea of a person: “We’re thus left to worry: What if the cost of gaining a second person within is losing the person altogether?” (p. 206).

Schofield goes on to answer this worry by insisting that the adoption of the second-personal standpoint only requires the capacity to occupy different practical or temporal perspectives. His proposal is not premised on the claims that one is composed of different selves interacting with one another at a time or that “person-stages, or time-slices” (p. 169) relate to one another over time. Perspectives are epistemic stances, not metaphysical entities, and it is always the person who relates second-personally to herself through the adoption of different perspectives. Schofield rightly emphasizes here a point of tension in his proposal, having to do with the notion of perspective. We might fear that by focusing on the task of telling us what these perspectives are not, Schofield leaves us wondering about what they exactly are and how we can access them. So, while his proposal is the most articulate response to the waivability objection to date, the key notion of a perspective remains somewhat elusive.

I suggested that some people may not have the kind of distant temporal perspectives required to ground duties across time. One may also fear that the appeal to practical perspectives is insufficient to explain why we have duties to ourselves at a time. Consider, for instance, a genius artist who is fulfilled by committing herself exclusively to her art or a monk who is perfectly content with his life of devotion. Arguably, these individuals do not have other practical perspectives, and there is thus no conflict of interest. Granted that they exist, can Schofield’s proposal acknowledge that people who are completely coherent, unified, or focused on the present are in a moral relationship with themselves? At first sight, the answer is “no” since his view presupposes the capacity to look at ourselves from a distance, or to see ourselves as someone else, so to speak. The same seems to be true of our moral relationships with other people: I have a moral duty to be nice to my friends even though I want to be nice to them anyway. We can have moral duties to the self and to others without conflicting interests.

A second worry is that Schofield’s focus on practical and temporal perspectives might be blurring the nature of the issue at stake. There might be alternative ways to gloss the necessary second-personal stance than by reference to temporal and practical perspectives. For instance, we sometimes regard ourselves second-personally when considering that what we did was morally wrong. The duty of self-respect may not be explained in terms of demands issued by one of our practical or temporal perspectives but only by reference to a moral perspective.

This brings me to my final point. Given its reliance on the notion of perspective, Schofield’s proposal may not have the resources to explain two prima facie central and interdependent aspects of the reflexive moral relationships: its relations to the aims of being happy and of becoming the best version of ourselves. Indeed, to which practical or temporal perspective do we owe the pursuit of happiness and of ethical or intellectual development? Given that these goods are desirable from any perspective, reference to perspectives seems here both unnecessary and insufficient in explaining why we have such duties to ourselves.² This is where Schofield’s model of moral self-relationship meets its limits. As I observed, Schofield is explicit that he does not seek to establish precisely what we owe to ourselves. Nevertheless, it seems reasonable to expect the metaphysics of the moral self-relationship to be consistent with how we intuitively conceive of the content of this relationship.

Chapters 5 and 6 of the book shift from the moral to the political sphere. Having shown how we can relate morally to ourselves, Schofield goes on to show that we can also relate politically to ourselves. In chapter 5 (“Might There Be Self-Directed Political Duties? Troubles for State Paternalism”), he lays out compelling reasons for thinking that we have no self-regarding political duties before arguing, in chapter 6 (“Defending Political Duties to the Self: The Possibility of Liberal Paternalism”), that we in fact have self-directed duties of right and justice and that paternalism is among the state’s functions. His aim, ultimately, is to challenge “a conception on which politics is for others exclusively” (p. 140).

In the seventh and final chapter (“Practical Philosophy After Duties to Self”), Schofield draws some important implications of his proposal, among which the fact that it should not be taken “as a mere addendum to whatever ethical theory is already in place” (p. 195). On the contrary, he insists that two difficult and hitherto ignored questions now arise. First, what to do when what we morally owe to ourselves conflicts with the moral interests of others? Second, what should we do when our own interests conflict with one another? These questions have been largely ignored so far, and practical philosophy will certainly be busy in the coming years trying to answer them. We should be grateful to Schofield for his stimulating proposal, which provides a stable foundation for future debates on these issues.

References