1. Note the vocabulary that seems so natural when characterizing a situation of this kind: there is a debt that is owed to another party, etc. This is the structure of relational normativity that will be central in the argument of this book.
1. I borrow here, of course, from Scanlon, What We Owe to Each Other. As I will discuss in chapter 5, Scanlon’s approach embodies a deeply relational conception of morality as a unified normative domain.
2. See Hohfeld, Fundamental Legal Conceptions. The Hohfeld-like structure that is important to the relational conception, as I shall develop it, is the basic connection between obligations and claims. Further elements in his scheme, such as permissions and powers, seem to me to be derivative from this basic connection, and to be determined by the specific shape or content of the claims and duties that relational morality defines. I have a moral permission to do X, for instance, just in case there is nobody who has a claim against me that I should do X, and also nobody who has a claim against me that I should not do X. Cf. Scanlon, “Reply to Wenar,” at pp. 401–2.
3. For an interpretation of tort law along these lines, see Ripstein, Private Wrongs.
4. See Thomson, The Realm of Rights, p. 117. In the same vein, see also Raz, The Morality of Freedom, pp. 170, 210–16; and O’Neill, “The Great Maxims of Justice and Charity,” pp. 224–33.
5. Cf. Foot, Natural Goodness, chap. 5.
6. See Scanlon, What We Owe to Each Other, pp. 6–7. In the same passage, Scanlon refers to this intermediate domain as the morality of what we owe to each other, which is in obvious alignment with my interpretation of the relevant requirements. Note, however, that there are relational requirements that arise in virtue of the special relationships we stand in to particular individuals, which I would not consider to be moral obligations in the sense here at issue. Interpersonal morality, on my account, is the domain of what we owe to each other just insofar as we are each persons, not insofar as we are friends, relatives, lovers, fellow-citizens, and so on.
7. See sec. 5.1 below for further discussion.
8. A pluralistic picture of this kind is suggested by Kamm in “Owing, Justifying, and Rejecting,” sec. 2; see also Frick, “Contractualism and Social Risk,” sec. 9. Kamm and Frick distinguish between moral requirements whose violation is wrong, and those whose violation wrongs someone. See also the works cited in note 4 above, which distinguish, somewhat differently, between the part of morality organized around rights and other moral requirements. (Raz’s pluralism, which goes together with skepticism about the interest or depth of any distinction between morality and other normative domains, is developed further in his Engaging Reason, chap. 11. Raz’s skepticism about what I call interpersonal morality might be compared to Foot’s in Natural Goodness, though it lacks the commitment to Aristotelian foundations that informs her position.) My own view agrees with those of Raz, Thomson, and O’Neill in taking rights to characterize only a subclass of requirements within the intermediate domain of interpersonal morality. Unlike these philosophers, however, I hold that there is a relational structure that is common to all requirements in this domain, and that lends those requirements coherence and unity.
9. The idea that there are extramoral forms of concern for moral persons is discussed in sec. 6.4 below. The dispute about whether to label these forms of concern “moral” or “extramoral” might appear merely verbal. But by the time they get to the final section of this book, readers will be able to see, I hope, that the considerations I label “extramoral” lack some of the defining normative features that unify the core requirements of interpersonal morality.
10. In keeping with this conception of the project, I have tried to keep the main text of this book as unencumbered as my scholarly conscience would tolerate, relegating many qualifications and details to the endnotes. The endnotes also contain extensive references to contemporary philosophical books and articles in which issues touched on in the text are discussed in greater detail.
11. Several examples are offered in sec. 3.2 below.
12. A stimulating general discussion of directed obligations, from which I have profited, is Thompson, “What Is It to Wrong Someone?”; see also May, “Directed Duties.”
13. There is a contrasting view, on which the directed duties generated by promises are unconditional, in the sense that having promised to do X, one owes it to the promisee to do X no matter what; but that these directed duties can sometimes be overridden by other kinds of consideration. See Thomson, The Realm of Rights, pp. 85–87, chap. 6; also Hart, “Are There Any Natural Rights?,” pp. 185–86; and Kamm, “Owing, Justifying, and Rejecting,” pp. 465–66. On this approach, the directed obligations defined by promises, though unconditional, are merely pro tanto or contributory; there is always, in principle, a further question to be answered about whether the promisor is morally obligated to do the thing that is owed to the promisee. The alternative that I favor treats these directed obligations as dispositive, albeit defeasible. I shall return briefly to the contrast between the two approaches in sec. 5.3.
14. For a summary discussion of this phenomenon, see Sreenivasan, “Duties and Their Direction.” Sreenivasan’s discussion is shaped by two assumptions that I shall eventually reject in what follows: that there are moral duties that are not directed; and that a directed duty might be overridden by other considerations in determining what it is that the agent ought to do, all things considered.
15. A theorist who apparently challenges the idea that directed moral obligations correspond to claims is Løgstrup; see his The Ethical Demand. On Løgstrup’s view, directed ethical obligations are grounded paradigmatically in natural relations of power and vulnerability between two parties. In the vocabulary I would prefer, these natural relations give the vulnerable parties personal interests that are apt to be affected adversely by what their powerful counterparts might do. But I differ from Løgstrup in thinking that the presence of such interests is not sufficient for a directed obligation; interests provide bases for directed obligations only when they also provide a basis for normative claims or entitlements on the part of the individual whose interests are at stake. (I am indebted to Robert Stern for stimulating exchanges about Løgstrup’s approach to directed moral obligation.)
16. See, for example, Sreenivasan, “Duties and Their Direction,” pp. 485–86. In the standard example of this phenomenon, A promises B that A will provide a benefit specifically to C; see, e.g., Hart, “Are There Any Natural Rights?,” p. 180. Theorists sometimes disagree with Hart about whether C has a claim against A that A provide the benefit in such cases. But few would hold that C has a claim if A’s fulfillment of the promise to B would benefit C merely accidentally.
17. Foot, Natural Goodness, pp. 47–51.
18. See Owens, Shaping the Normative Landscape, chap. 6. Strictly speaking, Owens talks of “bare wrongings,” not “bare claims”; but the possibility of bare wrongings presupposes that there are bare claims in the sense discussed in the text (i.e., claims that are not based in nonnormative interests held by the bearer of the claims, such as interests in avoiding harm or in securing the benefits of cooperation). Owens’s own view explains bare wrongings by appeal to our specifically normative interests in being able to create new obligations.
19. There is a gesture in the direction of these two aspects to the problem of the normativity of the moral in Scanlon’s distinction between the questions of the priority of moral reasons for the agent, and of the importance for other parties of the agent’s responsiveness to them; see Scanlon, What We Owe to Each Other, chap. 4. I find this distinction suggestive, but underdeveloped in Scanlon’s treatment of it in What We Owe to Each Other. Talk of the priority of moral reasons for the agent doesn’t precisely capture the distinctive features of them, as practical requirements; and the issue of importance is not connected clearly to interpersonal practices of accountability. Later work by Scanlon, however, including his Moral Dimensions, offers a more extensive treatment of the interpersonal significance of moral requirements.
20. Exemplary for both tendencies is Parfit’s brilliant and magisterial On What Matters, vols. 1 and 2. Parfit does not completely ignore either the obligatory character of moral rightness or its connection to responsibility and blame. But these topics are given perfunctory treatment (often in connection with discussions of Kant and other figures from the German tradition), and they are not taken to be central to an account of the normative structure of morality. Some examples: in vol. 1, Parfit notes that there is a “blameworthiness” sense in which actions can be wrong, but it does not figure prominently in his account of right and wrong in the core, “indefinable” sense (see vol. 1, secs. 22 and 54). Parfit also spends some time in vol. 1 arguing against the Kantian idea that people can deserve to suffer (see, e.g., vol. 1, pp. 257–72); he seems to associate robust practices of interpersonal accountability with this discredited idea, which perhaps helps to explain his tendency to downplay this interpersonal aspect of morality. In vol. 2, there is a discussion of normativity in the “imperatival” sense, in connection with some reflections on Nietzsche’s moral philosophy (see vol. 2, sec. 124). But Parfit dismisses imperatival conceptions of morality, arguing that they fail to distinguish normative claims from commands (a failure Parfit takes to have been prevalent culturally in Kant’s era). On the view I shall develop, by contrast, imperatival or voluntarist accounts may be understood as attempts to elucidate the character of moral norms as practical requirements (an issue that was clearly very important to Kant himself).
21. Issues of both of these kinds—about the role of morality as a source of obligations, and about its connection to practices of interpersonal accountability—are prominent in the recent work of Stephen Darwall; see, for instance, his The Second-Person Standpoint. Although I agree with Darwall about the importance of these issues for moral theory, the account I offer is ultimately very different from his. For relevant discussion, see my “Reasons, Relations, and Commands,” which argues that Darwall’s “second-personal” conception of morality slides between a relational account and a very different “voluntarist” model. In more recent work, Darwall makes clear that he understands obligation to be, at bottom, voluntarist rather than relational in character; see his “Bipolar Obligation.” I discuss the voluntarist approach in more detail in chapters 2 and 3 below.
22. One of the most impressive contributions of this kind is Nagel, The Possibility of Altruism. Some other (very disparate) examples of the general strategy include Gewirth, Reason and Morality; Hare, Moral Thinking; Gauthier, Morals by Agreement; and Korsgaard, The Sources of Normativity. For critical discussion of the general philosophical ambition to vindicate the authority of the moral, see Williams, Ethics and the Limits of Philosophy.
23. This conception of the challenge to moral theory is shared by at least some contemporary theorists. In addition to the works by Darwall referred to in note 21 above, see also Skorupski, The Domain of Reasons, part 3, which develops an intricate account of moral obligation and its connections to reasons for blame. The relational account I present differs from the accounts favored by both Darwall and Skorupski, as will emerge in later chapters of this book; but I share with them to some degree a conception of the normative features that we need a moral theory to illuminate.
24. These problems of resolution are illustrated, again, by Parfit’s On What Matters. Parfit discusses issues in metaethics extensively, devoting hundreds of pages to a defense of nonnaturalistic realism about reasons. (Parfit’s more recent On What Matters, vol. 3, is largely given over to additional extensive discussions of these same metaethical questions.) But the question of the specific normative significance of morality, as a source of obligations and a basis for accountability relations, is hardly mentioned, let alone addressed, in these pages. When Parfit turns to moral rightness and wrongness in particular, issues about their special normative significance are also invisible, and his argument zooms in closely to focus on first-order intuitions about specific cases (including intuitions elicited by an array of ingenious but highly artificial hypothetical examples). I don’t deny that first-order adequacy is a constraint on an account of moral rightness; but there are also theoretical constraints, deriving from the need to make sense of the special normative significance of moral rightness and wrongness themselves. I am trying to address this need directly, framing my investigation in terms that will keep these distinctive forms of normative significance clearly in focus throughout.
25. A conversation with Selim Berker helped me to see that pluralism about the normative is a commitment of the kind of view I favor.
26. In sec. 5.4, however, I do entertain the possibility of devising an illuminating (but nonreductive) philosophical account of the normative nexus at the heart of morality, considering moral contractualism as an account of this kind.
27. I have been deeply impressed by Scanlon’s contractualism since encountering his initial statement of it in his paper “Contractualism and Utilitarianism.” For an earlier attempt to come to grips with the theory, see, for instance, my “Scanlon’s Contractualism.” It has dawned on me only gradually that much of the power of contractualism derives from its articulation of a relational conception of the moral domain, and that this conception could in principle be extracted from the context of contractualism, and defended as a more general way of thinking about the basic normative features of the moral (with contractualism representing a specific theoretical elaboration of the more general relational conception). Scanlon himself, I should add, is sometimes diffident about the relational aspect of his view, and he certainly does not attach to it the significance that I see it as having. My argument, insofar as it defends contractualism, is therefore offered in a spirit of appropriation rather than faithful interpretation—one that takes much more seriously than Scanlon does his characterization of interpersonal morality as the domain of what we owe to each other.
28. The background assumption here might be that it is a constraint on any theory we might construct about morality and moral rightness that it accommodate our independent intuitions about first-order questions in this domain. The alternative that I favor is in the spirit of the kind of (wide) reflective equilibrium approach sketched by Rawls, which attempts to do justice both to our views about particular cases and to considerations of a more philosophical nature, such as the need to make sense of the characteristic deliberative and interpersonal roles that moral considerations are suited to play. See Rawls, A Theory of Justice, pp. 42–45.
29. For an influential example of a discussion in normative ethics that is couched in terms of intuitions about moral reasons and the comparative value of different outcomes, see the treatment of issues in population ethics in Parfit, Reasons and Persons, part 4.
30. Fred Neuhauser suggested to me that my argument might be framed in these terms; the suggestion seems to me apt.
31. A helpful recent discussion of this commitment to the “basic equality” of persons is Waldron, One Another’s Equals.
32. My reference to “most of us today” might be taken as an invitation to readers to reflect for themselves about whether they find the postulate of equal standing plausible, and whether they understand there to be requirements that are connected to the abstract insight it expresses. It may be that the community of those who share these assumptions, and view them as plausible starting points for ethical theorizing, is smaller than I perhaps optimistically take it to be. For a contrasting perspective on the characteristically modern, cosmopolitan conception of interpersonal morality operative in this book, see, e.g., Flanagan, The Geography of Morals. Flanagan usefully emphasizes the historical and cultural variety we encounter in understandings of selfhood and ethical attainment, and argues that we can profit, both individually and philosophically, from close attention to differences in ethical outlook across cultures and epochs. I do not disagree; but I also think there is value in working out systematically the modern idea that there is a common framework of relational requirements that are suited to structure our interactions with anyone, and that derive from our acknowledgment that their interests are no less important than ours.
33. See, for example, Singer, The Expanding Circle; and Greene, Moral Tribes.
34. Scanlon, “Contractualism and Utilitarianism,” p. 102. (Scanlon’s observation is, strictly speaking, about utilitarianism; but he would surely accept a similar claim about other forms of consequentialism.)
35. On the first issue, see, e.g., Sidgwick, The Methods of Ethics, pp. 497–98, and Rawls, A Theory of Justice, secs. 5–6; on the second issue, a classic treatment is Williams, “A Critique of Utilitarianism.”
36. See, for example, Parfit, Reasons and Persons, pp. 453–54; also the epigraph from Nietzsche’s The Gay Science, sec. 343, that appears at the start of that volume (“at last our ships may venture out again . . . ; the sea, our sea, lies open again; perhaps there has never yet been such an ‘open sea’ ”); a similar outlook is implicit in Greene, Moral Tribes.
37. Influential expressions of the view that the project of modern moral philosophy has failed are Anscombe, “Modern Moral Philosophy”; and MacIntyre, After Virtue. I discuss some of Anscombe’s concerns about moral obligation in chapter 2 below.
1. Anscombe, “Modern Moral Philosophy.”
2. See my paper “The Deontic Structure of Morality.”
3. Here and in what follows, I shall mostly use the expressions “practical requirement” and “obligation” interchangeably, to refer generically to normative considerations that function deliberatively as presumptive constraints on agency (as I shall go on to put it). The eventual argument of the chapter will be that moral obligations, in the more specific sense of directed duties, are among the normative considerations that are intelligible to us as obligations or practical requirements in this more generic sense.
4. See Scanlon, What We Owe to Each Other, chap. 1, for the idea of reasons as considerations that count in favor of attitudes. For the idea that normativity is to be understood fundamentally in terms of reasons, see Raz, Engaging Reason, chap. 4, at p. 67; also Skorupski, The Domain of Reasons.
5. See Scheffler, “Relationships and Responsibilities,” p. 100.
6. See Raz, Practical Reason and Norms.
7. See my “The Deontic Structure of Morality”; also Dancy, “Enticing Reasons”; and Greenspan, “Making Sense of Options,” sec. 3.
8. The classic statement of this view is Bratman, Intention, Plans, and Practical Reason.
9. Cf. Owens, Shaping the Normative Landscape, secs. 15–16; Owens assimilates conscientious responsiveness to obligations to motivation through habit, and attempts to make sense in these terms of some of the distinctive deliberative roles that I am calling attention to.
10. See, again, Bratman, Intention, Plans, and Practical Reason.
11. The general idea that there might be a plurality of different kinds of normative consideration is familiar from the work of Jonathan Dancy. See, for example, his Ethics without Principles, chap. 3, where Dancy contrasts the favoring relation with other forms of normative relevance, such as intensifying and enabling.
12. Cf. Thomson, Normativity, pp. 229–30. Thomson’s account is expressed not in terms of reasons and their comparative strength, but in terms of the different considerations that on her view ground “directive” claims about action. Those considerations have to do with being defective as a human being; the difference between directives expressed with “ought” and with “must,” on her account, can be traced to the severity of the defects that one would manifest in failing to act on the directives. But this is a matter of differences in degree, not differences in kind or quality.
13. Different versions of this approach are developed in Korsgaard, The Sources of Normativity, chap. 3; in Williams, Ethics and the Limits of Philosophy, chap. 4, and “Moral Incapacity”; and in Frankfurt, Necessity, Volition, and Love. For Korsgaard, the practical “must” expresses a class of genuinely normative considerations that are grounded in our “practical identities,” and that are aptly described as obligations. For Williams and Frankfurt, by contrast, it gives expression to psychological incapacities that in turn reflect central aspects of who we are; these authors tend not to speak of obligation in this context, but of practical or volitional necessity. For an illuminating discussion of volitional necessity, see Watson, “Volitional Necessities.”
14. See, e.g., Korsgaard, “The Normativity of Instrumental Reason.” This general approach could also be interpreted in terms of the voluntarist model, insofar as the constitutive condition of thought and action is taken to be a commitment to comply with the core requirements of rationality (where the commitment in question is understood as a kind of law that one legislates for oneself); cf. the brief discussion of Kantian versions of voluntarism about morality in sec. 2.2 below.
15. See, for instance, Kolodny, “Why Be Rational?”
16. It might be argued that one could be akratic with respect to these requirements, acknowledging their significance for one’s conception of who one is, but failing to live up to them on account of temptation or weakness. But it is hard to see much space for genuinely acknowledging something as existentially central to one’s conception of what one’s life is about while still being weak-willed in this way. The subjective conception of significance that is at work in the account seems to preclude this kind of backsliding. For Korsgaard, e.g., a practical identity is supposed to be a “description under which you find your life to be worth living and your actions to be worth undertaking”; see The Sources of Normativity, p. 101. (Korsgaard’s own attempt to accommodate the phenomenon of flouted duty is found on pp. 102–5 of this work. Her strategy appeals to the agent’s sense that occasional violations of the alleged duty are consistent with maintenance of the practical identity at issue; but this consideration seems to me to undermine the standing of the normative consideration as a genuine duty on the occasions when it is flouted.)
17. There is a different kind of identity-based strategy, which appeals not to agents’ conception of who they are, but rather to objective facts about identity of which the agents might be deeply mistaken. This kind of approach would not encounter the problem of flouted duty, though it is vulnerable to problems of other kinds. (The Aristotelian approach sketched in the following section might be understood as an identity-based strategy of this objective kind.)
18. Scanlon, “Wrongness and Reasons”; “Replies,” pp. 435–39; and Parfit, On What Matters, vol. 1, pp. 368–70; also sec. 22.
19. See especially Scanlon, “Wrongness and Reasons.”
20. See again Scanlon, “Wrongness and Reasons,” for a forceful presentation of this concern.
21. Compare the notion of morality in the narrow sense developed in Scanlon, What We Owe to Each Other, chap. 4, sec. 7. For an interesting expression of skepticism about the idea that there is a significant context-independent distinction to be drawn between moral and other kinds of reasons, see Raz, Engaging Reason, chaps. 11–12. A similarly skeptical attitude seems to me implicit in Williams’s critical reflections on what he calls “the morality system”; see his Ethics and the Limits of Philosophy, esp. chap. 10.
22. On the concept/conception distinction, see Rawls, A Theory of Justice, p. 5. On my understanding, a conception of moral rightness is an attempt to identify a high-level property that rightness might be taken to consist in, and that makes sense of the more abstract features that define the modern concept of the moral right.
23. In what follows, I shall (unless otherwise noted) use variants of the expressions “moral” and “morality” to refer to specifically interpersonal morality, in the sense I have roughly sketched here.
24. Nagel, The Possibility of Altruism, p. 14.
25. See Parfit, Reasons and Persons, p. 27, for this way of using the vocabulary of “agent-neutral” and “agent-relative” in relation to ethical theories. Compare the notion of an “agent-centered” restriction introduced and discussed in Scheffler, The Rejection of Consequentialism, e.g., p. 80; as Scheffler makes clear there, an agent-centered restriction in his sense need not involve an absolute prohibition on an agent’s performing actions of a certain specified type.
26. The discussion here might be considered an homage to Anscombe’s brisk critical survey of modern accounts of moral obligation in “Modern Moral Philosophy,” pp. 170–73. Where Anscombe devotes a couple of sentences to each approach she discusses, I offer a couple of paragraphs, in deference to contemporary scholarly expectations.
27. Reflecting on this aspect of their position, utilitarians might accept that theirs is a revisionist conception of the moral (perhaps by denying that agent-relativity is as important to the modern concept of the moral as I have contended). Alternatively, they might attempt to accommodate this element in the modern concept by showing that individuals will generally do better, by the lights of the utilitarian standard of right conduct, if they treat it in practice as giving them agent-relative goals (so that they strive to keep their own promises, e.g., rather than to promote the neutral value of promissory fidelity).
28. I have suggested that any reasonable conception of practical requirements should construe them as merely defeasible constraints on agency, insofar as circumstances can change unexpectedly in ways that alter what we are required to do. But defeasibility is a matter of degree. The difficulty with the utilitarian conception is not that it defines requirements that are defeasible, but that it renders moral rightness hostage in an exceptional degree to circumstantial fortune of this kind.
29. It is sometimes suggested that maximizing is constitutively rational, and that rationality intelligibly functions as a constraint on the formation of our intentions. But maximizing is crucially unlike the core elements of structural rationality that, as I suggested in sec. 2.1, intuitively strike us as constraints on deliberation. It is, for instance, difficult to assimilate it to the dominance and identity models that have proven helpful in thinking about demands of noncontradiction and means-end consistency. Moreover, it is controversial (to say the least) to assume that maximizing is a constitutively rational ideal.
30. A conclusion that at least some consequentialists have been happy to embrace; see, for instance, Norcross, “Reasons without Demands.”
31. Mill, Utilitarianism, pp. 47–48.
32. These arguments do not apply to the most prominent contemporary versions of rule consequentialism, such as those defended by Parfit in On What Matters, vol. 1, or by Hooker in Ideal Code, Real World. To the extent these approaches offer accounts of the obligatory character of morality, their accounts do not appeal to the idea of maximization.
33. Anscombe, “Modern Moral Philosophy.” The Aristotelian alternative that she recommends for contemporary ethics is one that does without the special notion of a “moral” obligation; I discuss this approach at the end of the present section.
34. For a sketch of such a view, see Wolf, “Moral Obligations and Social Commands.”
35. Wolf addresses these concerns in “Moral Obligations and Social Commands,” pp. 360–64. She notes that social commands might be said to give rise to obligations only when the behavior they command is morally valuable in some way. But this stipulation still leaves us with the problem of what she calls “false negatives,” which are apparent obligations that a given society neglects to enforce through the exertion of social pressures. The question of whether we are under a moral obligation to offer reparations to the descendants of slaves is not settled, as it seems to me, by observing that we do not as a society command each other to act in this way. Consideration of such cases moves Wolf to suggest that moral obligations might be determined not by what a society actually commands its members to do, but by the commands that it would be appropriate for a society to lay down, which moves the position in the direction of the hypothetical version of voluntarism that I go on to consider in the text.
36. Darwall, The Second-Person Standpoint; see also his “Bipolar Obligation.” Darwall understands the issuing of demands by representative members of the moral community to involve their being addressed to an agent through the reactions characteristic of blame. Two other approaches that interestingly tie moral obligation to critical reactions in the register of blame are Greenspan’s in “Making Room for Options,” sec. 4; and Skorupski’s in The Domain of Reasons, chap. 12.
37. See Darwall, “Bipolar Obligation,” pp. 36–37. For a more general discussion of some of these issues, see my “Reasons, Relations, and Commands.” I return to the question of the relation between obligation and responsibility relations in the next chapter; I argue, in sec. 3.4, that the voluntarist cannot do justice to the reasons we have to hold people to moral demands, which have to do with the fact that they define moral obligations in some sense that is independent of their being addressed by members of the moral community to each other.
38. Cf. Watson, “Morality as Equal Accountability,” p. 40.
39. For an overview of this historical development in the moral philosophy of the modern period, see Schneewind, The Invention of Autonomy. Freud, of course, also interpreted moral development as a process involving the internalization of social relations of authority.
40. A contemporary defense of this general approach is Korsgaard, The Sources of Normativity. The commentaries included in this volume raise versions of the questions I have briefly posed in the text for the general Kantian strategy. See, in addition, Street, “What Is Constructivism in Ethics and Metaethics?”; also my “Constructivism about Normativity.”
41. See Foot, Natural Goodness; also Thompson, Life and Action.
42. The conception of identity relevant to this development of the identity-based approach to obligation is different from that emphasized in sec. 2.1 above. It is the conception of one’s objective species nature, not a subjective conception of who one fundamentally is or what one’s life is about.
43. See Foot, Natural Goodness, p. 79.
44. Proponents of Aristotelian approaches are typically skeptical about the modern concept of moral obligation that I have been discussing in this section, preferring a framework that attaches little systematic significance to the distinction between moral and nonmoral reasons and requirements. Theorists of this stripe will be fairly untroubled by their inability to do justice to a concept whose coherence or importance they are anyway inclined to question. But the attractions of Aristotelianism as a theoretical framework for ethics may be diminished somewhat if there is an alternative approach that makes sense of the modern notion of moral obligation that Anscombe and others have rejected.
45. Foot, Natural Goodness, p. 48.
46. Her explanation for this dissatisfaction is that other human virtues, such as resourcefulness and persistence in pursuit of the truth, would lead a scholar such as Maklay to want to photograph the Malay servant for scientific purposes, in a way that is at odds with the force of the promissory reason; see Natural Goodness, p. 50. But there is a natural response to this worry within the framework of Foot’s approach, which is that the resourceful and persistent commitment to the truth is not a virtue if it leads an agent to act in ways that are incompatible with other virtues, such as trustworthiness. (This response would reflect a commitment to the thesis of the unity of the virtues, on at least one natural interpretation of it.)
47. Foot, Natural Goodness, p. 51 (emphasis mine). Foot draws here on Anscombe, “Rules, Rights and Promises.”
48. I return to the idea that directed duties are the joint property of the parties that they bind in sec. 4.2 below.
49. See, for instance, Feinberg, “The Nature and Value of Rights,” pp. 243–44; also Darwall, “Bipolar Obligation,” pp. 25–27.
50. See Scheffler, The Rejection of Consequentialism, chap. 4, for an especially forceful presentation of this puzzle.
51. Inducing the mysterious stranger to bring about the state of affairs in which promissory fidelity is maximized would count as my doing what I could to advance this agent-neutral goal, and so be a way of fulfilling my promise to you.
52. I offer some further comments about agent-relativity in sec. 4.4 below.
53. Proponents of different theories might understand them to be accounts of the modern concept of moral rightness, because they are attempts to do justice to the basic elements of that concept. But if the relational account is the most plausible such theory, then those who reason correctly about moral rightness will understand it in terms of the property of what we owe to each other, just insofar as we are persons.
54. Cf. Raz, Practical Reason and Norms.
55. I assume here provisionally, as it seems to me most natural to do, that these are claims about what it is morally right or wrong to do, in the relational sense I have been trying to explicate. As I explain in sec. 4.1 below, however, promissory obligations are sometimes understood to involve a distinctive, pre- or nonmoral form of normativity that is implicitly relational in structure. On this kind of view, there is a promissory form of rightness, involving self-standing obligations that are owed to another party, that might be registered in deliberation as a presumptive constraint on agency.
56. See, for example, Smith, The Moral Problem, chap. 3. Cf. Parfit’s discussion of “deontic reasons” in his On What Matters, vol. 1, pp. 448–51; also Owens, Shaping the Normative Landscape, p. 69: “[obligations] are not factors in the practical deliberations of the conscientious agent.” As noted earlier, Owens’s alternative assimilates responsiveness to obligations to motivation by habit; but this seems to me to give obligation too little work to do in the perspective of deliberation. (The conscientious agent should be aware of a feature of obligations that makes them suited to figure as defeasible constraints in reflection about what to do.)
57. The original buck-passing theory is Scanlon’s account of goodness, developed in chap. 2 of his What We Owe to Each Other. For an application of this terminology to debates about the normative significance of moral rightness, see Darwall, “ ‘But It Would Be Wrong.’ ”
58. This fetishism objection is pressed by Smith in The Moral Problem, sec. 3.5.
59. This point is helpfully emphasized by Darwall in “ ‘But It Would Be Wrong.’ ”
60. I believe this gives rise to a further potential problem for voluntarist approaches, concerning the content of the attitudes through which responsibility is attributed by one party to another (e.g., by the claimholders who address moral demands to agents). I return to this issue in sec. 3.4 below.
61. Private law systems of right figure prominently in Michael Thompson’s illuminating discussion of relational or (as he calls it) “bipolar” normativity, in “What Is It to Wrong Someone?” The idea that there are distinct “orders of right” (or systems of relational obligations and claims) plays a central role in the argument of Thompson’s paper; I discuss one strand in that argument in sec. 4.2 below.
62. To be clear, I am not appealing here to idea that the relational requirements of morality are obligations for us, because we are constitutively committed to complying with them in virtue of our fundamental identity as persons. There is an identity-based argument for moral requirements that takes this general form, but it is not the argument I am here sketching. My idea, instead, is along the following lines: if there are requirements that we owe to others, just insofar as they are persons who might be affected by our actions, then these requirements will be inescapable for us, because we necessarily occupy the situation that triggers them. Relational moral requirements differ, in this respect, from the relational requirements of certain games.
63. This claim is defended in my “Duties of Love.”
64. See my “Scanlon’s Contractualism,” sec. 3, for a discussion of the relational values enabled by both morality and friendship. The phrase “interpersonal recognition” is inspired in part by Scanlon’s talk of “mutual recognition” in What We Owe to Each Other, chap. 5. Scanlon understands mutual recognition to constitute the valuable form of relationship that is enabled by morality, and that helps to explain the reason-giving force of moral requirements. It strikes me as potentially misleading, however, to characterize the relational value at issue in these contexts as mutual recognition, since the value at issue can be realized even in cases in which the agent’s acknowledgment of the claims of another party is not reciprocated. I return to this issue in sec. 4.4 below.
65. A different possibility might be one in which a single individual has two distinct claims against the agent that cannot in the nature of the case both be satisfied. But this seems less plausible, as a model for intramoral conflict of obligation. The fact that the allegedly conflicting claims are assigned to a single person threatens to undermine their force, as self-standing entitlements. Thus prospective claimholders would not seem entitled to complain about my behavior if I have done what was necessary to live up to other claims that they themselves have against me.
66. I noted in sec. 2.1 above that obligations are typically registered in deliberation through the formation of intentions, which operationalize their function as presumptive constraints on agency. But it is a familiar fact that many of our intentions have a hierarchical structure, establishing clear relations of priority between our more fundamental goals and those subordinate to them. On this general point, see Raz, The Morality of Freedom, pp. 292–93; also Scanlon, What We Owe to Each Other, pp. 121–23.
67. This idea is interestingly defended by Scheffler, “Membership and Political Obligation.”
68. See (again) my “Duties of Love.” For a recent discussion of friendship that emphasizes its source in values distinct from the moral, see Nehamas, On Friendship. Also highly relevant here is Niko Kolodny, “Love as Valuing a Relationship.”
69. For this suggestion, see Scanlon, What We Owe to Each Other, pp. 164–66. This line of thought would assimilate friendship to the treatment of political membership sketched above, according to which the special values at issue in the two cases are conditioned by or subordinated to the requirements of morality. A different way to assimilate the cases would be to treat political membership, like friendship, as involving an inherent susceptibility to generate independent requirements that might conflict with those of morality.
70. See Bernard Williams, “Moral Luck,” for the discussion of Gauguin; and “Persons, Character, and Morality,” for his treatment of the compatibility of impartial moral obligation with the attitudes characteristic of intimate relationships.
71. We might put this by saying that the relational obligations of morality are pervasive, even if they do not necessarily override the other kinds of obligations with which they might conceivably conflict. For a thoughtful discussion of the pervasiveness and alleged overridingness of morality, see Scheffler, Human Morality, chap. 2.
1. They are not denying that it is a platitude that moral rightness is a source of obligations; to do that would be to offer an alternative interpretation of the abstract concept of moral rightness, rather than the kind of skeptical position I am sketching in the text.
2. I develop some of these themes at greater length in my paper “Rightness and Responsibility.”
3. Strawson, “Freedom and Resentment,” pp. 79–80.
4. Strawson, “Freedom and Resentment,” pp. 79–80.
5. I discuss this and other aspects of Strawson’s position in my paper “Emotions and Relationships.”
6. See my Responsibility and the Moral Sentiments, chap. 2.
7. The idea that accountability relations involve a form of “moral address,” whereby demands or expectations are addressed by one agent to another, is an important and interesting theme in the work of Gary Watson on responsibility. See, for instance, Watson’s “Responsibility and the Limits of Evil,” and also his “Two Faces of Responsibility.” I discuss this theme in my paper “Moral Address.”
8. A recent collection of papers on this topic, which reflect the complexity of the phenomenon and the range of approaches that may be taken to it, is Coates and Tognazzini, Blame.
9. See my “Dispassionate Opprobrium.” For a similar view, see Wolf, “Blame, Italian Style.” The reactive approach is, I believe, implicit in Strawson’s “Freedom and Resentment”; it is also a theme in my Responsibility and the Moral Sentiments.
10. On the importance of adjustments in one’s ways of relating to people who have acted wrongly, see Scanlon, Moral Dimensions, chap. 4.
11. A still different aspect of the distinctively modern concept of interpersonal morality, to which I shall return, is its commitment to the postulate of the equal standing of all.
12. For a defense of the idea that intentions or attitudes are generally irrelevant to questions about the permissibility or wrongness of what we do, see Scanlon, Moral Dimensions, chaps. 1–3. For purposes of discussion in this chapter, we do not need to insist that permissibility is in all cases independent from questions of intent or attitude, only that it is in many cases independent. For a penetrating discussion of this general issue, see Kolodny, “Scanlon’s Investigation.”
13. For all I have said here, these actions might sometimes include purely mental performances, such as a failure to manage one’s malicious or antagonistic attitudes toward others in accordance with reasonable expectations. For discussion of this possibility, see (again) Kolodny, “Scanlon’s Investigation.”
14. I would emphasize that the attitudes relevant to blame reactions must be construed widely, to include such things as thoughtlessness, lack of due care or consideration, negligence, lack of attention, and so on. These sorts of characteristics might be understood as global features of a person’s attitudes, and they sometimes render blame reactions warranted or fitting.
15. This point is an important theme in recent work on accountability by Pamela Hieronymi; see, for example, “The Force and Fairness of Blame.”
16. I have elsewhere suggested that traditional forms of global skepticism about responsibility may best be understood as appeals to a concern about the fairness of holding people responsible, given the implication of our actions in a larger world of natural causal processes; see my Responsibility and the Moral Sentiments, chap. 4. Standards of fairness govern actions rather than attitudes; they bear on our accountability practices, then, at the moment of reflection described in the text, where we acknowledge our implication in the economy of disesteem, and face a practical question about whether to affirm or to renounce it, as a basis for our continuing interactions with the persons who are its target. I return to this theme in sec. 3.3 below, in connection with the question of the nature of forgiveness.
17. A very different basis for distancing ourselves from participation in the economy of disesteem might be that doing so is counterproductive, serving as an obstacle to the establishment of a more just social order, going forward. This might conceivably be the case, even if it is granted that wrongdoing makes it fitting to respond with reactive or other forms of blame. For a forceful (if, to my mind, unconvincing) statement of this critical position, see Nussbaum, Anger and Forgiveness; also relevant here is Flanagan, The Geography of Morals, chaps. 8–9. A more positive assessment of the familiar practice of angry blame is offered by Miranda Fricker in her “What’s the Point of Blame?”
18. Foot, Natural Goodness.
19. See, e.g., Singer, Practical Ethics, pp. 213–15.
20. Cf. Sidgwick’s sympathetic discussion of the possibility of an esoteric morality, in his The Methods of Ethics, book 4, chap. 5.
21. By the same token, a conscientious effort to comply with the utilitarian principle will notoriously not suffice to provide protection from the opprobrium of others. We can easily imagine circumstances in which it might be optimific, and therefore required by utilitarian lights, to react to such a conscientious agent with the reactions characteristic of blame.
22. I offer some additional remarks about utilitarian or consequentialist conceptions of rightness, and their significance for interpersonal accountability, at the start of sec. 3.3 below.
23. It is not in general true that the traits constitutive of virtues are responses solely to the reasons provided by the fact that those traits are virtuous. People who are just, for instance, correctly take their promissory commitments to be serious reasons for action, and respond to those reasons appropriately. In the present case, however, it seems there is no independent account on offer of what the first-order reasons for blame might be, which leaves us only with the reasons that are provided by the fact that such reactions are virtuous; these reasons, I maintain, are reasons of the wrong kind. To anticipate the argument that follows: if one could take for granted a relational account of interpersonal morality, then one could argue that virtuous agents respond correctly to the reasons for blame that are provided by deliberate offenses against relational requirements. But the account of ethical virtue, on this approach, would not be an individualistic alternative to the relational story, but a supplement to it.
24. Note, again, that for all I have said here, these blameworthy performances might include some purely mental actions, involving failure, e.g., to manage malicious or derogatory emotions and feelings toward another person.
25. There are versions of both perfectionism and consequentialism that might come closer to accommodating the point I have emphasized about the social significance of rightness. Consider rule consequentialism, in the version that holds that actions are wrong just in case they would be prohibited by general rules that it would be optimific for the members of community to internalize and to follow (where internalization precisely involves a tendency to blame people when they exhibit attitudes of indifference toward those general rules). This theory seems to do a better job than the perfectionism considered in the text at acknowledging the idea that moral rightness needs to be understood in relation to interpersonal practices of accountability. But it still gives a distorted account of this relation. Reactive and other forms of blame are directly responsive to reasons, construed as considerations that make them individually fitting or warranted, whereas the rule consequentialist appeals to evaluative considerations of a very different kind, about whether it would generally be optimific to hold agents responsible for complying with a given set of requirements. For a recent statement and defense of the kind of rule consequentialism under discussion here, see Hooker, Ideal Code, Real World.
26. An impressive recent statement of this general approach is Adams, Finite and Infinite Goods.
27. See Adams, Finite and Infinite Goods, pp. 252–53 and 257, on this point. On this way of developing the divine command theory, the obligations created by determinations of the divine will ultimately trace back to relational duties of gratitude toward God, which are prior to and independent of the issuance of commandments by God. So interpreted, the divine command theory is not an alternative to the relational account of interpersonal obligation, but presupposes it. I return in sec. 3.4 below to the ways in which voluntarist approaches distort our reasons for complying with moral obligations and addressing them to other parties through our accountability practices.
28. Cf. Watson, “The Trouble with Psychopaths,” pp. 315–16.
29. Fricker, “What’s the Point of Blame?,” p. 173.
30. Watson, “The Trouble with Psychopaths,” p. 316.
31. Hieronymi, “Reflection and Responsibility,” p. 31.
32. For a historical example of the same basic point, compare the following characterization by Adam Smith of the object of such reactive attitudes as resentment: “To bring him [sc. the wrongdoer] back to a more just sense of what is due to other people, to make him sensible of what he owes us, and of the wrong that he has done to us, is frequently the principal end proposed in our revenge, which is always imperfect when it cannot accomplish this”; see Smith, The Theory of the Moral Sentiments, p. 115 (part II, sec. III, chap. I).
33. Strawson, “Freedom and Resentment,” p. 77.
34. Strawson, “Freedom and Resentment,” p. 80.
35. Strawson, “Freedom and Resentment,” p. 84.
36. For a more detailed development of these points, see my “Emotions and Expectations.”
37. I return to this theme in sec. 4.4 below.
38. Many culpable failures of causal reasoning can be traced to our natural tendency to overweight our own interests within moral reflection. Flawed reasoning of this kind is often strategically rooted in partiality; the faulty empirical reasoning of most climate change skeptics, for instance, seems clearly to be motivated by their concern to protect their own narrowly economic and political interests. (It is in this way a form of ideology, in the pejorative sense.) Cases of this kind could therefore be described as ones in which agents disregard the consequentialist requirement to attach equal weight to the interests of those potentially affected by their actions.
39. This aspect of apology is emphasized in Helmreich, “The Apologetic Stance.”
40. Recent works that take up these issues include Allais, “Wiping the Slate Clean”; Garrard and McNaughton, “In Defense of Unconditional Forgiveness”; Griswold, Forgiveness; Hieronymi, “Articulating an Uncompromising Forgiveness”; and Pettigrove, Forgiveness and Love.
41. This claim is a minor exaggeration, insofar as there are contexts in which we seem to attribute the standing to forgive to certain third parties. Thus, close friends of the victim might intelligibly say to the perpetrator of a serious wrong that they will never forgive the perpetrator for what was done to the victim. This kind of standing, however, is arguably still positional, insofar as it presupposes that the person who possesses it stands in a special relation of some kind to the party who was wronged. Third-party forgiveness is not exactly bestowed on behalf of the victim; but it proceeds out of recognition of the victim’s privileged entitlement to complain, which may be taken to extend to the third party in virtue of their special relationship to the victim. For an interesting discussion of this elusive and complex phenomenon, see Chaplin, “Taking It Personally.”
42. See Scanlon, Being Realistic about Reasons, chap. 2, on “pure normative claims,” which often articulate relations of dependency between normative and nonnormative facts or circumstances.
43. It has been argued that there are cases in which a person is wronged by an action, without having a right or claim against the agent that is directly flouted through the action. See Cornell, “Wrongs, Rights, and Third Parties.” On Cornell’s view, the cases at issue involve wrongful actions that cause adverse effects for parties who do not have rights or claims against the agent. But the notion of wrongful action in play here is relational rather than individualistic, insofar as Cornell assumes that someone’s rights have been violated by the action that is wrongful; the suggestion is that such actions wrong not only the agent whose rights are violated, but also other parties who are adversely affected by those actions. I explain in sec. 6.1 why I do not find this proposal plausible; for the present, the important point is that, though it attaches significance to the fact that third parties are harmed by wrongful actions, it does this within a relational rather than an individualistic framework.
44. See Ripstein, Private Wrongs, chap. 8.
45. This may ultimately be what Ripstein means with the somewhat misleading talk about rights surviving their own violation. See, e.g., Private Wrongs, pp. 240–41, where he writes, “The thought that something can be repaired does not mean that damage and repair is just as good [as no wrongdoing at all]; it means that whatever can be done to repair a wrong must be done because it is a wrong that requires repair.” See also pp. 249–51, where Ripstein argues, against a proposal due to John Gardner, that the duty of repair has a directional element that cannot be explained merely in terms of the survival of reasons for action that originally counted in favor of the primary duty that was breached by the tortfeasor. These ideas are very close to what I have in mind in characterizing the secondary duty of repair as a residue of a claim that was flouted by the original wrong.
46. Scanlon talks in this connection about the “impairment” of the agent’s relationships through acts that disregard moral claims, in his Moral Dimensions, chap. 4. I find this somewhat misleading, since it implies that there is an antecedent relationship between the agent and the claimholder that is eligible to be impaired. I come back to this issue in secs. 3.4 and 4.4 below.
47. Cf. Fricker, “What’s the Point of Blame?,” where blame is characterized as serving to bring about a convergence in moral understanding. If we are looking for a constructive function of blame, however, it seems to me important to emphasize convergence of understanding within the nexus defined by relational requirements. It is through blame and remorseful acknowledgment of wrongdoing that the relational rupture constituted by wrongdoing can be repaired, as the parties converge in a shared affirmation of what they owe to each other.
48. See Wolf, “Moral Obligations and Social Commands.”
49. An analogous point applies to divine command theories that emphasize the sanctions attached by God to failures to comply with divine law. Helpful in this connection is the discussion of the “sanction theory” of obligation in Owens, Shaping the Normative Landscape, sec. 13.
50. See Wolf, “Moral Obligations and Social Commands”; and Adams, Finite and Infinite Goods, chap. 11.
51. See Adams, Finite and Infinite Goods, pp. 252–53 and 257, on the primacy of the relationship to God in a divine command theory of obligation. For Adams, the primary reasons that are associated with obligations are reasons of gratitude to God, rather than reasons associated with the avoidance of divine sanctions. One potential problem for this approach, discussed in note 27 above, is that it threatens to ground voluntarist obligations in a primordial obligation of the relational kind, namely a duty of gratitude that is owed by the individual to the divine lawgiver. This would obviously undermine the voluntarist’s ambition to offer an alternative theoretical account of obligation. (It is interesting, however, that in ordinary cases of gratitude, directed obligations of gratitude have a discretionary character, something that is hard to reconcile with the role Adams ascribes to gratitude in relation to compliance with divine commands; compare the discussion of imperfect duties of gratitude in sec. 6.2 below.) But however these reasons of gratitude are understood, they still seem to be reasons of the wrong kind for compliance with moral obligations, insofar as they involve the subject’s relation to the lawgiver rather than to the persons potentially harmed or injured when the obligations are flouted.
52. See Darwall, “Bipolar Obligation,” pp. 36–37; also Wolf, “Moral Obligations and Social Commands,” p. 250.
53. A version of this objection also applies to the interesting account of obligation offered by John Skorupski in The Domain of Reasons, chap. 12. Skorpuski argues, very roughly, that an action is right or obligatory if it would be blameworthy for agents to perform it in their current epistemic condition. But—as Skorupski is well aware—on this approach, thoughts of right or obligation are transparent to the person who might blame the agents for what they have done. To me, by contrast, it seems that our paradigmatic forms of reactive blame, such as resentment, precisely reflect an understanding of the targeted agents as having acted wrongly, indeed as having wronged another party through their action.
54. For a statement of what is essentially this picture, see Darwall, “Bipolar Obligation.” On the general contrast between private and criminal law as involving a contrast between relational and monadic conceptions of “deonticity,” see Thompson, “What Is It To Wrong Someone?,” sec. 4.
55. Strawson, “Freedom and Resentment,” pp. 83–84.
56. For a contrasting view, see Lacey and Pickard, “To Blame or to Forgive?” Despite their best efforts (see especially sec. 3), Lacey and Pickard do not succeed in establishing that something recognizable as forgiveness is really intelligible in the criminal justice context. Whatever public authorities might be doing in forswearing negative emotions and attitudes toward an offender, their doing so would lack the quality of forgiveness, precisely insofar as it is not the action of someone who is understood to have been wronged by the offense.
57. Cf. the discussion of “authoritative motivation” in Scheffler, Human Morality, chap. 5.
58. See Adams, Finite and Infinite Goods, p. 257.
59. In sec. 5.1 below, I discuss in more detail the extension of moral status to individuals who, for a variety of reasons, may not be able to assert claims against others on their own behalf.
60. See Scanlon, What We Owe to Each Other, chap. 4, sec. 8, on the idea of “trusteeship” and the scope of morality.
61. See Scanlon, Moral Dimensions, chap. 4.
62. On this point, see my “Dispassionate Opprobrium,” secs. 3–4.
63. Indeed, it is something of a challenge for the nonreactive understanding of blame to make sense of the extended forms of blame, such as those intuitively available to third parties and to the agents of wrongdoing. If A wrongs B, then A may have acted in ways that give B reason to modify B’s behavior toward A, going forward. But A’s behavior toward B need not reflect any disregard for third parties, and so it will not necessarily give those parties reasons to modify their behavior toward A. The extended or vicarious forms of blame, it seems to me, are much more naturally interpreted in reactive terms.
64. Feinberg, “The Nature and Value of Rights.”
65. Feinberg, “The Nature and Value of Rights,” p. 250. Feinberg does not explicitly relate the performative sense of claiming to the practice of interpersonal accountability and blame, thinking of it rather in juridical terms (see, e.g., p. 251). But for reasons that have emerged in this section, it seems to me that reactive and other forms of blame are the most basic ways in which claims are asserted performatively in our moral practices.
1. Williams, Ethics and the Limits of Philosophy, pp. 174 and 180.
2. Williams, Ethics and the Limits of Philosophy, p. 187.
3. This aspect of promissory obligation is emphasized in the classic account found in Scanlon, What We Owe to Each Other, chap. 7.
4. A solution to this problem is sketched in Kolodny and Wallace, “Promises and Practices Revisited.”
5. See, for example, Raz, The Morality of Freedom, chap. 7; and Owens, Shaping the Normative Landscape.
6. A different nonmoralized account of promissory obligation, due to Margaret Gilbert, traces it to the nature of joint commitment rather than to some existing practice or convention. I consider Gilbert’s approach in sec. 4.2 below.
7. See Thompson, “What Is It to Wrong Someone?”
8. Thompson himself emphasizes this potential difficulty with the Humean conception of the distinctively moral order of right; see “What Is It to Wrong Someone?,” sec. 14. I return to his position in sec. 4.2 below.
9. Not only are there relationship-based moral obligations that the parties stand under in cases of this kind; there may also be independent “duties of love” that are constitutively connected to the valuable forms of interpersonal relationship that are at issue. For a discussion of this possibility, see my paper “Duties of Love.”
10. Raz, “Numbers,” p. 210. Raz identifies three kinds of relationship that can give rise to directed duties: commitments or undertakings, thick social bonds, and the kinds of interaction between people that generate debts of gratitude. Compare Hart’s comment, in his “Are There Any Natural Rights?,” at p. 179, note 7, that (directed) obligations arise “out of the relationship of the parties.”
11. Tomasello, A Natural History of Human Morality.
12. Tomasello, A Natural History of Human Morality, p. 86.
13. Greene, Moral Tribes.
14. On the distinction between System 1 and System 2 processes, see Kahnemann, Thinking, Fast and Slow.
15. See, for instance, Greene, Moral Tribes, chaps. 9–10. Cf. Parfit, Reasons and Persons, pp. 85–86.
16. Tomasello, A Natural History of Human Morality, pp. 158–63.
17. This would be to follow Williams and Tomasello, and to reject the more skeptical attitude of Greene.
18. See Greene, Moral Tribes, pp. 202–4; also Singer, The Expanding Circle.
19. Scheffler, “The Practice of Equality.” See also his “What Is Egalitarianism?,” and “Choice, Circumstance, and the Value of Equality.”
20. Scheffler, “The Practice of Equality,” pp. 25–31.
21. Scheffler, “The Practice of Equality,” pp. 35–37. For a different take on the application of the ideal of equality to relationships between intimates, see Shiffrin, “Promising, Intimate Relationships, and Conventionalism.”
22. Cf. Scheffler, “Membership and Political Obligation.” Though not specifically about demands of political equality, this article defends the idea that political obligations arise from the value of a specific kind of social relation, one that involves membership in a common political community.
23. Cf. Scheffler, “What Is Egalitarianism?,” p. 191: “human relations must be conducted on the basis of an assumption that everyone’s life is equally important, and that all members of society have equal standing.” Scheffler’s primary concern is with contexts, such as political membership and friendship, in which there are thick “human relations” to be “conducted” on an egalitarian rather than an inegalitarian basis. My point is that a recognizably similar idea can be situated within the more abstract context of morality, where it is a question of comporting oneself on the basis of “the assumption that everyone’s life is equally important.” The deliberative constraint that operationalizes this assumption is one that acknowledges limits on one’s conduct that flow from the fact that others have claims against one, just insofar as they are persons.
24. Thompson thinks of the relational part of morality as a conception of justice; there are other ethical requirements that are not bipolar in the same sense, determined, e.g., by the other virtues that humans need in order to do well. See “What Is It to Wrong Someone?,” p. 337. In this Aristotelian vocabulary, the relational account of interpersonal morality I am developing in this book could be described as an account of justice. Unlike some Aristotelians, however, such as Philippa Foot and perhaps Thompson himself, I resist the suggestion that we should think of justice in this sense as fundamentally continuous with the other ethical virtues; see the discussion of the perfectionist strategy in sec. 2.2 above. The relational features that distinguish demands of morality or justice from other ethical reasons, on my approach, are essential to their standing as first-personal obligations and as a basis for relations of accountability; these features constitute the unity of the normative domain that I have been referring to as interpersonal morality.
25. Thompson, “What Is It to Wrong Someone?,” pp. 361–62, 373.
26. Thompson, “What Is It to Wrong Someone?,” p. 373.
27. Thompson, “What Is It to Wrong Someone?,” sec. 11 (this condition appears to be the “chief lemma” to which Thompson refers in the heading to this section).
28. Cf. Thompson, “What Is It to Wrong Someone?,” p. 351 (for talk about “foothold”), p. 364 (for the reference to a “common source”).
29. Thompson refers to this as the “received” conception of justice; see “What Is It to Wrong Someone?,” sec. 10.
30. Thompson, “What Is It to Wrong Someone?,” secs. 14–15.
31. Thompson, “What Is It to Wrong Someone?,” sec. 16.
32. The following criticisms are fairly hastily sketched, and do not do full justice to Thompson’s remarkably rich and suggestive paper, which I strongly recommend that readers engage with for themselves.
33. Thompson, “What Is It to Wrong Someone?,” p. 372.
34. See, for instance, Thompson, “What Is It to Wrong Someone?,” p. 379.
35. Thompson considers a similar story in “What Is It to Wrong Someone?,” pp. 380–82. He complains that this strategy for isolating a common order of right threatens to be circular, insofar as it attempts to specify a form of bipolarity “by reference to the manifold it induces” (p. 381). But I fail to see anything circular in the movement of thought described in the text. (Indeed, since I reject Thompson’s first condition, it seems to me open to us to specify a still more extensive form of moral bipolarity, by reference, e.g., to a manifold that includes all individuals capable of concept-governed agency, regardless of whether they are competent with distinctively bipolar concepts.)
36. I draw primarily on the following texts by Gilbert: “Obligation and Joint Commitment,” “Three Dogmas about Promising,” and “Scanlon on Promissory Obligation.”
37. See, for example, Gilbert, “Obligation and Joint Commitment,” sec. 6, where she contrasts obligations of joint commitment with the obligations that appear to arise in other contexts, and suggests that the latter, insofar as they do not involve joint commitments, may be obligations only in a different (perhaps nondirected) sense.
38. See, for instance, Gilbert, “Three Dogmas about Promising,” pp. 306–8, 312–13.
39. Gilbert, “Three Dogmas about Promising,” p. 313.
40. Cf. Gilbert, “Three Dogmas about Promising,” p. 306: “Olive owes Roger her going to Chicago tomorrow, if and only if, in an as yet unspecified intuitive sense, Roger can appropriately regard Olive’s act of going to Chicago tomorrow as his.” So being able to regard an action as one’s own, in the intuitive sense, is necessary and sufficient for having a right against the agent to its performance.
41. Gilbert, for her part, apparently sees joint commitments in a much more extensive array of situations than seems to me intuitively plausible. See, for example, her “Mutual Recognition and Some Related Phenomena,” sec. 3.
42. It is not entirely clear that this is the sense in which Gilbert takes it to be intuitive that the actions of agents involved in joint commitments belong to both parties. Some of her formulations, at least, leave open an interpretation according to which, e.g., Roger can regard Olive’s going to Chicago as his, just insofar as he has a claim against her that she so act; see, for instance, “Three Dogmas about Promising,” p. 313. But if this is what she means, I see no explanatory work being done by the appeal to joint commitment.
43. For a different and to my mind more plausible account of the relation between joint commitment and directed obligation, see Alonso, “Shared Intention, Reliance, and Interpersonal Obligations.”
44. See, for instance, Scanlon, Moral Dimensions, p. 140, who speaks of the relationship we stand in to others as “fellow rational beings.”
45. Thompson, “What Is It to Wrong Someone?,” pp. 351 and 364. Cf. my “Dispassionate Opprobrium,” secs. 3–4.
46. See, for example, Williams, “Internal and External Reasons”; also Schroeder, Slaves of the Passions.
47. Examples include Korsgaard, The Sources of Normativity; and Street, “Constructivism about Reasons.”
48. We might put this by saying that there are reasons that both agent and claim-holder have in virtue of the obtaining of the directed duty, even if they can “have” those reasons only if there is epistemic warrant for people in their position to believe that they obtain.
49. On this kind of position, we should be reluctant to say that “there are” reasons that the moral nexus grounds, given the lack of the attitudes in the two parties that constitute reasons in the first place.
50. See, again, Street, “Constructivism about Reasons.”
51. We could also think of this position as a kind of reductionism, insofar as it reduces relational normativity to individual normativity.
52. An impressive example of an argument to this effect is Korsgaard, The Sources of Normativity. For critical discussion, see my “Constructing Normativity.”
53. On this point, see my paper “The Argument from Resentment.”
54. It is an interesting further question what the best alternative is to the individualist accounts that render directed obligations and claims problematic. The most promising approach, in my view, would be a nonnaturalist realism about the normative that leaves open the possibility that reasons are not conditioned by the subjective attitudes of the agent, and that tolerates a plurality of distinct normative relations (including those at issue in practical requirements, in aspirational reasons for action, and in fittingness-based reasons for attitudes).
55. Scanlon, What We Owe to Each Other, chap. 4.
56. See Williams, “Utilitarianism and Moral Self-Indulgence.”
57. See my “Dispassionate Opprobrium,” secs. 3–4.
58. These remarks apply only to the individuals who are in the core of the manifold of moral persons, insofar as they are both bearers of moral claims, and capable of acting in compliance with directed moral obligations themselves. As I noted earlier, however, it is possible that this manifold extends to include individuals who have claims against us, even though they are not subject to reciprocal obligations toward us. In these cases, too, our concern to honor their claims cannot be derivative from anything like mutual recognition, since the claimholders are not capable of reciprocating our acknowledgment of their claims. For further discussion, see sec. 5.1 below.
59. A prominent example of this tendency is John Stuart Mill’s appeal to the “desire to be in unity with our fellow creatures” in chap. 3 of Utilitarianism, p. 30; see also p. 33, where Mill glosses the object of this desire in terms of a “harmony” between one’s aims and feelings and those of the others with whom one shares social life.
60. We might think of this as the “method” of running it up the flagpole to see if they salute.
61. This idea figures saliently throughout Scanlon’s What We Owe to Each Other; see, e.g., p. 5. See also Nagel, “War and Massacre,” pp. 67–68; Nagel here contrasts justifications that might be launched to the world at large with justifications that are targeted specifically at the potential victims of one’s conduct, and suggests that moral permissibility is plausibly associated with justifications of the latter kind. A similar conception of justifiability to other individuals figures in the essays collected in Forst’s Das Recht auf Rechtfertigung.
62. On eudaimonistic reflection and morality more generally, see my “The Rightness of Acts and the Goodness of Lives.”
63. Cf. the discussion of friendship and morality in my “Scanlon’s Contractualism,” sec. 3.
64. See, especially, Scanlon, What We Owe to Each Other, chap. 4, sec. 5; for further discussion, see (again) my “Scanlon’s Contractualism,” sec. 3.
1. See Thompson, “What Is It to Wrong Someone?,” p. 354.
2. In the same section, however, I also reject the idea that the possibility of convergence in thought of this kind is a condition for the possibility of a bipolar order of right.
3. The absurdity of this idea reflects a commitment to the idea that “ought” implies “can,” in the most basic and plausible version of that idea. The normative significance of moral obligations—both for agents, as presumptive constraints on deliberation, and for others affected by their conduct, as bases of interpersonal accountability relations—presupposes that those subject to them should have the capacity to grasp that the requirements obtain and to regulate their conduct accordingly. These capacities are missing in agents who, at a given point in time, do not possess basic facility with relational normative concepts.
4. An illuminating treatment of cases in which claims might be assigned to individuals who are not able to assert them on their own behalf is Scanlon, What We Owe to Each Other, chap. 4, sec. 8. Scanlon’s discussion is couched in terms of the idea of justification to the individual (via the mechanism of a trustee); I focus instead on the idea of vicarious assertion of claims on behalf of the claimholder. These ideas converge, insofar as there is no basis for vicarious blame in cases in which the agent’s behavior can be justified appropriately (by reference to the interests of the individual affected, considered as bases for specific moral claims against the agent).
5. Cf. Thompson, “What Is It to Wrong Someone?,” pp. 359–60, 376–77. Thompson thinks it is a general condition for membership in a manifold of persons that thoughts about the corresponding order of right should be a typical or normal attainment for the individuals who belong to it. I reject the general condition, but feel that reference to what is normal for members of our species is helpful in thinking about individual humans who suffer the kind of severe disabilities or diseases that are here at issue.
6. See, for example, Scanlon, What We Owe to Each Other, chap. 6, sec. 5, and his Moral Dimensions, chap. 4. See also Smith, “Attributability, Answerability, and Accountability.”
7. Some questions about it are raised in my “Scanlon’s Contractualism,” sec. 2; see also Watson, “The Trouble with Psychopaths.”
8. We might plausibly understand reason to involve capacities to respond to normative considerations as such, in particular the capacity to adjust one’s attitudes to the recognition that there are reasons for making such adjustments. See, for example, Raz, “Reason, Rationality, and Normativity.” The protests to which I refer in the text are not rational responses in this sense, insofar as they do not involve the capacity to grasp normative contents. But they are responses to facts about actions and attitudes that we can recognize as constituting wrongs; this is what I meant in saying that we can understand the protesting creatures to be incipiently in touch with such normative considerations.
9. Prudence is often contrasted with morality, insofar as it involves one’s treatment of oneself rather than one’s treatment of other individuals. But within a relational conception of it, the moral should be understood as matter of what we owe to individuals as persons; and at least some duties to oneself, if intelligible, would seem to count as moral in this sense.
10. For a stark expression of this kind of puzzlement, see Hart, “Are There Any Natural Rights?,” pp. 181–82. Hart considers duties to oneself “absurd,” in part because he thinks the rights to which they correspond give one normative authority over the actions of the party who is bound by the duty. But in the cases at issue, it appears that one “authorizes” the actions that allegedly violate duties to oneself through one’s voluntary choice to perform them.
11. It has been questioned whether one can be wronged only if one has an antecedent claim; I return to this issue in sec. 6.1.
12. For this general conception of the issue, as well as some comments on the theoretical alternatives, see Sreenivasan, “Duties and Their Direction”; “A Hybrid Theory of Claim Rights,” pp. 262–64; May, “Moral Status and the Direction of Duties”; and Wenar, “The Nature of Claim-Rights.”
13. See Hart, Essays on Bentham, p. 183.
14. For a “definition” of rights that influentially emphasizes the interests of right-holders, see Raz, The Morality of Freedom, chap. 7.
15. Examples are Sreenivasan, “A Hybrid Theory of Claim Rights”; and Wenar, “The Nature of Claim-Rights.”
16. See Thompson, “What Is It to Wrong Someone?,” sec. 5, for an interesting discussion of the prospects for a reductionist account of bipolar notions (with particular attention to Raz’s analysis in terms of individual interests). In saying that I take relational norms as a theoretical primitive, I do not mean that there is no prospect of an illuminating analysis or account of their structure in a given domain, such as that of morality. The point, rather, is that any such analysis or account will implicitly deploy relational notions, and hence not be well understood as a reduction of the relational to the nonrelational. I come back to this issue in sec. 5.4 below.
17. I do not mean to suggest that we can never act in ways that affect the deceased in morally relevant respects. Failure to honor the wishes of one’s deceased parents in some matter of importance to them (as expressed, e.g., in their will) could plausibly be described as wronging them. As our temporal distance from the deceased greatly increases, however, it becomes correspondingly more difficult to affect their personal interests in significant ways. (An exception might be cases in which a distant predecessor established a significant institutional legacy, on terms that much later generations might fail to respect or to honor.)
18. I address the topic of moral argument in more detail in the following sections.
19. Classic treatments include Scanlon, “Preference and Urgency”; and Nagel, The View from Here, chap. 8.
20. Owens, Shaping the Normative Landscape, chaps. 5–6.
21. Cf. Scanlon, What We Owe to Each Other, pp. 211–13.
22. The Raz quotation is from his The Morality of Freedom, p. 191. Since Hohfeldian rights are claims against some other parties, it seems that Raz’s point applies to the justification of at least some claims; it does not matter, for present purposes, whether Raz would agree with me that there are some claims that do not involve assignable moral rights. For a contemporary account of promissory obligation that appeals to normative interests of this specific kind, see (again) Owens, Shaping the Normative Landscape.
23. For instance, Owens’s account of promissory obligation, which appeals to claim interests, has at least two features that distinguish it from the kind of moral justification I am interested in elucidating. First, Owens makes clear that he is not offering a specifically moral account of the obligations generated by promises; and second, he attributes explanatory significance to claim interests only in conjunction with facts about the social practices through which claims are recognized. See Shaping the Normative Landscape, especially chap. 6. I am not sure I understand the explanatory project that is meant to be shaped by these assumptions. But it is surely very different from the explanatory task I am here sketching, which is to make sense of the assignment of a moral claim to a specific individual.
24. Cf. F. M. Kamm, Morality, Mortality, Volume II, especially chap. 10.
25. It seems to me important to distinguish what we might call pure claim interests—whose object is just one’s possession of a normative claim—from a distinct interest we might have in possessing normative claims that are socially recognized. Once this distinction is drawn, it begins to seem doubtful to me that we really have pure claim interests. (I am not convinced that it matters that much to us to have specific claims that are not honored if they are not at least acknowledged by the members of our social world, as considerations that give us warrant for resentment and complaint.) But interests of neither kind will, by themselves, be capable of rendering intelligible our concern that our claims should actually be honored by the agents against whom they are held.
26. See, for example, Thomson, The Realm of Rights, chaps. 3–5. Thomson frames her discussion in terms of the question of whether rights or claims are absolute, denying that this is the case. She puts this by saying that it is not always the case that agents ought all things considered to respect or “accord” the claim right. But she also says that it is sometimes “permissible” for the agent to infringe claim rights that are held against them; see The Realm of Rights, p. 123. The latter formulation suggests that the issue is at least in part whether there is a specifically moral objection to infringing claim rights. See also Sreenivasan, “Duties and Their Direction,” sec. 1; Hart, “Are There Any Natural Rights?,” pp. 185–86; Kamm, “Owing, Justifying, and Rejecting,” pp. 465–66; and Skorpuski, The Domain of Reasons, p. 312.
27. See Ross, The Right and the Good. For a helpful recent discussion of Ross’s conception of a prima facie duty, see Hurka, British Ethical Theorists from Sidgwick to Ewing, chap. 3.
28. Cf. Ross, The Right and the Good, p. 20, where Ross notes that a prima facie duty “is in fact not a duty, but something related in a special way to a duty.” What is “morally significant” to Ross, it appears, is not being a prima facie duty per se, but being an act “of a certain kind (e.g. the keeping of a promise)”; cf. The Right and the Good, p. 19. But I confess that Ross is not as clear on these points as one might have wished. For discussion, see (again) Hurka, British Ethical Theorists from Sidgwick to Ewing, chap. 3. As Hurka notes, Ross developed a different conception of prima facie duties in his later Foundations of Ethics, one that appeals to relations of fittingness. But on neither approach do deontic concepts, as such, appear to play a contributory role in relation to obligation sans phrase.
29. A helpful discussion of this general point, in application to the positions of both Ross and Prichard, is Dancy, “More Right Than Wrong.”
30. For a different take on the relation between claims and Hohfeldian rights, see Feinberg, “The Nature and Value of Rights,” pp. 253–54. Feinberg takes claims to be considerations that provide prima facie grounds for asserting a claim in interpersonal practice, which however might not be conclusive. Rights are specified by valid claims, i.e., claims that there is conclusive (epistemic and practical?) reason to assert.
31. Ross allowed that there are some prima facie duties that are conditional in this way (including some of those generated by promises), and duties of this kind are not “outweighed” when the conditions of their validity are not satisfied. See Ross, Foundations of Ethics, pp. 94–99.
32. These points bear on the characterization of the deliberative role of directed obligations that was offered in chap. 2, where I suggested that they properly function as presumptive constraints on further deliberation and action. One might have thought that the constraints are merely presumptive, because in any given case there is the further question of whether it might be permissible to infringe the claim that was acknowledged. But in fact the idea is a different one: that even if a moral claim and the connected directed obligation obtain in the present circumstances, things might change, going forward, in such a way as to undermine the claim that previously obtained (e.g., through the occurrence of an emergency).
33. Residues are also sometimes said to include the appropriateness of compunction or regret about the action that one has performed in cases of this kind; but on no plausible view is there a duty to feel these things.
34. It would also be natural in such circumstances to apologize to the person whose property rights one had violated. This is somewhat puzzling, however, since—as we saw in sec. 3.3 above—apology literally involves an acknowledgment of wrong-doing, and there is by hypothesis no wrong that has been committed in the kinds of cases we are discussing. I think the kind of apology that seems to make sense in such cases reflects the “nameless virtue,” identified by Susan Wolf, whereby we take responsibility in an expansive sense for our actions and the harms they may cause. Trespassers who refuse to apologize, insisting legalistically that their action involved no strict wrongdoing, seem to lack the generosity characteristic of this virtue. See Wolf, “The Moral of Moral Luck.”
35. Cf. the discussion of “traces” of outweighed prima facie duties in Hurka, British Ethical Theorists from Sidgwick to Ewing, sec. 8.3. See also Kamm, Morality, Mortality, Volume II, chap. 12, p. 317.
36. This is a point emphasized by Hurka, British Ethical Theorists from Sidgwick to Ewing, pp. 70–72, 78.
37. The same consideration might explain the appropriateness of compunction or regret in these cases. Conscientious agents will be concerned about the effects of their actions on the personal interests of those affected by them, as bases of moral claims against them. But once one has regard for those interests, that concern might intelligibly be a source of regret that one was forced to act in a way that would damage them (a regret or compunction that may persist, even after one has honored all of the fallback claims that those same interests may generate). This kind of compunction or regret, however, will have a very different basis from the positional responses to wrongdoing discussed in sec. 3.3 above, such as remorse and the associated duties to apologize and make amends. These duties are genuine residues of claims, insofar as they presuppose that moral claims were in place that the agent flouted (so that the agent could be said to have wronged the bearer of the claims). In the cases here at issue, by contrast, compunction or regret reflects an independent concern for the interests that ground a person’s moral claims, one that might persist even under conditions in which there is no claim that has been flouted.
38. Though the relational conception can make sense of non-absolute deontological constraints of various kinds, I should acknowledge that it may have difficulty with one particular case in which familiar deontological constraints are sometimes understood to be defeated or undermined. This is the case in which a violation of moral constraints on our treatment of individuals could avert a disaster that would otherwise befall a large number of other persons. To make sense of such cases—which are staples in discussions of so-called threshold deontology—it would be most natural to assume that the interests of the people who are threatened by the disaster could somehow be aggregated together in the moral reasoning that goes to determining what we are permitted or required to do. But reasoning of this kind does not seem consistent with the relational conception, as I have been developing it in this book. I discuss in the next section the significance of the distinction between personal interests and impersonal reasons to the relational account of moral reasoning; secs. 6.3–6.4 take up in more detail some cases in which essentially aggregative considerations might appear relevant to reflection about what it is permissible for an agent to do.
39. For discussion of cases with this structure, see Kamm, Morality, Mortality, Volume II, chap. 12.
40. See Scheffler, Human Morality, pp. 31–33, for a very helpful survey of different ways in which an effective responsiveness to moral considerations could be realized through an agent’s deliberative processes and dispositions.
41. Relevant here are the discussions of utilitarianism in secs. 2.2 and 3.2 above.
42. A further respect in which the position differs from some forms of intuitionism is that the propositions grasped through rational intuition do not describe abstract principles, but claims about the rightness of particular acts. An intuitionism of this kind might aptly be compared to the “particularism” developed by Dancy; see, e.g., his Moral Reasons.
43. In support of this conclusion, it might be said (in the spirit of particularism) that there are too many distinct ways in which personal interests might bear on questions about the assignment of claims to an individual, and that the normative significance of such interests in the context of relational moral reflection interacts too multifariously with other relevant factors to permit illuminating general procedures or principles to be specified.
44. The primary statement of this view is Scanlon, What We Owe to Each Other.
45. On “personal” reasons that people have for objecting to candidate principles, see Scanlon, What We Owe to Each Other, chap. 5.
46. There is, strictly speaking, a gap between the interests that figure in contractualist reasoning and the concrete interests of individuals who have actual moral claims. Contractualist reasoning, about principles for the general regulation of behavior, is somewhat idealized, and it considers the personal interests associated with representative positions or roles that individuals might occupy under the normative regimes that are up for assessment (such as the position of recipient of a promissory commitment, or addressee of a lie). Cf. Scanlon, What We Owe to Each Other, chap. 5, sec. 4, on “generic” reasons for objecting to principles. The actual moral claimholder, in real life, will be an individual who occupies such a position or role in fact. Such individuals will, in virtue of occupying these roles, have personal interests that ground reasonable objections to principles that permit agents, e.g., to break promises made to the claimholders, or to address to them claims known to be false. (For further discussion of this issue, to which I am indebted, see Jonker, “In Defense of Directed Duties,” chap. 1.)
47. Scanlon’s own account of promissory obligation, which appeals to the value of assurance in support of a principle of fidelity, is presented in his What We Owe to Each Other, chap. 6. For discussion, see Kolodny and Wallace, “Promises and Practices Revisited.”
48. Thus, though promisees and third-party beneficiaries have interests of similar kinds that might adversely be affected by the promisor’s failure to perform, it makes a crucial difference that the adverse effects on the third party are not foreseeable by the promisor at the point in time when the promise is undertaken. This tells against the assignment of primary claims to third parties not to be disadvantaged by the actions of promisors, which is the idea that is under consideration in the present context. But the issue of foreseeability does not necessarily tell against the different idea that there is a generic secondary claim not to be harmed by actions that violate primary claims, once these have been assigned to individuals; I return to this idea in sec. 6.1 below.
49. Gilbert, “Scanlon on Promissory Obligation.” A similar criticism is developed by Darwall in his “Demystifying Promises.”
50. Gilbert, “Scanlon on Promissory Obligation,” p. 282.
51. See Gilbert, “Scanlon on Promissory Obligation,” sec. 3, for this argument.
52. Gilbert, “Scanlon on Promissory Obligation,” pp. 284–85. Gilbert might have added, as well, that having the power to release the agent from a directed duty is also not necessary for having a right or a claim. Some claims might be inalienable, and this possibility poses a challenge to the so-called will theory of rights or claims.
53. Cf. Scanlon, “Reply to Wenar,” p. 404: “in a contractualist account of promises promisees are singled out in two ways: by the central role that their interest in assurance has in justifying principles of fidelity and by the way in which those principles must make promissory obligations sensitive to their wills.” Scanlon mentions both of these aspects of the contractualist account of promising because he wishes to emphasize its ability to accommodate ideas that are prominent in the “interest” and “will” theories of rights. But I believe the first aspect is the one that is ultimately fundamental to the implicitly relational structure of contractualist morality.
54. Scanlon, What We Owe to Each Other, pp. 11–12.
55. Scanlon, “Wrongness and Reasons”; see also Scanlon, “Replies,” pp. 435–39.
56. These different conceptions are in competition with each other, in part, because they are attempts to identify different high-level properties that make actions right or wrong in the to-be-done or not-to-be-done sense. But they are also in competition because they differently fill in the abstract and somewhat indeterminate concept of moral rightness and wrongness. Thus competing substantive conceptions will emphasize different aspects of the abstract concept of moral rightness that they are trying to specify. (Consequentialists, for instance, typically attach less importance to the interpersonal aspect of the moral than do some other theoretical approaches—to a degree, I believe, that gives their accounts a somewhat revisionist character.) Note that Scanlon himself originally spoke of his theory as an account of morality in the narrower sense “having to do with our duties to other people”; see What We Owe to Each Other, p. 6. This suggests that his account might be one that answers to a specifically relational concept of morality, a suggestion that would make it hard to see it as a competitor to other moral theories (insofar as they do not similarly construe the domain they are describing in relational terms). It therefore seems to me more fruitful to think of the account as a substantive conception that fills in a more abstract concept of the moral right (construed, perhaps, as a set of duties that concern other people, considered as individuals whose interests are equally important, and that also have the deliberative and interpersonal features discussed in chaps. 2 and 3 above).
57. Construed as a theory of moral rightness, contractualism might appear to have reductionist ambitions, insofar as the main elements of the theory (including personal interests, reasons for rejecting candidate principles, and so on) seem intelligible within a nonrelational context. But the way these elements are combined in the theory seems to incorporate an essentially relational structure, in the ways I have tried to explain in the text. The result is not, I think, helpfully understood as a reduction of the relational to the nonrelational, but rather as an elucidation of relational morality (via an account of the nature of moral principles and the form of reasoning through which they are justified). An even more clearly nonreductionist approach to understanding directed obligations and their correlative claims might be to see them as specifications of some essentially relational ideal. For examples of this general strategy, see, for instance, Ripstein’s account of the private law of torts, in his Private Wrongs, which derives specific claims and directed duties from the fundamentally relational idea that no person is in charge of another person and his or her means. See also the account of rights sketched in Zylberman, “The Very Thought of (Wronging) You,” which appeals to a basic ideal of independence and nondomination. These ambitious approaches are attractive in many respects; but I worry that the relational ideals to which they appeal do not provide a perspicuous basis for the full range of directed obligations that make up interpersonal morality. (The ideals are in some ways too abstract, so that it is difficult to derive from them specific moral obligations, and in other ways too determinate, describing values that do not seem to apply to some ways of wronging others.)
58. He sometimes characterizes specific moral duties, such as the duty of promissory fidelity, in relational terms, as duties owed to another party; see, e.g., What We Owe to Each Other, p. 316. But it is striking that the terms “claim” and “right” do not even appear in the index of this book.
59. See Scanlon, “Reply to Wenar,” pp. 402–3. Note that I have not suggested that we might explain why certain actions are obligatory by appeal to the claims of other parties. This is connected to a point emphasized in sec. 5.3, namely that claims represent conclusions of moral deliberation, rather than inputs into reflection about what people are obligated to do. (The specific moral reasoning that explains why a duty is owed to another party will also, itself, explain the assignment of a claim to that party.) It does not follow that the relational structure of the moral is not significant, only that its significance lies elsewhere, in the light it sheds on larger normative features of morality.
60. Parfit, On What Matters, vol. 2, secs. 21–23.
61. For a critical discussion of this debate, see Kumar, “Contractualism on the Shoal of Aggregation.”
62. See Scanlon, “How I Am Not a Kantian.”
63. See Scanlon, “Contractualism and Justification.”
64. One might stipulate that each of the persons on Rock 2 has a claim to be rescued, in virtue of the fact that they are members of the larger group of individuals with symmetrical personal interests in being saved. But on this interpretation of it, the contractualist account of reasonable rejectability would not shed much light on this assignment of claims, precisely insofar as it becomes detached from the idea that claimholders have a distinctive personal objection or complaint about actions that violate their claims. (The persons on Rock 2 have personal objections, on Scanlon’s latest proposal, to the actions that violate their stipulated claims. But their personal objections are not distinctive, but precisely comparable to the objection that the individual on Rock 1 has to the rescuer’s saving the Rock 2 people, an objection that by hypothesis does not correspond to a moral claim.) In sec. 6.3 below, I develop a different relational explanation of the idea that there is a moral requirement to save the individuals on Rock 2.
65. Parfit, On What Matters, vol. 1, secs. 22, 54. Parfit at one point concedes that Scanlon’s theory will lose the idea that moral obligations are owed to specific individuals, and that those individuals would be wronged if we flouted them, if the individualist restriction is given up; see On What Matters, vol. 2, pp. 241–42. My point is that in losing these things, the theory is deprived of precisely the elements that contribute in large measure to its interest and appeal.
1. Raz, “Numbers,” p. 210.
2. This would be a specifically moral obligation that is owed by children to their parents, one that applies in virtue of more general principles that specify what we owe to each other. I think there are, in addition, sui generis directed duties that are owed to each other by the parties to relationships of this kind. For discussion, see my “Duties of Love.”
3. For further discussion of such issues, see Scheffler, “Relationships and Responsibilities”; and Kolodny, “Love as Valuing a Relationship”; also my “Duties of Love.”
4. Shiffrin has argued that promises serve to redress imbalances of power and vulnerability that naturally develop in the context of relationships between intimates; see her “Promising, Intimate Relationships, and Conventionalism.” She appeals to this consideration, however, to explain why participants in such relationships require the power to undertake promissory commitments, not to elucidate the force of such commitments once they are made.
5. Cornell, “Wrongs, Rights, and Third Parties.”
6. Hart, “Are There Any Natural Rights?,” p. 180.
7. Cf. Shiffrin, Speech Matters, pp. 19–26. Note that the situation would be different if the third party overheard the lie as a result of covert surveillance that could not have been anticipated by the agent.
8. Compare the treatment of the Palsgraf case, and others with this structure, in Ripstein, Private Wrongs, chap. 4. Ripstein defends and develops Cardozo’s thought, in Palsgraf, that the defendants cannot have wronged the plaintiff, because there is a contradiction in “the idea of owing [a] duty to someone who is unforeseeable” (p. 90).
9. See Cornell, “Wrongs, Rights, and Third Parties,” note 26, pp. 121–22. Cornell quotes Anscombe, who writes, in her “Modern Moral Philosophy,” at p. 184, that “a man is responsible for the bad consequences of his bad actions.” (I note that Anscombe’s main interest, in the larger passage in question, is in the somewhat different claim that people are not responsible for the bad consequences of good actions; also, that her core conception of good and bad actions is very different from the relational one that Cornell himself presupposes.)
10. See Cornell, “Wrongs, Rights, and Third Parties,” p. 140. Insofar as this is the case, it seems, after all, that “every wrong can be traced to a rights violation,” which is a claim Cornell disavows on p. 142. One theoretical peculiarity of Cornell’s favored approach is that it operates with a disunified conception of wrongs. Some wrongs essentially violate rights; others are not understood to consist in rights violations, though they presuppose that the harmful consequences in which they consist can be traced to rights violations. It is an advantage of the alternative I propose that it sees every wrong as itself the violation of a claim.
11. It might be asked why, if we are willing to go this far, we do not also assign to people claims not to be harmed in consequence of actions that fall short of what the agent ought to do, even if those actions do not flout primary moral claims. Thus, I might akratically sleep in rather than going to the library first thing in the morning, as is my customary practice; and this might, unbeknownst to me, cause my neighbor to get fired from her job. (Perhaps she developed the habit of leaving for work only when she saw that I was on my way to the library, and so was late on the fateful day, missing an important meeting with her boss.) But I feel no intuitive pull toward thinking that there might be secondary moral claims in cases of this kind. Bad consequences of bad actions seem imputable to the agent, at most, only in cases where the badness of the actions involves neglect of the social responsibilities of agency. Moral claims not to be harmed thus seem secondary in the way described in the text, insofar as they are parasitic on primary moral claims that are flouted through the actions that violate them. Their assignment perhaps reflects a general concern we have, as moral persons, that agents should be attentive to their interpersonal responsibilities.
12. On enforceability as a condition of rights, see Wenar, “The Nature of Claim-Rights,” sec. 10. On identification of an agent to whom the corresponding duty can be assigned, see O’Neill, Towards Justice and Virtue, p. 129. On general conditions of enforceability, see James, “Rights as Enforceable Claims,” pp. 133–47.
13. For the suggestion that enforceability might apply to juridical but not necessarily to moral rights, see Zylberman, “The Very Thought of (Wronging) You,” p. 162.
14. Compare defensive liberty rights, such as our right to freedom from bodily trespass, which is held against any agent we might come into contact with. The class of people who are bound by the correlative directed duty in such cases, though extensive, seems much more clearly defined than the class of individuals against whom secondary claims not to be harmed might be held.
15. On inalienable moral rights, see Sreenivasan, “A Hybrid Theory of Claim Rights,” pp. 259–61.
16. Specificity is of course a matter of degree. Even in cases of rights to positive performance, there is some discretion that agents have to determine specifically how they are going to fulfill their directed duties. Having promised to meet you at the airport, I have leeway to decide for myself which bridge to take to get there, how long in advance I will arrive, etc. The general idea is that moral claim rights will seem firmly in place in proportion to the degree to which the directed duties connected with them are more determinate. The duties at issue in cases of gratitude, it seems, are at the less determinate end of the spectrum.
17. Note that this conception obviously does not transfer over to the context of an individual’s relationship to God, where—as I observed in secs. 3.2 and 3.4—some voluntarists appeal to gratitude to explain our reasons for complying with divine commands. But whatever conception of gratitude is thought to be in place there, it will lack the element of discretion characteristic of the gratitude we owe to other human agents. (The normative considerations that support our compliance with divine commands should not be such as to leave it up to us whether or when to so comply!) The reasons of gratitude that appear to figure in some divine command theories therefore represent a different normative phenomenon from the imperfect duty of gratitude that often obtains in interpersonal relations, one that would require a very different kind of account.
18. I am indebted here to the illuminating discussion in Herman, “Being Helped and Being Grateful.”
19. See Herman, “Being Helped and Being Grateful,” p. 398.
20. See, for example, Singer, The Most Good You Can Do.
21. A starkly contrasting position, in the spirit of the Effective Altruism movement, is defended by Pummer in “Whether and Where to Give.” Whereas I maintain that we have discretion to act suboptimally in discharging our moral obligations of mutual aid, Pummer contends that if we are making charitable contributions, we (often) have to give to the most effective aid organizations, even when it would not be wrong to give nothing.
22. This is an important theme in the work of Bernard Williams; see, for instance, his “A Critique of Utilitarianism.”
23. The resulting account would have affinities with the approach sketched by Murphy in Moral Demands in Nonideal Theory. For critical discussion of this approach, see Cullity, The Moral Demands of Affluence, chap. 5.
24. A powerful statement of the maximally demanding view is offered by Unger in Living High and Letting Die. Note that the question of the level of contribution expected of individuals is distinct from the question of whether individuals have discretion to direct their efforts in suboptimal (but still effective) directions. A perfectly coherent position might be that we are each required to contribute most of our income (beyond the minimal level necessary to support our basic needs) to charitable causes that address acute human welfare deficits; but that it is open to us to devote all of those contributions (say) to local programs that support the homeless in our community, rather than to efforts to prevent malarial infections farther afield.
25. For any given agent, we might ask two questions: (a) what is the range of eligible actions open to the agent to provide assistance to others in acute need (i.e., actions that are over some reasonable threshold of expected effectiveness)?; (b) which specific individuals would benefit from the assistance that is made possible by each of those available actions? It is a commonplace that the resulting class of potential beneficiaries, for those of us in the modern world, is extremely extensive, since there are so many easy actions open to us that would make available benefits to people who are extremely remote from us. (Consider online donations to aid organizations with global reach.)
26. See, again, Murphy, Moral Demands in Nonideal Theory, for a broadly similar view. Murphy emphasizes the fact that our duties of mutual aid give us a common, agent-neutral aim; my discussion differs in stressing the relational character of those duties, and the nature and extent of the claims with which they correspond.
27. Cf. Cohen, “Who Is Starving Whom?”
28. Compare the comment of Theano Laoumis of Lesbos, Greece, describing the response of fellow islanders to the thousands of migrants who washed up on their shores in 2015: “You didn’t know who to save first, there were so many people. But we did save them. It was only natural.” Quoted by Alderman, “Greek Villagers Rescued Migrants.”
29. Cf. Cullity, The Moral Demands of Affluence, pp. 75–77. Though he admits that a fair share view applies more naturally to collective projects than to direct emergencies, he ultimately rejects it even there. But it may be relevant that he draws the distinction somewhat differently than I have done. For me, the idea is that we need different principles to explain our duties in regard to ongoing collective efforts to address structural problems of acute human need, and in direct emergency situations in which only a few individuals are even in a position to respond. Cullity, by contrast, distinguishes collective endeavors from individual direct action, and notes that some situations that require a collective response have the character of direct emergencies. For me, the latter are not situations that call for a fair share solution, so the fact that that approach would deliver some counterintuitive verdicts in the emergency context is not an objection. But the issues are complicated, and there is of course much more to be said about them.
30. See Cullity, The Moral Demands of Affluence, e.g., secs. 5.3–5.5, also chap. 10.
31. The locus classicus for discussion of this problem is Parfit, Reasons and Persons, part 4. Parfit’s discussion is shaped by the assumption that any theory must recognize that beneficence is at least a part of morality, and that beneficence in turn involves “our general moral reason to benefit other people, and to protect them from harm” (Reasons and Persons, p. 371). The challenge is to develop a principle of beneficence that can accommodate our intuitions about a range of hypothetical cases involving populations of different sizes and welfare levels (what he calls “Theory X”). His final attempts at meeting this challenge are his “Can We Avoid the Repugnant Conclusion?,” and “Future People, the Non-Identity Problem, and Person-Affecting Principles.” I myself am doubtful that moral philosophy really needs a Theory X in Parfit’s sense, in part because the essentially consequentialist conception of beneficence latent in his discussion—as a matter of our general “moral” reason to benefit people and protect them from harm—seems to me questionable. On the latter point, see Scheffler, Why Worry About Future Generations?, chap. 1.
32. An important challenge in thinking about this issue is posed by the fact that our personal behavior, considered on its own, would apparently make little difference to the overall trajectory of environmental degradation or climate change. As we saw in the discussion of mutual aid above, contractualism invites us to consider, in the first instance, the effects of general compliance with candidate principles by the people whose behavior they purport to regulate. It is clear that continued complacency on the part of all of us who live in affluent and resource-intensive parts of the world would have significant cumulative effects on the life circumstances of future generations.
33. Where quality might take into account not only the aggregate well-being or happiness of the people who inhabit each of the future scenarios we might help to bring about, but also the distribution of happiness and suffering among those people.
34. See, for example, the approach implicit in Shiffrin, “Wrongful Life, Procreative Responsibility, and the Significance of Harm.” Broadly similar positions are defended, e.g., by Harman, “Can We Harm and Benefit in Creating?”; Woodward, “The Non-Identity Problem”; and Kumar, “Who Can Be Wronged?”
35. Parfit, On What Matters, vol. 2, chap. 22.
36. Parfit, On What Matters, vol. 1, p. 419; cf. On What Matters, vol. 2, pp. 234–35.
37. On the plausibility of a noncomparative conception of harm, see Parfit, On What Matters, vol. 2, pp. 233–34. What Parfit fails to appreciate is that this noncomparative conception of harm can be combined with an essentially comparative account of moral reasoning to yield a relational framework for thinking about the moral dimension of cases involving the non-identity problem. See, e.g., his “Future People, the Non-Identity Problem, and Person-Affecting Principles,” pp. 125–26, where he argues that in many cases of this kind, we cannot restrict ourselves to thinking about the effects of our actions on people who will actually exist, but must also think about how other, merely possible people would have fared if we had acted otherwise. But comparisons of this sort are entirely consistent with the relational approach I have been defending. Indeed, as I argued in sec. 5.4, they are built into the contractualist account of relational moral reasoning, which is essentially comparative, and hence requires us, in cases that involve the non-identity problem, to compare the objections of the actual individuals whom our actions cause to exist with the objections that different people would have had if we had acted otherwise. For a sympathetic recent discussion of contractualist reasoning about obligations to future generations, see Kumar, “Wronging Future People.” Kumar rightly emphasizes the generality of the moral principles determined by contractualist reasoning, as well as the fact that they define duties that are owed to future people; but he somewhat downplays the essentially comparative character of moral justification, which seems to me important for understanding cases involving the non-identity problem in relational terms.
38. For discussion of this case, see Frick, “Conditional Reasons and the Procreation Asymmetry”; also his “Zukünftige Personen und Schuld ohne Opfer.”
39. See Parfit’s discussion of the “no difference view,” in Reasons and Persons, pp. 366–69.
40. For discussion of some other cases that involve similarly eccentric retrospective attitudes, see my book The View from Here. Consideration of such cases lends support to the idea that one can have a moral claim against someone, even if it is not the case that one rationally prefers, all things considered, that the claim should not have been violated; contrast Frick, “Zukünftige Personen und Schuld ohne Opfer,” sec. 11.
41. See, e.g., Greaves, “Population Axiology.” For further discussion of issues in this area, see, e.g., Broome, “Should We Value Population?”; Broome, Weighing Lives; McMahan, “Causing People to Exist and Saving People’s Lives”; and McMahan, “Problems of Population Theory.”
42. Parfit, Reasons and Persons, p. 369.
43. While this inference is tempting, I note that it elides two very different kinds of judgments, those about the value of outcomes and those about the rightness of actions, in ways characteristic of consequentialist approaches. See also Parfit, On What Matters, vol. 2, chap. 22, which similarly bases conclusions about what it is right for individuals to do on somewhat elusive intuitions about, e.g., “the goodness of outcomes and peoples’ complaints” (p. 227).
44. See, for example, Parfit, Reasons and Persons, pp. 393–94. Also relevant here is Jan Narevson’s telling observation that “we are in favor of making people happy, but neutral about making happy people,” in his “Moral Problems of Population,” p. 80. (Even the first part of Narveson’s slogan is potentially misleading, however: we may be in favor of making people happy, but it isn’t clear that we have a moral obligation to do so. To think that we do is arguably to fall back on the questionable assumption that beneficence, construed in broadly consequentialist terms, is at least part of any sane moral outlook.)
45. For this way of thinking about the content of our duties regarding future generations, see Scheffler, Why Worry About Future Generations? Scheffler argues, among other things, that our concern to meet this standard has important sources outside of morality, e.g., in our relation to the values around which our own lives are organized. While I agree with him about this, I also think there are obligations of interpersonal morality that derive from the relation we will stand in to future individuals, and that these moral obligations are similar in content to the nonmoral norms that Scheffler describes.
46. Scanlon originally suggested that there is a personal objection that could be brought by an individual in tiebreakers; when it is a question of saving one person on Rock 1 vs. saving two people on Rock 2, the two people on Rock 2 could each object, on his or her own behalf, to a principle that permits you to go to either rock; the personal objection would be, roughly, that his or her presence makes no difference to the moral verdict about what it is permissible to do. See What We Owe to Each Other, chap. 5, sec. 9. But there are many situations in which the presence or absence of a given individual makes no difference to the question of what it is permissible for an agent to do. If aggregation is permitted, the presence of the single individual on Rock 1 makes no difference of this kind; similarly, in the rescue situation described in the text, where there are several on Rock 2, each of them could individually say that his or her presence makes no difference, even if it is morally required to save the greater number. Cf. Raz, “Numbers,” pp. 204–6. Scanlon now favors a different approach to cases of this kind, arguing that the number of individuals on Rock 2 with equivalent personal interests in being saved might make it unreasonable for the person on Rock 1 to reject principles requiring that you go to Rock 2; see his “Contractualism and Justification.” I offer a brief critical discussion of this proposal in sec. 5.4 above, noting that it appears to sacrifice the relational structure that makes contractualism attractive in the first place.
47. See, e.g., the classic discussions in G. E. M. Anscombe, “Who Is Wronged?”; and Taurek, “Should the Numbers Count?”
48. For an interesting suggestion of this kind, see Munoz-Dardé, ”The Distribution of Numbers and the Comprehensiveness of Reasons.” In particular, she proposes that the impersonal value of human life might represent a good reason for action, one that renders it intelligible that the rescuer might choose to save more rather than fewer, without making it obligatory so to act. See also Anscombe, “Who Is Wronged?”; and Scanlon, “Contractualism and Justification.” (Note: though I shall continue to refer to impersonal reasons of human life and human well-being in what follows, the relevant considerations should be understood to include, potentially, the well-being of all members of the manifold of moral persons, including nonhuman claimholders, if there are such.)
49. In a different kind of case, of course, there might be compelling personal reasons that tip the balance in the other direction, in favor of saving the person on Rock 1. Maybe that person is a close friend of yours, someone who has a claim against you as a friend to be given special consideration. This is a consideration that figures in Taurek’s discussion in “Should the Numbers Count?”
50. Cf. Kumar, “Contractualism on Saving the Many.” Kumar argues that once the symmetrical objections of the individuals who find themselves on the different rocks have been determined to neutralize each other, one for one, we are to consider the objections of the remaining individuals on Rock 2 in a second stage of contractual-ist reasoning. I am suggesting, by contrast, that the earlier objections of everyone to principles permitting the rescuer to save fewer are dispositive, once we recognize that the later objections of individuals to principles requiring the rescuer to save those on the other rock are inconclusive.
51. For a brief critical discussion of similar proposals, see Kamm, Morality, Mortality, Volume I, p. 120; and Otsuka, “Saving Lives, Moral Theory and the Claims of Individuals,” pp. 123–24. It is perhaps an advantage of this approach that it does justice to a desideratum articulated by Scanlon in “Contractualism and Justification,” namely that the explanation for why we should save more in a “disjoint” case such as Rock 1 vs. Rock 2 should also explain why we should save more rather than fewer in a “subset” case, where there is a single group of potential rescuees, and saving more rather than fewer members of this group would not impose significant additional costs on us. The ex ante argument sketched in the text applies equally to cases of both kinds. Note, however, that there is a different objection to saving fewer in the subset case that does not carry over to disjoint cases. In the subset case, some individuals in the larger set of prospective rescuees have a personal objection, at the time of the rescue operation, to principles that permit us to save fewer members of their group; and there is no similar objection at that time that individuals in the subset have to principles that require us to save more.
52. This is a common objection to the general approach I have sketched; see the texts by Kamm and Otsuka referred to in the preceding note.
53. Note that something like this assumption may be necessary to explain, in relational terms, why the rescuer is not permitted to save none in cases in which, e.g., there is the same number of individuals on each rock. Those individuals have symmetrical personal objections, which cancel each other out, to principles that require the rescuer to save those on the other rock. But what is the personal objection to principles that permit the rescuer to save nobody? One basis for it would be the idea that all of the individuals could reject such a principle, since their ex ante likelihood of being saved in such a situation will be enhanced if the rescuer is required to save those on one rock or the other. But this seems to assume, again, that the occurrence of the emergency functions in the way described in the text, as a kind of natural lottery.
54. This is of course an echo of Anscombe’s question in “Who Is Wronged?”
55. This would be another case, like that of mutual aid, in which individuals have claims against agents to do things that will not necessarily benefit them in particular in the end. Cf. Munoz-Dardé, “The Distribution of Numbers and the Comprehensiveness of Reasons,” p. 196: “Justifying to you a duty to save does not mean there is a duty to save you.”
56. See, again, my The View from Here.
57. Two classic early contributions are Foot, “The Problem of Abortion and the Doctrine of the Double Effect”; and Thomson, “Killing, Letting Die, and the Trolley Problem.” A helpful recent discussion, which includes commentary from leading contributors that provides something of an overview of the current debate, is F. M. Kamm, The Trolley Problem Mysteries.
58. Sometimes the intuition about the trolley driver is put in terms of a conflict of negative duties, with a duty not to kill one rationally outweighed by a duty not to kill the five; see, e.g., Foot, “The Problem of Abortion and the Doctrine of the Double Effect,” pp. 27–28. But these would not be directed obligations, as the relational account understands them, since (like Hohfeld-style rights) they figure as inputs into moral reflection rather than outputs of such reasoning; cf. sec. 5.3 above.
59. Kamm, The Trolley Problem Mysteries, pp. 16–18.
60. For expression of skepticism that these different ways of causing harm to one in the course of saving others make a moral difference, especially in the context of contractualist reasoning, see Parfit, On What Matters, vol. 1, sec. 53.
61. Indeed, this has become one of the fixed points in discussions about trolley cases. But Thomson, who introduced this case in the first place, has since had second thoughts about whether it really would be permissible for the bystander to switch the trolley onto the siding; see her “Turning the Trolley.”
62. Similar considerations apply to Scanlon’s Case One, in which Grey would lose a few years of life by giving up an organ to White, and White would gain many more years if the donation were made; see Scanlon, “How I Am Not a Kantian,” p. 136; also Parfit, On What Matters, vol. 2, p. 192. In considering a principle requiring people to sacrifice organs in situations of this kind, it is important to take into account not merely the years of life that Grey and White would each lose under this principle and the alternatives to it, but the objections that both of them would have to a set of social arrangements in which their bodies were constantly available for beneficent public appropriation.
63. Cf. Scanlon, “How I Am Not a Kantian,” p. 138, on the possibility of a “moral standoff” within contractualist reasoning. Scanlon is considering here what he calls Case One, described in the preceding note. The idea is that Grey and White might each have comparably powerful personal objections to principles that either require Grey to donate the organ, or permit Grey not to make the donation. For further discussion of this case, see Scanlon’s more recent “Contractualism and Justification.”
64. For a vigorous statement of this worry, see Fried, “Can Contractualism Save Us from Aggregation?”
65. See Frick, “Contractualism and Social Risk”; see also James, “Contractualism’s (Not So) Slippery Slope”; and Kumar, “Risking and Wronging.” The vaccine example, and the terms in which it is discussed, are borrowed loosely from Frick’s persuasive discussion of this general approach. Note that the contrast between “ex ante” and “ex post” objections here refers to objections that individuals have before and after the actions up for assessment are carried out. The ex ante objections at issue in my discussion of the two-rock rescue situation in the preceding section, by contrast, are objections that individuals have at a still earlier point in time, before they know what rescue situations they will find themselves in at all.
66. See, yet again, Wallace, The View from Here.
67. Frick, “Contractualism and Social Risk,” p. 212.
68. Frick, “Contractualism and Social Risk,” sec. 8.
69. Frick, “Contractualism and Social Risk,” sec. 9.
70. Parfit, On What Matters, vol. 2, p. 197.
71. Parfit, On What Matters, vol. 2, p. 232.
72. See Parfit, On What Matters, vol. 2, p. 234. The reasoning is questionable, however. Note that humans have noncomparative interests in being able to live as long as it is typical for members of their communities to survive, something that is a condition for their being able to sustain projects and relationships through their different natural stages. Even people who will enjoy forty happy years have a powerful objection on their own behalf to being brought into a scenario in which it is known in advance that their lives would be cut short in medias res. (To say nothing of the comparative and noncomparative objections that their friends, parents, children, and life partners would have, on their own behalf, to losing the person to whom they are related in these ways in the person’s prime of life.) This objection seems more powerful than the comparative objection that the X-people would have to scenario B (involving the loss of a single year of happy life); and of course, there is no similar noncomparative objection that any of the Z-people could bring against our choosing B. For these reasons, it seems to me that Case Seven is not genuinely problematic for a relational interpretation of contractualism, and that Case Four and Frick’s Miners (1 vs. 1,000) better illustrate the limits of the kind of reasoning implicit in this approach to interpersonal morality. I shall therefore mainly focus on the latter cases in the discussion that follows; but the remarks I make about those cases could also apply to Case Seven if readers disagree with the relational treatment of it I have just sketched.
73. Parfit, On What Matters, vol. 2, p. 234.
74. Cf. Scanlon, “How I Am Not a Kantian,” p. 139. Speaking of decisions by public officials, Scanlon here writes, “Producing the best consequences might be the correct standard in these cases not because it is the basis of morality but because it is what is owed to people in situations of that kind, by agents who stand in a certain relation to them.” This suggestion would admittedly need careful development. If there is a directed special obligation here, for instance, it will presumably be owed to all of the individuals in the jurisdiction of the public officials, considered collectively. Thus, each member of the group might have a basis for complaint if, for instance, the officials were to underinvest in mine safety programs, in order to support overly robust search and rescue teams. But these claims would be based in interests that individuals share, insofar as they are members of the polity to whom consideration is owed, by officials whom they have entrusted to make decisions about the allocation of public funds. So all members of the polity might be wronged by the decision to overinvest in rescue rather than prevention, including people who end up benefiting personally from such a decision.
75. Cf. James, “Contractualism’s (Not So) Slippery Slope,” sec. 5.
76. Cf. Munoz-Dardé, “The Distribution of Numbers and the Comprehensiveness of Reasons,” secs. 7–8. Munoz-Dardé proposes that there are limits on the claims that individuals can reasonably make on scarce public resources in contexts of social choice, limits that are sensitive to the numbers of individuals who stand to benefit in other ways if resources are made available to address their needs. Considering Frick’s Miners (1 vs. 1,000), Scanlon tentatively explores a similar proposal in “Contractualism and Justification”; but he does not limit the reasoning to contexts of social choice by authorized public officials, which undermines the cogency of the proposed form of argument. (Considered as a question of interpersonal morality, I don’t see how the impartial value represented by even twenty statistical lives can make it unreasonable for the trapped miner to reject principles that would almost certainly entail his or her death.)