PEOPLE MEAN MANY DIFFERENT THINGS when they talk about morality. In a familiar modern sense, however, morality may be thought of as a set of normative constraints on attitudes and actions that stem from the fact that we inhabit a common world together with other agents. More specifically, and more controversially, it may be thought of as a normative nexus that links us individually with each of the persons who might potentially be affected by what we do. According to what I will call the relational interpretation of it, morality involves a set of requirements on action that are constitutively connected to claims that others have against us, just insofar as they are persons. Requirements that are connected to claims in this way have a built-in directionality, specifying things that we owe it to others to do. So on the relational interpretation, morality could be said to be fundamentally a matter of what we owe to each other.1
This book offers a statement and defense of the idea that morality collects a set of fundamentally relational requirements. A leading idea of the discussion is that moral standards have some significant normative features that can be made sense of only if we interpret them in relational terms. They function to define practical requirements, which regulate the deliberations of agents in the distinctive manner of obligations; and they also have interpersonal significance, providing a normative basis for relations of moral accountability. I argue that the relational approach is better able than the alternatives to illuminate these significant aspects of the moral. In addition, I shall highlight and defend the major philosophical commitments and presuppositions of the relational interpretation, which have not been subjected to sustained critical investigation. I shall also discuss some of the first-order implications of the relational approach, for questions about what, specifically, we owe it to each other to do, and about the nature of the reasoning that goes into deciding issues of this kind.
The idea of relational requirements in the most general sense is implicit in talk of moral and legal rights. More specifically, it is familiar from discussions of claim rights, in the sense made familiar by Wesley Newcombe Hohfeld. These rights are commonly understood to be complexes of claims, privileges, and powers that are invested in agents, and that correspond to duties on the part of other agents.2 My right to a piece of property that has legitimately come into my possession, for instance, involves a claim against other people that they should not use it without my permission, where that claim defines a duty that others are under; it involves, furthermore, a permission to do with the piece of property as I wish, as well as a power to transfer my claims in it to others as I see fit. According to this way of thinking, those who make off with my property without my authorization will violate a claim I have against them, thereby transforming their relationships to me in a way they will not change their relationships to other parties. They will have not only have acted wrongly; they will also have wronged me in particular, providing me with what we might think of as a privileged ground for objecting to what they have done.
Hohfeldian rights of this general kind are a familiar part of our normative repertoire, deeply embedded in our thinking about (for instance) the structures of private law. Contracts, for instance, seem to generate a complex of directed duties and corresponding claims, and a similar structure of claim rights is arguably implicit in the law of torts.3 Whether there are similar claim rights at the most fundamental level of moral thought is a more controversial suggestion, one that has been questioned, in different ways, by consequentialists and proponents of certain virtue-theoretic views. Even those who are open to the idea that there are basic moral claim rights, however, naturally tend to think of them as constituting just a part of morality; the “realm of rights” that Judith Jarvis Thomson has written about, for instance, is seen by her as a sub-region within a larger moral territory. Thus, there are many moral duties that people seem to be under that do not correspond to any Hohfeldian claim rights in the familiar sense, including imperfect obligations of mutual aid, duties of gratitude, environmental imperatives, and sundry requirements of moral virtue.4
I agree that morality cannot be understood exclusively in terms of moral rights in the narrow, Hohfeldian sense. But it nevertheless strikes me as promising to interpret morality in relational terms, as a set of requirements on agents that are like the obligations of the Hohfeldian domain in being constitutively connected to claims that other individuals have against us. Relational elements are pervasive in many significant features of modern, secular morality, and it is my belief that these elements can be brought together into a comprehensive interpretation of the moral realm. The account I defend is not a theory of moral rights, as these are conventionally understood; rather it extracts a relational core from ordinary talk about rights and directed duties, and proposes that this relational structure can be extended, in an illuminating way, into a general framework for understanding the nature and normative significance of moral requirements.
The extension of the relational model that I shall defend, however, is not meant to capture everything that might intuitively be understood to be a reason or requirement of morality. There is a broad conception of the moral according to which it collects all standards of deliberate human conduct, whatever their source. In this broad sense, it is a moral defect if someone acts with disregard for the beauty of the natural world, or attaches importance to an activity that is out of proportion to its significance and value, regardless of whether the behavior in question otherwise affects the interests or welfare of persons or other sentient individuals. Moral failings, in this very capacious usage, are to be contrasted with deficiencies that do not directly involve the will, such as chronic bodily illnesses or infirmities that interfere with an individual’s biological good functioning.5 It is no part of my brief in this book to maintain that all moral standards, in this maximally capacious sense, are defined by relational requirements that are owed to individual claimholders. The relational account I shall develop is meant to capture a moral domain that is broader than the realm of moral rights, but narrower than the set of all standards that are applicable to the rational will. This is the intermediate domain, roughly speaking, that T. M. Scanlon has referred to as “the morality of right and wrong.”6 I shall call it interpersonal morality, to emphasize the fact that the standards that are in question derive directly from the effects of an individual’s actions on the interests and well-being of persons (where personhood is understood in a manner, still to be defined, that potentially diverges from membership in our biological species7). Interpersonal morality, in this intermediate sense, might be thought of as a set of requirements that reflect the fundamental insight that we share a world with other individuals whose interests are in some sense neither more nor less important than our own.
Not everyone who accepts this way of dividing up normative standards needs to agree about the desiderata to which an account of interpersonal morality is answerable. Some philosophers favor pluralist interpretations of interpersonal morality, tracing its requirements to fundamentally distinct values rather than imputing to them any underlying substantive unity.8 There is room for disagreement, as well, about the precise boundaries of this intermediate domain (disagreements, that is, about some particular requirements on human agency, and whether they are to be included within the domain). The approach I favor starts from the observation that interpersonal morality intuitively exhibits more specific normative features that have been neglected in recent treatments, and that a relational conception of obligation captures the underlying unity of the interpersonal domain that exhibits these important features. A consequence of this approach is that there might be some standards of rational agency that derive from the effects of action on other persons, but that are not standards of interpersonal morality as I understand it. Standards of this kind are not obligations in the sense in which the core requirements of interpersonal morality can be understood to be, nor do they function to structure accountability relations with other persons. They define forms of what I shall call extra-moral concern for moral persons—taking “morality,” as I shall do in what follows, to refer to the intermediate domain of interpersonal morality, rather than to the broad set of all standards that might apply to the rational will.9
My discussion will at many points raise more questions than can be answered in the compass of a single volume; as will become plain in what follows, the relational interpretation touches on many large issues of both theory and practice that are worthy of extended treatment in their own right. But I write in the conviction that it is important to have a sense of the big picture before we get too bogged down in matters of fine detail and nuance.10 Writers on moral philosophy frequently fall into a relational idiom when they talk about particular normative and philosophical issues. They assume, for instance, that individuals are typically wronged by behavior that is morally impermissible, and proceed to reflect on the implications of being treated in this way for the attitudes and behavior of the person who is wronged.11 But the relational interpretation, even when it comes naturally to us, is also philosophically distinctive; it is fundamentally opposed by some of the most influential traditions of reflection about morality, which treat moral requirements in individualistic rather than relational terms. There is need for an overview of the relational approach to the moral that highlights its distinctive features, so that we may better appreciate both the philosophical and normative advantages of understanding morality in these terms and the obstacles that stand in the way of such an interpretation. My hope is that the discussion in the present volume will go some way toward addressing this need.
In the present section, I would like to flesh out my initial sketch of the relational conception of morality by saying a bit more about what I take to be its basic elements.12 For purposes of exposition, it will be helpful to take as an example a case of a moral requirement that it is natural to understand as having a relational structure: that of promissory obligation. There are three important features that appear distinctive of cases of this kind, which I shall call directed obligation, claim, and normative injury; let us consider these in turn.
Relational norms, on an intuitive understanding of them, serve to ground obligations, specifying things that an individual agent must do. Thus, someone who makes a promise has undertaken an obligation, one that would not obtain in the absence of the promissory exchange. If the promise was to do X, then it seems, at a minimum, that the fact of the promise gives rise to a new reason for the agent to do X. That is, there is a consideration that speaks in favor of doing X that was not in place before the promise was made. But this understates the change in the normative situation that is effected by the promise. We normally think that promisors, in offering someone else promissory assurance about what they will do, are now under a duty to fulfill the promise they have made. Promisors exercise a normative power that is available to agents to create obligations where such were not antecedently in place, binding themselves to do what they have promised. Promises are indeed among the most salient and familiar examples of the phenomenon of moral obligation, even if they differ from many alleged obligations in being created through voluntary acts.
In saying this, I do not mean to suggest that it is always morally impermissible for promisors to fail to do the very thing that they have promised. Our understanding of the morality of promissory obligation implicitly acknowledges circumstances in which promisors do not have to follow the letter of their promises. Emergencies sometimes come up, for instance, that could not have been anticipated, even by a conscientious agent, at the time when the promise was made. Under these conditions, I think it would be natural to say that it is not wrong for the promisor to fail to perform the promised action. There may be some residual obligation that the promisor is under to provide compensation for losses that the promisee might have suffered in virtue of having relied on the promisor to do the thing that was promised; strictly speaking, however, the original promissory obligation (to do X, for example, if that is what was promised) no longer obtains under such circumstances. We might put this by saying that promissory obligation is not reasonably understood to be unconditional.13 It is, however, defeasible: so long as exceptional circumstances do not obtain, the promisor is under a moral obligation to do the thing that was promised, an obligation that was entered into through the promissory act.
It is further characteristic of these obligations that they have a built-in directionality.14 The promissory exchange brings into existence a normative nexus between the promisor and the promisee, whereby the former owes it to the latter to do the thing that was promised. Other people, who were not themselves parties to the promissory exchange, might well take an interest in whether the promisor fulfills the obligation. But this is not something that the promisor owes it specifically to them to do. The promise creates a special relationship between the promisor and the promisee, making it the case that the new obligation that obtains is directed specifically to the promisee. Indeed, as I shall argue more extensively in chapter 2, our sense that the promise creates an obligation is connected to the fact that it links the promisor and the promisee in a new normative nexus of this kind. A requirement that is owed to another individual in particular is not the exclusive property of the agent whose actions it governs; rather it is held in common by the two people whom it links. We are “bound” when we stand in a normative relationship of this kind, in the specific sense that the requirement that we are under, as agents, binds us to another party.
A directed obligation corresponds, on the side of the person to whom it is directed, to the notion of a claim. The party to whom the agent owes compliance with the obligation is someone who has a claim to such compliant behavior. Indeed, the claim in question has a built-in directionality that mirrors that of the obligation to which it corresponds; it is a claim that the party has against the agent to the latter’s compliance with the directed requirement. In the promising case, for example, it is the promisees who have a moral claim of this kind. We might understand this as an entitlement, held against the promisors, to their seeing to it that the promises are kept.15
Like the directed obligation with which it is linked, the claim that is held by the other party is not necessarily unconditional. As we saw above, the promise to do X does not generate an obligation on the promisor to do X under any and all possible circumstances that might eventually come to obtain. The claim on the part of the promisee, insofar as it corresponds neatly to the promisor’s obligation, is therefore likewise defeasible in character. It is a claim that the promisor should do X, barring unforeseeable conditions of the kind that would generally be understood to defeat the promisor’s obligation so to act. Just as the promisor might well have various secondary obligations under such circumstances, for example to provide compensation for losses suffered, so too would the promisee have claims against the promisor to those secondary performances.
A claim, in the sense that dovetails with a directed obligation, should be distinguished carefully from the notion of an interest. To have a claim against another party or parties that they should do X is not the same as having an interest in whether they will so act. Suppose that A has promised B to do X, and that there is another person, C, whose professional projects will be furthered if A in fact does X. Under these circumstances, C has an interest in A’s doing X (making C what is sometimes called a “third-party beneficiary” of A’s X-ing16); but it does not follow from this that C has a specific claim against A that A should do X. The directed claim, insofar as there is one in this case, resides in B, the promisee. Having an interest in someone else’s doing something is thus not sufficient for having a claim against that individual, in the sense that is here at issue.
It is perhaps more plausible to suppose that interests are at least necessary conditions for normative claims, but even here caution is required. When people have claims against other parties that they do X, it seems that there must be something in their situation and outlook, as individual agents, that provides a basis for their claims, and enables us to understand the claims as residing in them in particular. In some cases, this will be the fact that the claim-holders’ interests would be affected negatively if the other parties were to fail to do X; consider in this connection our claims against people not to be physically assaulted by them, which presumably have something to do with the effects of assault on our basic interests in bodily integrity, self-determination, and freedom from pain and suffering.
But the general relationship between interests and normative claims is more complicated than this example might suggest. Returning to the phenomenon of promising, take the case that Philippa Foot introduced into recent discussions: a Malay servant extracts from the anthropologist Mikluko-Maklay a promise that the latter will not photograph him, believing that having his picture taken would cause harm to his spirit.17 It seems plausible that the promise gives the servant a claim against Maklay not to be photographed by the anthropologist; and yet it seems that the servant would not really be harmed if Maklay were to break the promise surreptitiously (for example, by snapping some pictures of the servant while he is asleep).
Some have suggested that we have in this case a bare normative claim, one that is not grounded in any genuine nonnormative interests that are held by the promisee.18 But this seems to me a questionable inference. Even if the servant would not by hypothesis be harmed if Maklay were to photograph him, he has a legitimate interest in his own autonomy—in determining for himself, and in accordance with his own convictions, how others will make use of his body and person. He also has an interest in achieving assurance from others that they will respect his wishes about such matters of personal self-determination. There are thus important interests of the servant’s, in the sense of things that he takes a legitimate interest in concerning the character of his own life, that provide a basis for his claim that Maklay should keep the promise that was made in this case.
Generalizing boldly from this single example, it is tempting to conjecture that normative claims have to be anchored somehow in the interests of the claimholder, even if interests are not on their own sufficient to ground claims of the relevant kind. But is there anything in the relational interpretation of morality that would support this way of thinking about the bearing of interests on moral claims? What, more specifically, is the nature of the interests that are relevant to our specifically moral claims? And how exactly do we get from such interests to the determinate assignment of moral claims to individual claimholders, given that the bare possession of an interest does not on its own suffice to ground a normative claim? These are important questions for the relational interpretation of morality, to which I shall return in chapter 5 of this book.
In the meantime, I would caution against equating moral claims with the notion of moral rights. As I noted above, it is commonplace in the philosophical literature on rights to assume that at least some moral rights involve directed claims that are structurally like the claims that I have been discussing, insofar as they are held against other agents, correspond to directed duties on the part of those agents that are owed to the claimholder, and so on. If we are going to interpret the entire moral domain in relational terms, however, then we should not assume from the start that all moral claims involve what we would intuitively recognize as assignable individual rights. There are duties of gratitude, for instance, but it would be strange to say that these correspond to moral rights on the part of the potential beneficiaries of the duty; and we have duties to aid those who are in severe distress, without it being the case that specific rights are held by each of the potential millions of beneficiaries that each in particular should be assisted by us. We also have moral obligations to uphold and support valuable practices or conventions that we may have participated in, and yet these, too, do not seem to correspond to assignable individual rights against us.
Rights, I believe, represent a subclass of the normative claims that morality invests in individuals, but their distinguishing features are not present in all cases in which we have claims against others that they should comply with moral requirements. This will become a prominent theme in chapter 6 of this book, which addresses, inter alia, moral duties that have not seemed to be relational in character, precisely because they do not correspond to what we conventionally think of as individual moral rights.
The third element in the kind of relational conception that interests me is that of a normative injury. This concept is in place in situations in which agents have flouted the requirements that they stand under. Insofar as the requirement is genuine, specifying what people have to do, those who have violated it will have done something wrong. But if the requirement is a directed one, owed specifically to another party who has a claim against the agent to compliance, something further will be the case as well. The action that flouts the requirement will not merely be wrong; it will change the agent’s normative relationship to another individual, wronging the person to whom compliance with the requirement was owed. This is, as it were, the ex post facto residue of the individual’s claim against the agent to performance of the required action, in the case in which the claim has not been honored.
Just as normative claims are not to be equated with interests, normative injury is not the same as harm. Wrongful actions can have harmful effects on other persons without wronging them in particular, and those who are wronged by an action need not specifically be harmed by it, taking everything into account. Consider a case in which A promises B to stay away from a reception that B will be attending, but A ends up going to the reception all the same. A’s wrongful presence at the reception might end up being disadvantageous to the professional interests of a third person in attendance, C, but it wouldn’t necessarily follow that A had thereby wronged C. (Perhaps A ends up in a lengthy and mutually absorbing conversation with a potential client whom C had been hoping to cultivate at the reception, without it being the case that the promise was originally made by A to B out of consideration for C’s professional interest in having unimpeded access to any potential clients who might be in attendance.) By the same token, A’s action might wrong B, even if it leaves B better off on balance than B would otherwise have been. (Maybe B extracted the promise from A out of a concern for B’s own access to potential clients at the reception, but A’s presence and easy banter about fly fishing ends up facilitating B’s plan to cultivate new business relationships.) If, as I suggested above, moral claims can always be traced to some interest or other on the part of the claimholder, then disregard of moral claims will involve a slight of some kind to the claimholder’s interests. But it needn’t be the case that claimholders are harmed, on balance, by the actions that wrong them.
These points, or analogues of them, are familiar from the literature on rights, claims, and directed obligations. But there is a further aspect of moral injury that has seldom been remarked on, namely its dependence on the attitudes of the agent. Directed obligations, and the claims that correspond to them, do not in general involve demands on the attitudes of the agents who are bound by them. In the case we have just been discussing, where A promised B to stay away from the reception, A will count as complying with the obligation, and satisfying B’s literal claims, so long as A does not appear at the location of the event during the time when it is taking place; the motives out of which A satisfies this condition make no difference to the question of whether the claims and obligations have been satisfied. By the same token, A will have broken the promise if A shows up at the reception, regardless of A’s reasons for so acting (assuming that no unanticipated emergencies have arisen in the meantime, of the kind that are capable of modifying the original promissory commitment).
When it comes to moral injury, by contrast, things are otherwise. It matters to the question of whether A has wronged B with what attitudes A acted. When I introduced the notion of normative injury above, I associated it with cases in which agents have flouted a directed obligation that they stand under. But to flout a requirement is to act with a distinctive attitude toward it, one of knowing and even open disregard. Furthermore, given the constitutive connection between the directed obligation and the claim of another party, this attitude of disregard for the obligation is eo ipso an attitude of disregard for the specific claimholder. But to be treated with this kind of disregard is crucial, I submit, to the relevant notion of a normative injury; the wrong that has been visited on one who suffers such an injury consists, at least in part, in the attitude of indifference to or contempt for one’s specific claims.
If this is right, however, then there can be cases in which agents fail to fulfill the letter of the directed obligations that they stand under, without their having thereby wronged the individuals who had claims against them to performance. Having promised not to show up at the reception, A might end up putting in an appearance inadvertently, arriving at the gallery where it is taking place out of ignorance that it was relocated there by the organizers at the last minute. The promisee, B, would not have been wronged by A’s action under this scenario, precisely insofar as A’s action does not reflect an attitude of disregard toward B’s specific claims (though this might change if A does not leave the gallery as soon as it becomes clear that the reception is taking place there).
An attitude of disregard for another person’s claims seems in this way to be a necessary element in the analysis of moral injury. Whether such an attitude is also sufficient to give rise to a moral injury is a further question, which does not need to be resolved here. To think that it is would be to suppose that people might suffer moral injuries through actions that are not morally wrong. Thus, suppose A goes to the gallery, believing that that is where the reception will be happening, and not caring about the fact that she promised B not to attend, but it turns out that the reception was all along scheduled to take place in a different venue on the other side of town. Here B would naturally feel unnerved about A’s performance, if not outright resentful, despite the fact that A complied in the end with the obligation that was owed to B. But it is not clear that we would want to say that B was actually wronged by A’s behavior, strictly speaking. Perhaps people suffer moral injuries, in the relevant sense, only through actions that in fact violate duties that are owed to them; or perhaps we can allow that there are cases of moral injury that do not violate any duties that are owed to the person who suffers the injury.
Whichever way we come down on this question, however, it is important to note that there will be something unsettling to claimholders about agents who comply objectively with their claims, but without acknowledging them as important constraints on their behavior. The agent’s attitude of indifference to other peoples’ claims is something significant that these cases have in common with the central examples of moral injury, and there will be continuities in our responses to the different cases, however we decide to classify them in the end. This is an issue to which I shall return in chapter 4 of the present volume.
In the remainder of this book, I shall develop an interpretation of interpersonal morality as a domain of normative requirements that exhibit the three relational elements just sketched. Understanding morality in these terms, I shall argue, enables us to make sense of central features of the moral realm that are otherwise mysterious, while also shedding light on the character and content of the reasoning we engage in about particular moral issues.
The argument of the book begins, in chapter 2, with a discussion of the deliberative significance of moral considerations. The focus here is on the character of such considerations as obligations. Conclusions about what it is morally right or permissible to do enter the field of deliberation in a distinctive way, as presumptive constraints on the agent’s activities. It is an important desideratum for moral theory to make sense of this aspect of morality, which involves (as I contend) a sui generis form of normative relation. I show that a relational interpretation of morality can illuminate the force of moral considerations as obligations of this kind, ones that derive from the basic fact that we inhabit a common world together with other individuals. Duties that are owed to another party are paradigmatically suited to the distinctive deliberative role of presumptive constraints on a person’s agency. So if there are things that we owe to other persons just in virtue of their standing as persons, the ingredients will be in place for an account of interpersonal morality as a set of obligations or practical requirements on the will. In the course of developing this idea, I argue, further, that the resulting account is superior to the alternative theories of moral obligation bequeathed to us by the philosophical tradition.
In chapter 3, I turn to a different but equally important aspect of morality, which involves its social significance. Moral norms, I suggest, characteristically ground relations of accountability between individuals. Thus, we understand a group’s morality, in part, by identifying the norms whose violation attracts blame and opprobrium within their community; these are norms that members of the community hold each other accountable for complying with. I argue that the features that explain the standing of moral considerations as obligations should equally shed light on their suitability to structure interpersonal accountability relations of this kind. That is, the moral qualities that function normatively in deliberation as presumptive constraints on agency should equally provide other parties with reasons to adjust their attitudes and behavior toward the agent when the constraints are flouted.
I contend that the relational interpretation is uniquely equipped to render intelligible this interpersonal dimension of the moral. If moral obligations have their basis in the things that we owe it to other individuals to do, then to act with disregard for such considerations is to display disregard for other individuals, as persons who have claims against us. To flout moral obligations, so understood, is not merely to do the wrong thing, but to wrong someone else, causing that individual what I referred to above as a moral injury. But this is the sort of thing that gives the person who is wronged reason to resent the agent, in the way that is characteristic of relations of accountability. I show, further, that our accountability practices themselves have a relational deep structure. Reactive and other forms of blame, as well as the subsequent responses they set in motion, seem to presuppose a relational interpretation of moral requirements. Thus, wrongdoers are expected to apologize to those who have been wronged by what they did, and people in this position have a power to forgive wrongdoers that does not extend to other parties. We can make sense of these features of our interpersonal practices of accountability only if the moral norms that structure them are relational in character.
Chapters 2 and 3 together make a positive case for interpreting morality as a domain of directed obligations. The argument turns on the importance of elucidating the characteristic role of morality, both within individual deliberation and as a basis for a social practice of interpersonal accountability. Both aspects of this challenge have been somewhat neglected in recent discussions.19 To be sure, the general problem of normativity looms large in contemporary moral philosophy. But contributors to debates about this issue often ignore the distinctive role of moral considerations in practical reflection about what to do. Considerations of this kind are not merely reasons, in the now familiar sense of things that count for or against candidate actions that the agent might perform, but obligations, which structure reflection in a very different and more peremptory fashion. Similarly, the role of morality as a basis for relations of interpersonal accountability is crucial to human life, but it is often completely neglected in treatments of the nature of morality; whole treatises are written on this subject in which the topic of interpersonal accountability hardly comes up at all.20 The lived experience of morality is as a domain of considerations that make demands on us as agents, and that also have a special kind of importance for the attitudes of other parties toward us. My contention is that the relational approach is uniquely able to shed light on these neglected aspects of interpersonal morality, and that this constitutes a powerful presumptive case for understanding the moral domain in relational terms.21
In the not too distant past, philosophical accounts of morality were expected to establish in some way or other the authority of moral norms to govern the activities of the individual agents to whom they apply; at least that was the expectation if they were vindicatory rather than revisionist in tenor. The background was a climate of mild skepticism about independent normative notions, and a corresponding feeling that moral requirements required some special justification if they were to succeed in prescribing what the agent is to do.22
More recently, the intellectual climate has evolved somewhat. Many contemporary moral philosophers are willing to take for granted some form of nonreductive realism about the normative in their philosophical investigations. Those who are not tend to focus on the metaethics of normativity in general, not the credentials of moral reasons and requirements in particular. The idea that there might be a special problem that is posed by the normativity of the moral starts to seem peculiar against the background of these assumptions. But I think there is a special problem here, one that it is important for a philosophical account of morality to address.23 In particular, the challenge is to make sense of the fact that considerations of moral right and wrong have two very distinctive kinds of normative significance: they represent obligations or practical requirements in the first-person perspective of deliberation, and they also structure our interpersonal relations of accountability. My initial case for the relational approach is based on its success at meeting this important challenge to moral theory.
Every philosophical investigation takes some things for granted. In developing my argument for the relational approach, I shall not attempt to vindicate normative commitments in general in the face of skeptical or naturalistic worries about them. Rather, I shall assume that we can safely operate with normative notions of various kinds, and address the more specific questions sketched above, about how moral theory can make sense of the distinctive normative bearing that morality has both on individual deliberation and on our social relations. These issues, like many others in philosophy, come into sharp relief only when we focus on them at the right level of resolution. Zoom out too far—for instance, by taking up the metaethical project of placing normativity in general within the larger landscape of natural objects and processes—and the specific normative features that distinguish interpersonal morality become indistinct. These features also disappear when we zoom in very closely on the fine structure of moral requirements, as happens with many investigations in so-called normative ethics, where fantastic variants on hypothetical cases are constructed in order to elicit intuitions about the morality of right and wrong.24 Ethical theory, as I shall be pursuing it here, cannot ignore questions in metaethics or normative ethics; but it equally cannot allow the pursuit of these questions to pull us out of the intermediate range within which the distinctive normative features of interpersonal morality become both visible and puzzling.
The background framework that I shall adopt for purposes of this intermediate investigation does not merely countenance normative commitments in general; it allows that there can be irreducibly distinct forms of normative relation.25 Philosophers sometimes assume that normativity is exclusively a matter of reasons, where reasons are in turn considerations that count for and against attitudes and actions. But as noted earlier, this idea does not seem to fit very well with the idea that there are obligations. Considerations that practically require an agent to do something seem to figure very differently within deliberation from the kind of reasons that merely count in favor of doing the same thing. Similarly, reasons for the reactive attitudes, such as resentment or indignation, seem different in kind from the considerations that count in favor of actions we might perform; the former are considerations that render the attitudes fitting or intelligible, whereas the latter have to do with the various ways in which actions might be valuable or worthy of pursuit. My suggestion will be that the relational account is well suited to elucidate the different kinds of sui generis normative significance we attribute to morality: its standing as a source of practical requirements on the individual will, as well as its role in making it fitting or appropriate for others to respond to infractions with reactive and other forms of blame.
Directed obligations and the claims that are connected with them might be understood to constitute a normative nexus, and this is how I shall speak of the relational elements in the theory of morality that are my main target in this book. They represent a normative nexus, just in the sense that their elements have the different kinds of normative significance for agents and for claimholders that were distinguished in the preceding paragraph. But idea of a relational nexus itself—of a complex of directed obligations and claims that are linked to them—is a further element in the larger theory that I shall largely take for granted in what follows. I assume there is a range of familiar cases that we all intuitively understand to have this kind of relational structure, including the example of promissory obligations that recurs throughout the book, as well as numerous other examples from private law, institutional practice, and even games of various kinds. My aim will be to draw out some of the significant features of a normative nexus of this familiar kind, and to defend and develop the suggestion that a nexus of the same kind can be understood to constitute the deep structure of the realm of impartial morality. But I shall not endeavor to provide a reductive account of relational structures of this general type (one that might, for instance, attempt to identify nonrelational elements that together constitute necessary and sufficient conditions for the obtaining of such a structure).26
To summarize this catalogue of defensive stipulations, I shall develop a normative argument for the relational interpretation of the moral, but one that differs from approaches familiar in the philosophical tradition. I shall not attempt to demonstrate that all agents have reason to comply with relational moral requirements, nor do I have a transcendental argument to offer, to the effect that a commitment to relational obligations is implicit in the structure of rational agency (for instance, as a constitutive condition of its possibility). Rather, taking basic normative concepts largely as given, I shall focus on some of the things that set moral norms apart from normative notions of other kinds, including their standing in deliberation as practical requirements and their significance to our practices of interpersonal accountability. The gist of the argument is that we will best be able to understand these aspects of the moral domain if we think of it as a set of relational obligations that link us with other individuals in a pairwise normative nexus.
The normative argument of chapters 2 and 3 has a conditional structure. The general suggestion is that, if the relational model can be applied to the entire moral domain, the result will be an interpretation of it that renders fully intelligible its distinctive normative features. But it still needs to be shown that the relational model can be extended in this way. This is the task of the remaining chapters of the book. In chapter 4, I discuss some general issues that are raised by the ambition to interpret all moral obligations in relational terms. The most salient and familiar examples of directed duties arise from transactions and other forms of causal interaction between the parties that they link. A promise comes into existence through some kind of exchange between the parties to it, and duties of gratitude are created when a kindness is bestowed by one person on another; similar patterns of nonnormative interaction are found in relationships between family members and friends, which characteristically give rise to networks of claims and directed obligations.
Against this background, it is natural to wonder whether directed obligations presuppose antecedent personal relationships. If so, this would prevent the application of the model to the entire moral domain, as interpersonal morality, on the modern conception of it, is meant to define obligations that govern our conduct toward people with whom we have never before interacted. A complementary line of thought is suggested by reflection about the evolutionary history of morality, which plausibly first emerged as a set of tendencies that encourage cooperation in contexts involving close and sustained social interaction. The relational model that seems well suited to contexts of this kind might not provide a plausible framework for thinking about our duties in cases that do not involve face-to-face interaction or tribal identity.
I argue, in response to these questions, that some reflective extension of more elementary moral resources is required in any case, if we are to devise principles that are adequate to regulate our conduct in the full range of situations that must be covered by a comprehensive moral scheme. A scheme of this kind is one that acknowledges the basic modern insight that morality is a cosmopolitan phenomenon, regulating our relations to a maximally inclusive notional community of individuals whose interests are considered equally important. I suggest that there is a strong presumptive case for extending the relational model into a comprehensive framework that applies to this extensive notional community, given the ability of the model to illuminate the normative features that are distinctive of the moral realm. Doing this requires that we think of morality as a set of self-standing directed obligations, which are not grounded in any antecedent relationship that the parties to them have with each other. I consider and reject some general arguments that have been advanced for thinking that there could not be a self-standing normative nexus of this kind. I also explore some of the metaethical presuppositions of the moral nexus, construed in this way, which include an anti-individualist conception of normativity. Individualists might not be able to accept that there are self-standing relational obligations that link agents and claimholders in a common normative structure; but I show that they will equally be skeptical about ideas that are basic to any cosmopolitan conception of morality, including ideas that the relational approach is especially well equipped to illuminate.
The chapter concludes with some further reflections about the agent-relativity of moral obligations and about the values enabled by relational morality. It might initially seem that obligations are agent-relative in character in virtue of the fact that they are grounded in patterns of historical interaction between individuals, which serve to distinguish the people to whom agents specifically owe compliance with moral obligations from others who merely stand to be affected by what the agent does (perhaps in ways that also involve the agency of third parties). But this is not correct. The deeper feature that explains the agent-relativity of many standard moral requirements is simply their directed character, which connects them constitutively to the claims of other individuals. This same feature of relational morality sheds light on the values that are enabled through compliance with moral requirements, which I suggest should be understood in terms of an ideal of interpersonal recognition. We understand it to be a valuable thing to relate to others in a way that serves to realize this ideal in our own conduct, and this can help us appreciate the contribution that morality can make to the goodness of the agent’s own life.
In chapter 5, I take up some questions about the structure and scope of relational morality. If the moral nexus is construed as a domain of self-standing relational obligations, the question arises as to the manifold of individuals who are potentially linked under such obligations. Prior relationships or ties are not preconditions for linkage under the relational duties and claims of morality, so what other principles might enable us to decide the issue of inclusion in the set of moral persons? I suggest that we begin by thinking of this class as including those individuals who are capable of entertaining relational moral thoughts. This group includes all normal adult members of our species, but it might conceivably include other individuals as well. I then consider various ways in which the cosmopolitan manifold of persons might be extended outward from this core, including possible extensions that result in asymmetries among the members of the manifold (where, for instance, claims are assigned to individuals who do not themselves stand under corresponding directed obligations).
It is implausible to think, however, that there are concrete duties that we owe to all of the members of the maximally extensive class of moral persons. Some individuals, for instance, lie outside our temporal or spatial reach. In thinking about the specific conditions that have to be satisfied before a concrete nexus of moral duties and claims can be in place, I suggest that we focus on the effects that our agency can potentially have on the personal interests of other individuals. Personal interests in the relevant sense are interests people have in how their own lives go. I go on to propose that interests of this kind will play a prominent role in justifications for the assignment of concrete moral claims to an individual, and that moral reasoning can be understood, in general terms, as a movement of thought that takes us from personal interests to the identification of claims. The resulting picture differs from some conventional conceptions of moral rights and duties, which conceive of these things as inputs into reflection about what it is morally permissible to do, rather than outputs of specifically moral thought.
Chapter 5 ends with a discussion of the prospects for a theoretical account of relational morality. I note that the relational approach might be advanced in an intuitionistic variant, which would hold that the movement from personal interests to claims is effected through an exercise of particularistic judgment that cannot be subsumed under any general procedures or principles. While acknowledging this possibility, however, I also think that there is a promising moral theory that can be interpreted as offering an account of morality’s implicitly relational structure. That theory is moral contractualism, in the form familiar from the work of T. M. Scanlon. Contractualism offers an account of moral reasoning, describing what we might think of as a general template for extracting assignable moral claims from the personal interests of those who might be affected by an agent’s actions. It can also be understood as a substantive conception of morality that specifies, in illuminating terms, what it is for a nexus of moral directed duties and claims to be in place between two individuals. Its relational character is, in my view, essential to the power and plausibility of contractualism as a substantive moral theory, something that has been lost sight of in recent debates about the role of personal interests within contractualist reasoning.27
The final chapter addresses some first-order practical implications of the relational interpretation. As noted above, there are some important moral duties that do not intuitively seem to be grounded in the claims of individuals. Examples include duties to future generations, where the identity of the people who will exist in the future depends on what we now decide to do; imperfect duties of mutual aid, where our actions benefit people who do not have specific claims to the goods that we provide; and situations in which the numbers count for moral reflection, which have seemed difficult to make sense of in terms of moral requirements that are owed to other individuals. I show that the relational approach can plausibly be extended to challenging cases of these kinds, though only by modifying the familiar idea of a moral claim.
The chapter begins with a discussion of some of the most familiar examples of relational obligation, which I show to be very diverse in character. Some, but by no means all of them, involve transactions through which we understand ourselves to have incurred a literal or figurative debt that stands to be repaid. But they all involve situations in which our actions have potential effects on individuals who are foreseeable at the time when we act. I go on to consider a recent suggestion to the effect that foreseeability of this kind is not a necessary condition for an action to count as one that wrongs another party. Though the suggestion seems implausible to me, I note that it could be accommodated by postulating secondary moral claims not to be harmed by the wrongful agency of another. Claims of this kind would be unlike paradigm moral rights; but on any account of relational morality, there will be numerous claims of this sort. These include the considerations involved in cases of so-called imperfect moral duty, which leave agents with considerable discretion as to how they are to be fulfilled. I discuss the cases of gratitude and mutual aid, suggesting that claims are in place even here, though they are not necessarily claims whose satisfaction would redound to the benefit of the claim-holder in particular.
The remaining sections of the chapter offer an extended discussion of some of the moral situations that have traditionally been thought especially difficult to accommodate within a basically relational framework. A particular crux is the significance of numbers and aggregation for moral thought. I recommend a form of relational morality that gives prominence to the ex ante objections that individuals might have to principles for the general regulation of behavior, and observe that it provides resources for incorporating aggregative considerations within the framework of what we owe to each other as individuals. There remain, however, some cases that cannot be understood in these terms, where philosophers have maintained that aggregate well-being has independent importance for practical thought. I note that intuitions about this matter may trade in part on a failure to distinguish between contexts of individual agency and of collective decision-making about matters of democratic public policy. I also argue that the independent importance of well-being for individual agency, to the extent it persists, should be understood to involve requirements that conflict (at best) with those of relational morality, rather than undermining or outweighing them, as some have alleged that they do.
It is not my aim in chapter 6 to resolve the thorny practical questions that are there discussed. The objective is instead to explore the resources of the relational approach for understanding the moral dimension of these important questions. As noted earlier, many contemporary discussions in normative ethics proceed through constructing elaborate hypothetical cases, which are meant to test our intuitions about what it is right or permissible to do, but in a way that is independent of theoretical debates about the nature of moral rightness or permissibility.28 One implication of the earlier chapters of this book is that this is a questionable strategy. It is not at all clear that there is a stable and convincing conception of moral rightness, one that is suited to the deliberative and interpersonal roles that a conception of this kind may reasonably be expected to satisfy. Until a plausible such conception is articulated and developed, reflection on elaborate hypothetical cases threatens to become undisciplined, appealing to elusive ideas about morality that we do not fully understand. It is unpromising, for instance, to argue about what we have “moral reason” to do, or about which of various hypothetical outcomes would be “better (from the moral point of view),” until we know what conceptions of moral reasons and moral value might be in play.29 So one objective of chapter 6 is to situate some of the controversies in normative ethics within the context of an independently compelling conception of what morality is about in the first place.
Interpersonal morality, according to the relational interpretation of it, collects a set of requirements that derive from the inherent challenges of our social life. These requirements constitute a coherent domain of interpersonal obligations and claims, specifying what we owe to each other insofar as we are moral persons who stand to be affected by exertions of each other’s agency. Chapter 6 argues that this way of thinking about interpersonal morality provides a fruitful framework for understanding what is at stake in some of the challenging practical questions that are discussed. Seen in this light, the argument of chapter 6 dovetails with the earlier chapters of the book, which identify important normative features of our moral practices that are fully intelligible only on the relational interpretation of them, and which show how the relational framework can be extended to encompass our relations to anyone whose interests might conceivably be affected by what we do.
The book as a whole might thus be thought of as making an interpretative case for the relational account.30 It sketches a distinctive approach to understanding the unity of the moral realm, highlighting normative and practical features of morality that best make sense when morality is understood in these terms. The burden of the argument is not to counter the kind of skepticism about morality that questions whether we can have reason to do things that do not directly promote our own welfare and interests; nor do I aim to establish that rational agents have to think of themselves as subject to relational moral requirements, on pain of inconsistency or some other form of rational instability. My discussion is targeted at those who are prepared to accept the basic idea, already mentioned several times, that no individual is either more or less important than any other.31 This is a characteristically modern idea, as it seems to me, one that may not have been fully acknowledged at all times and places, even by agents who were otherwise thoughtful and conscientious. But for most of us today, it is an idea that seems extremely difficult to deny.32 Even those who are willing to take on board this substantive commitment, however, might reasonably wonder how the postulate of equal standing could give rise to something recognizable as interpersonal morality: a set of obligations on agents that are also suited to structure accountability relations with other persons, and that provide an illuminating framework for negotiating first-order questions about the ways in which it is permissible for them to conduct their lives.
Thus, consequentialism in some form strikes us as an exceptionally natural way of thinking about morality, once we take seriously the idea that the circle of moral concern must expand beyond the boundaries of parochial attachment, so that we grant the interests of each individual to be equally important.33 As T. M. Scanlon has written, consequentialism is for a wide range of people “the view towards which they find themselves pressed when they try to give a theoretical account of their moral beliefs.”34 If the interests of all persons matter equally, then it is very tempting to suppose that morality will enjoin us to maximize the impartial good, taking the good of each to be no more or less important than the good of any other. But the resulting interpretation of morality is also deeply problematic. It notoriously delivers verdicts about many questions of normative ethics that are wildly at variance with our considered convictions, failing (among other things) to acknowledge the significance for moral thought of the differences between persons and the relation between their agency and their own projects and attachments.35 It also, as I argue in chapters 2 and 3 below, leaves us without resources to understand the character of morality as a set of obligations on agents that at the same time structure relations of accountability with others.
Some have tried to make of such apparent deficiencies a kind of theoretical virtue, arguing that secular moral philosophy is a subject in its infancy, and that it is only to be expected that the comparatively recent insights it attempts to accommodate might lead to radical revisions in received ideas about its nature and first-order consequences. It can be thrilling to think of oneself as embarked on an inquiry that is in this way untethered from conventional wisdom.36 A different and no less plausible response is that the project of modern moral philosophy would be a failure if consequentialism (or something similar) is what becomes of it when we try to take seriously the postulate of the equal standing of everyone.37 Both of these attitudes—the heroically optimistic as well as the pessimistic and gloomy—take for granted that central features of our interpersonal moral practices will not survive the attempt to adapt them to the cosmopolitan insight that the interests of all persons matter equally. But perhaps this common assumption is unwarranted.
I hope to show that the relational approach offers a promising paradigm for thinking about the significance of people’s interests for our agency. It distills those interests into a universal normative system whose features align with our reflective understanding of interpersonal morality, construed as a unified set of obligations that equally function to structure relations of accountability. To adopt this interpretation is to see ourselves as standing in a distinctive relation to the other members of a notional domain of equal moral beings, acknowledging that we are linked to each of them through a moral nexus of directed obligations and corresponding claims. We become intelligible to ourselves, as agents who are subject to genuine moral requirements, and who rightly hold each other accountable for living up to them, when we think of the realm of interpersonal morality along these lines; we also gain insight into the first-order structure of this realm when we understand it as a collection of duties that are owed to others, just insofar as they are persons with equal standing. That, in essence, will be my argument for the relational approach.