Hohfeld’s concept “duty” is a concept that attaches to two people: there are in every case of its application the one who is under the duty and the one toward whom he or she is under it. We might summarize this in the words: it is a two-hat concept. “Ought” is not a two-hat concept; we might call it a one-hat concept.
- Judith Jarvis Thomson (Reference Thomson1990, p. 62)
1. Introduction
A body of work has emerged in anglophone philosophy in the last 20 or so years that has been dubbed relational ethics. Footnote 1 This label is intended to stick to a family of views in metaethics and deontological normative ethics which hold that ethical reasons are in some fundamental sense an expression of agents’ relatedness to one another. In broad terms, relational ethics sees a central sphere of practical reasons as comprised of facts about what agents owe to one another. This view is a rival to consequentialism, and to traditional, non-relational versions of deontology.
As with many labels for intellectual movements, there is a diversity of views among so-called relational ethicists. There are differences over what explains the existence of interpersonal duties; over what kinds of things those duties are; over which duties there are; over whether we owe our moral conduct to everyone, or in the first instance just to those whose rights and interests are at stake in our action; and so on.Footnote 2 Given this diversity, a challenge for anyone who wants to use the label “relational ethics” is to articulate what unifies the field.
One compelling candidate for a unifying characteristic is what I will call the irreducibility thesis. This is the thesis that there are some practical reasons that are irreducibly relational. Exactly what it means for a reason to be irreducibly relational is something I will seek to clarify shortly. Roughly, it is for a reason to be such that it cannot be expressed except with reference to some relation between the agent for whom it is a reason and some other person or people. For example, one might think that a reason to keep a promise is essentially an expression of the relation between promiser and promisee, so that to lose sight of that relation would be to lose sight of the distinctive kind of reason that the promise purported to create. In different ways, the irreducibility thesis has been defended by three high-profile figures on whose work I will focus my discussion: Stephen Darwall, Margaret Gilbert, and R. Jay Wallace.
In this article, I argue that the normative force of any reason must be explained by the value of the action for which it is a reason, so that the irreducibility thesis is false. This claim is consistent with recognizing some ways that relations between agents ground the value of various actions. As such, my goal is not so much to reject the insights of relational ethics, which are many, but rather to clarify their significance for the theory of practical reason. In Section 2, I will, as already promised, explain in better detail what is meant by the irreducibility thesis, and draw out what has made it an attractive view for relational ethics. In Section 3, I will elaborate on the quite general reason just mentioned for thinking that the thesis must be false. In Section 4, I will consider the most salient line of reply to my argument, which invokes the priority of the right over the good.
2. The Motivation for the Irreducibility Thesis
In the work of relational ethicists, the irreducibility thesis has been advanced in different contexts. In summarizing three of these approaches, I aim to draw out the common motivation of the thesis. All three philosophical enterprises seek to explain the normative basis of a certain class of reasons. For Darwall and Wallace, that class is the set of moral obligations, and so in that respect they are contributing to the tradition of seeking to account for the force of morality. Gilbert’s project is somewhat different—on which, more below.
In The Second-Person Standpoint, Darwall argues that moral obligations are grounded in the authority that members of the moral community have to hold one another accountable for compliance with moral norms. Part of his defense of this view involves the irreducibility thesis (Darwall, Reference Darwall2006, p. 12):
These notions—second-personal authority, valid claim or demand, second-personal reason, and responsibility to—therefore comprise an interdefinable circle; each implies all the rest. Moreover, I contend, there is no way to break into this circle from outside it. Propositions formulated only with normative and evaluative concepts that are not already implicitly second-personal cannot adequately ground propositions formulated with concepts within the circle.
The quoted passage expresses a commitment to the irreducibility thesis because it holds that there are some reasons (those which Darwall calls “second-personal reasons”) that are irreducibly relational in the sense that they are not explained by non-relational facts, such as facts about what is valuable. I will return shortly to discuss what irreducibility amounts to, once the commitment of the other relational ethicists to some version of the thesis is made clear.
In The Moral Nexus, Wallace (Reference Wallace2019, see especially chapter 2) pursues a similar explanandum to Darwall: namely, he sets out to explain the basis and force of moral obligation. Rather than being grounded in the “representative authority” that members of the moral community have over one another, Wallace argues that any given moral obligation is grounded in (or, better, arises from the same relational nexus as) the bipolar claim rights of some particular other.
On Wallace’s view, moral obligations (or “practical requirements”) are sui generis expressions of a way that agents are related to one another. Like Darwall’s second-personal reasons, a practical requirement is “distinct from, and not reducible to, the aspirational form of normativity that is at issue with ordinary recommending reasons, of the kind that count for and against candidates for action” (Wallace, Reference Wallace2019, p. 29). Rather, moral obligations have an “inherently relational structure,” in which (Wallace, Reference Wallace2019, pp. 49–50):
one’s reasons for doing something are constitutively connected to claims to performance on the part of another person. The values which these reasons are based on are not purely monadic; they do not exclusively concern the agent, but essentially implicate the person to whom the agent is related, whose own normative situation will be altered if the agent fails to respect the value that is at issue. We might say that they are held in common by two different parties, the agent who stands under the directed duty and the claimholder to whom it is directed. […] The reason in question is part of a normative complex that does not belong to agents alone, and so it isn’t for agents to determine how much weight it is to be given in their practical deliberation.
Wallace (Reference Wallace2019, p. 3) holds that moral obligations are directed duties, that is, actions the performance of which is owed by an agent to a claimholder. On his analysis, the reasons expressed by directed duties are not reducible to the value of the actions for which they are reasons. It is in this way that Wallace is committed to the irreducibility thesis.
Unlike Darwall and Wallace, Gilbert’s relational ethics is not committal over whether the set of irreducibly relational reasons includes moral obligation. Her project seeks to account for the normativity of all directed obligations, irrespective of whether moral obligations are among them. However, like Wallace, Gilbert advances an analysis of the nature of directed obligations according to which their normative significance (or force, or basis) cannot be explained except by reference to the relation between the agent who faces the directed duty, and the claimholder toward whom the duty is directed. She says (Gilbert, Reference Gilbert2018, p. 55):
[I]n spite of the inclusion of the word “duty” in the phrase “duty towards” the referent of that phrase should not be assumed at the outset to be, or to include, what I have called a plain duty. Though this suggestion runs contrary to the usual practice of rights theorists, it has some initial plausibility. On the face of it, a directed duty is a relation that holds between two parties. In other words, a directed duty, as such, appears to be internally relational. A plain duty as such is not internally relational, nor can the fact that Y’s plain duty relates to X in some particular way convert that plain duty, as such, into an internally relational duty.
It is by virtue of this interpretation of the nature of directed duties that Gilbert too is committed to the irreducibility thesis. Her view holds that the kind of reason agents face when they face directed duties are such as not to be reducible to any more basic normative fact than the relation between the relevant agents.
It seems, then, that Darwall, Wallace and Gilbert endorse the view that there are some practical reasons that are irreducibly relational. This much is vague on the question of what it means to say of a reason that it is irreducibly relational, or, more fundamentally, what it would be for a reason to be reducible. There are competing interpretations of this thought, so it will be worth spelling those out in order to clarify the nature of the thesis to which the relational ethicists are committed.
One interpretation of the concept of reduction is semantic. On this conception, for X to be reducible to Y is just for it to be the case that propositions about X can be expressed without reference to X, but instead with reference to Y. For example, it is in this semantic sense that one might say that the distance in miles between two places is reducible to a distance in meters. Similarly, it is this sense of reduction that is invoked in mathematics when an expression is said to be reduced from a more complex to a simpler form.Footnote 3
A competing interpretation of the notion of reduction, rather than being semantic, is metaphysical. On this conception, for X to be reducible to Y is for it to be the case that propositions about X are really just a matter of propositions about Y. For example, one might think that facts about thermal energy are reducible to—in the metaphysical sense of being really just a matter of—facts about kinetic energy. Metaphysical reduction is a kind of explanation. To show that X is reducible to Y in the metaphysical sense is to reveal something substantive about X that makes possible a deeper description of X-phenomena.
If something is reducible metaphysically, it must be reducible semantically too. For example, suppose that we are entertaining the hypothesis that the spooky events in the house are really just a matter of the doings of a single malicious agent. If this hypothesis about the spooky noises is correct, then the sentence “the spooky events are happening again,” must entail that “the doings of the malicious agent are happening again.” If the first sentence does not entail the second, then it turns out that the spooky events are not—or at least are not all—really just a matter of the doings of a single malicious agent. With this point noted, we can ask which of the following two versions of the irreducibility thesis is the one at the heart of relational ethics:
Semantic version Certain practical reasons cannot be formulated as reasons except with reference to some relation between agents.
Metaphysical version Certain practical reasons are such that it is not the case that they are really just a matter of anything other than some relation between agents.
Because the semantic version entails but is not entailed by the metaphysical version, the latter is a more modest claim. In the remainder of this section, I wish to discuss the basic motivation for the irreducibility thesis. In so doing, I will argue that—despite some suggestions to the contrary—it is this more modest claim, the metaphysical version of the thesis, that is at the heart of relational ethics.
I have noted that relational ethicists are united in seeking to account for the normative force of certain reasons, where those reasons apparently involve a relation between agents. A paradigmatic example of such a reason is the reason A has to φ, when A owes it to B to φ. In Darwall’s view, this might be a moral responsibility that A owes to B in B’s capacity as a representative member of the moral community; in Wallace’s, a moral responsibility owed to B in B’s capacity as the individual claimholder; in Gilbert’s the reason, moral or not, may be the product of a joint commitment between A and B as individual agents. In each case, there is some reason for action the normative force of which seems to have something to do with the relation between A and B.
A question arises here of inherent philosophical interest. What is the “something” just mentioned? What exactly do reasons have to do with the relevant relations between agents? It is to this type of question that relational ethics offers an answer. The answer is the irreducibility thesis. That is, the core claim of relational ethics is that what explains the connection between certain reasons for action, and certain relations between agents, is that such relations can themselves constitute reasons, and can do so in such a way as to be irreducible to any other level of explanation.
As noted, irreducibility claims can be made with respect to descriptions of phenomena. The ambiguity between semantic and metaphysical interpretations can emerge in many cases. A first step toward resolving the ambiguity in any such case is to consider the most salient candidate level of description to which the initially given description might be thought to reduce. For instance, a theistic response to the Euthyphro dilemma might hold that facts about the will of the gods are irreducible. One might reasonably wonder whether this claim is semantic or metaphysical. The claim may be best understood as the claim that there is no level of description, Y, with which descriptions of facts formulated in terms of the will of the gods (X) could be substituted to yield an adequate alternative expression. Alternatively, the claim may rather best be understood as the more modest point that whereas the prior level of description, X, may be expressible in some other description, Y, it is not the case that the phenomena described in terms of the will of the gods are really just a matter of something else that could be captured in a contrasting vocabulary. To evaluate which interpretation is preferable, it is helpful to hold two things in view: the most salient candidate for a level of description, Y, to which X might be said to reduce; and the explanandum of which X is held to be an explanans.Footnote 4
In the case of relational ethics, I submit, the most salient candidate description to which facts formulated in terms of relational reasons might be supposed to reduce is the same across the gamut of views in relational ethics. The candidate description is simply in terms of value. Darwall mentions “normative and evaluative concepts that are not already implicitly second-personal”; Wallace mentions “the aspirational form of normativity that is at issue with ordinary recommending reasons”; and Gilbert mentions “plain duty.” Each of these are candidate levels of explanation to which relational reasons might be thought to reduce, but where that is what is being denied by the relational ethicists. To simplify things, we can see each of these candidates for X as being ways of referring to the one principal candidate: practical value. By “practical value” (or otherwise just “value,” or “goodness”), I will mean the quality of an action’s being to be done.Footnote 5 The core claim being made in the irreducibility thesis is that the fact of A owing it to B to φ is not reducible to the fact that φ-ing has a certain value or to-be-done-ness.
This core claim does not require the semantic version of the irreducibility thesis. The explanandum to which the irreducibility thesis is intended to be the explanans is that relational reasons have a certain deliberative significance, or normative force. The semantic version of the irreducibility thesis holds that there is no way to formulate a relational reason which conveys its normative significance but does not refer to the relevant relation between agents. Since their explanandum is about the normative force of a certain description, the question raised by the semantic version of the irreducibility is not strictly relevant to the core claim of relational ethics. Relational ethicists may independently hold the semantic version, but this is not the version that is core to their view qua relational ethicists. If a description of a reason could be expressed in some other level of description, taken alone this would have no bearing on whether the initial description explained the normative force of the reason.
By contrast, of course, the metaphysical version of the irreducibility thesis is crucial to the explanatory claim that relational ethics makes. If one were to hold that A’s reason to φ when A owes it to B to φ is really just a matter of the value of φ-ing, then this would be to deny the core claim of relational ethics. Indeed, this denial of relational ethics is entailed by rival analyses of moral obligation. Consequentialism holds that A’s reason to φ is really just a matter of the value of the states of affairs brought about by standing by this duty to B. Kant’s deontology holds that reason is really just a matter of what is required by pure reason. Natural law theories hold that it is really just a matter of what is called for by the natural order of things. Divine command theories hold that it is really just a matter of God’s will. It is in opposition to these views that relational ethicists posit the irreducibility thesis as an analysis of the normative force of apparently relational reasons.
I have sought to clarify the irreducibility thesis and convey the motivation on which relational ethicists are drawn to the thesis. A clarified version of the thesis is the view that there is no alternative level of description, Y, such that the normative force of relational reasons is really just a matter of Y. In particular, it is not the case that the normative force of relational reasons is really just a matter of the value of the actions for which they are reasons. The motivation for this claim is that, if true, it explains the normative force of relational reasons and the apparent contrast between the role that such reasons play in practical deliberation, and the role played by other, non-relational reasons.
In the next section, I will argue that it must be the case that the normative force of relational reasons is really just a matter of the value of the actions for which they are reasons.
3. The Commensurability of Practical Reason
The argument that I will discuss that speaks against the irreducibility thesis represents a general and simple case against any version of the thesis. After presenting the case, I will consider a central line of reply.
The simple reason that the irreducibility thesis must be false is that it is contradictory. The thesis holds that there are some facts (namely, deontic interpersonal relations such as “that you owe it to me to φ”) which are reasons for actions, but which are not facts about the value of those actions. But reasons for action just are facts about the value of actions. Naturally, proponents of the irreducibility thesis do not agree with that last point. I will offer a rudimentary argument in its support. While I do not expect this argument to persuade those who already doubt its conclusion, I do think that the argument transfers the burden of explanation onto those who do not share this conception of practical reason.
The rudimentary argument goes like this. Practical reasoning is the process of deciding what to do. Practical reasons are considerations that figure in practical reasoning. The way that they figure is that reasons favor actions—sometimes competing reasons favoring competing actions—and the deliberating agent compares the reasons in such a way as to yield a choice about which action is to be done.Footnote 6 “Value” refers to the quality of an action in virtue of which it is to be done.Footnote 7 So, practical reasons must be comparable with one another in terms of the extent to which the actions they recommend are to be done. This just is their being comparable in terms of the value of actions that they favor. So, reasons for action just are facts about the value of actions.
What I have just presented amounts to the outline of a theory of practical reason.Footnote 8 It specifies definitions of some key terms—value, reasons, practical reason—and, using those terms, the theory offers an explanation of what it is to choose what to do, and how reasons figure therein. Choosing what to do is comparing reasons for action. Decisions about what to do are the conclusions of such comparisons. It is incumbent on anyone who wishes to reject this theory of practical reason to provide an alternative explanation of what it is to choose what to do, and how reasons figure therein. This is the explanatory burden that I think my rudimentary argument transfers to proponents of the irreducibility thesis.
Consider the following example. Suppose that Alys and Bryn are elected representatives for a union of security staff and a union of cleaners, respectively, preparing for a high-stakes negotiation meeting with their shared employers in the context of ongoing strike action by both unions over the terms and conditions of their employment. Ahead of the meeting, Alys and Bryn go for a walk. Bryn says that he is contemplating holding out. That is, he is considering insisting, in the meeting, that his members, the cleaners, would not consider balloting on any deal that failed to meet certain stringent demands of theirs. He asks for Alys’s backing. After considering the matter, Alys promises Bryn that she will back him up: she will also commit her union, in the meeting, to refraining from balloting on any deal that fails to meet those certain stringent demands.
Let φ represent that action of backing Bryn up in the negotiation meeting. The fact that Alys has promised Bryn that she will φ makes it the case that Alys owes it to Bryn to φ. The fact that Alys owes it to Bryn to φ is a relational reason. For the relational ethicists, the relational reason entails and is entailed by the fact that Bryn has a claim against Alys that she φ. For Wallace and Gilbert, that claim is grounded in Bryn’s status as the one to whom the promise was made; for Darwall, since generally we are all morally required to keep our promises, Bryn also has a valid claim against Alys that she φ, not just qua the individual claimholder, but qua representative of the moral community (see Darwall, Reference Darwall2013, p. 24).
With this relational reason in view, a question can be raised about semantic reducibility. Namely, can the relational reason be expressed as a fact about φ having value (i.e., being to-be-done)? In response, it seems that it must be possible for the relational reason to be expressed in terms of the value of φ-ing, on pain of it being a reason at all. If there was some part of the semantic content of the relational reason which is not amenable to being incorporated into an expression about the value of φ-ing, then to that extent, the relational reason would not be a matter of what is to be done, and so would not be a practical reason. The explanandum which relational ethics seeks to explain is the practical normativity, or normative force, of certain considerations. So, the answer that these theorists provide in terms of relational reasons must be a matter of practical normativity, and so must be conveyed by expressions about the value of the relevant actions. That is to say, by their own lights, relational ethicists should allow that relational reasons, as reasons, are semantically reducible to—which is to say expressible as—facts about the value of the relevant actions.
Happily for proponents of the reducibility thesis, I have already shown that the version of the thesis required to meet their explanatory goal is not the semantic version of the thesis. Thus, they ought to be prepared to allow that relational reasons might be semantically reducible. What matters for their thesis is that while relational reasons may be expressed in terms of the value of the relevant action, it is not the case that such reasons are really just a matter of such value. According to relational ethics, what relational reasons are, at the most fundamental level of description (beyond which no further explanatory light can be cast), is the relation between the agent for whom they are reasons, and some other agent.
The problem for the irreducibility thesis that I am presenting is that the simple picture of practical reason outlined above suggests that relational reasons are really just a matter of the value of the action for which they are reasons. There are two steps to illustrating this problem. The first step is to show that revealing a relational reason to be valuable constitutes an explanation, one that may go further than the explanation provided by the thought of A owing it to B to φ. The explanatory importance of the concept of value is in the deliberative significance of relational reasons, where they are understood to be commensurable for agents with other practical reasons. The second step is to confirm that the explanatory role played by the understanding of relational reasons as facts about the value of actions is a role that in principle could not be grasped without relational reasons being reduced to facts about the value of actions. Perhaps ironically, the argument for this second step resembles an argument offered by relational ethicists in favor of the irreducibility thesis.
Only facts about the value of actions are commensurable with other facts about the value of actions, of the sort that must be compared in practical deliberation. This is the explanatory significance of the reduction of relational reasons to such facts about value: the reduction reveals the commensurability of relational reasons with other practical reasons. Consider again the case of Alys and Bryn. Suppose now that, in the course of the negotiation meeting, the employers have made an offer in an attempt to sow division between the two unions. Their offer does not meet the stringent demands for which Alys and Bryn had agreed to hold out, but it seeks to placate the security workers by offering them advantageous conditions, without offering the same for cleaners.
Now, whatever one might think about the merits of such an offer, as so described, it is presumably something that Alys should consider. Considering the offer is an activity of deliberation that should include comparing her previous plan to hold out with at least the alternative course of action to propose that the security union hold a ballot on the offer. The theory of practical reason outlined above holds that the comparison between these courses of action just is the comparison between their value, which is to say the extent to which they are to-be-done. Thus, the revelation that a relational reason to be a fact about the value of an action is a revelation that explains how the relational reason can be compared with other reasons.
There are two routes via which proponents of the irreducibility thesis may reply to the argument at this point. One route would be to reject the theory of practical reason that I have outlined. I will discuss this line of reply in the next and final section below. The other route would be to accept the theory of practical reason but maintain that the commensurability of relational reasons with other practical reasons is explained already without their being revealed as facts about the value of actions, and so to deny that relational reasons are really just a matter of such facts. This second route of defense faces a challenge that is structurally similar to the argument offered in favor of the irreducibility thesis.
Defenses of the irreducibility thesis hold that relational reason is a sui generis domain: that the concept of a relational reason and its cognates cannot be explained except in their own terms. Irrespective of the efficacy of my argument for the idea that relational reason is not in fact sui generis in that way, an analogous point can be made about the conceptual domain of practical reason. Practical reason is sui generis, and irreducible. No propositions that are not already about what is valuable, or to be done, can entail any propositions that are. This is an echo of Hume’s (Reference Hume, Norton and Norton2007, bk. III, §1) famous observation about the descriptive–normative distinction that no propositions that are not already about what ought to be the case can entail any that are.
Could it be the case that relational reasons are semantically reducible to facts about the value of actions, and that relational reasons figure in practical deliberation, which is comparing the value of actions, but that relational reasons are not really just a matter of the value of actions? Perhaps one might think so, on the grounds that the deliberative role of relational reasons as comparable with other reasons can be recognized without revealing relational reasons to be facts about values. One might think, that is, that relational reasons just are the kinds of propositions that can be compared with other reasons, that is, with facts about the value of actions.
However, given the sui generis nature of the domain of practical normativity, this latter line of defense fails. The only propositions that can be compared with facts about value are themselves about value. Relational reasons entail facts about value only if they are themselves already facts about value. There is no way into the domain of practical normativity from the outside. My point here is an elaboration of the insight from Thomson that is my epigraph in this article. Ought simpliciter is a “one-hat” concept. If relational reasons cannot be entertained in practical deliberation, then they cannot explain the practical-normative force of the deliberative phenomena they are supposed to explain. But if they can be entertained in practical deliberation, then they are facts about the value of actions. Revealing as much explains how relational reason can figure in practical deliberation. Since the revelation has that explanatory function, it is more explanatorily fundamental than an alternative conception of the same phenomena which lacks that explanatory function. So, relational reasons are really just a matter of the value of the actions for which they are reasons.
4. The Priority of the Right Over the Good
The theory of practical reason on which my argument has relied is not the only game in town. A rival view holds that practical reason is, at least sometimes, an exercise not of comparing the value of actions, but rather of being moved by obligations, where these are different in kind from any countervailing facts about such values. The famous slogan for this Kantian-spirited view is that the right is prior to the good.Footnote 9 In this section I wish to note a central concern for the idea that the right is prior to the good. Without resolving the debate, my hope is that these remarks will maintain the dialectical pressure, where the burden of evidence is on critics of the conception of practical reason as value comparison.
There are two bases on which a proponent of the irreducibility thesis may wish to deny that relational reasons should be compared with facts about the value of actions. The first of these bases to consider (because it can be more straightforwardly addressed) is that such facts about values are inputs into practical deliberation, whereas relational reasons are not; they are outputs (Wallace, Reference Wallace2019, p. 19). If the term “relational reason” is restricted to referring to something that is a conclusion of practical deliberation, then it is true that it should not be thought to be comparable—and so needn’t be commensurable—with facts about the value of actions.
However, on that construal, the fact that Alys owes it to Bryn to hold out is not a relational reason—at least not insofar as that fact is something that Alys wishes to consider in relation to the idea that it may be best for her members for her to propose a ballot. If facts of the form “A owes it to B to φ” are not classified as relational reasons by some theory of relational reasons, then that theory will not explain the distinctive normative force of such facts. But explaining the normative force of those facts was the chief motivation of at least one central strand of relational ethics. So to retreat from seeking to explain the normative force of relational reasons qua inputs to deliberation is to retreat from that core project. While that strategy does seem to avoid the bite of the argument presented in Section 3, it may thus come at too great a cost for proponents of the irreducibility thesis.Footnote 10
The other basis on which one may deny that relational reasons should be compared with facts about the value of actions is the idea that relational reasons are indeed inputs to practical deliberation, but that they interact with other inputs in some way other than through comparison. A chief candidate for an alternative interaction between deliberative inputs is the notion of silencing. Some deontologists hold that certain considerations—principally obligations—figure as inputs to deliberation which silence other inputs (see e.g. Bader, Reference Bader, Hirose and Olson2015). Whereas a fact about the value of some action may generally have the normative force of favoring that action, if the agent faces an obligation not to perform that action, then the obligation silences the fact about the value and thereby cancels the normative force that the fact would otherwise have had.
On this deontological view, the right is prior to the good. Obligations and facts about the value of actions can both be entertained from the deliberative standpoint, but the former take priority over the latter in settling the question of what to do. Crucially, in order for this view to avoid collapsing back into the conception of practical reason as value comparison, it must be the case that when an obligation silences some other reason, this silencing cannot be explained by the obligation being to be done to some greater extent than the mere fact about value.
This is worth considering through an example, so let us turn one last time to the predicament in which Alys finds herself: a negotiation meeting at what might feel like a crucial moment in their industrial dispute. Among the inputs to Alys’s deliberation at this moment, two are salient for our purposes. Alys owes it to Bryn to hold out. And yet, it may be that balloting on the offer that has been tabled would be good for her members. If we understand the former to be a relational reason and an obligation, then on the line of defense currently under consideration, we might think that this consideration silences the other.Footnote 11 Crucially, the obligation silences the fact about the value of a rival action not because it is more valuable, or more to be done, but because the two considerations are different in kind and obligations silence mere facts about value.
The central problem for this silencing theory of practical reasoning is the point that has emerged through this example: the theory must deny the possibility of comparing the value of obligations and non-obligations. This is a problem because Alys can compare the to-be-done-ness of holding out with that of proposing a ballot, including in that comparison the force of her relational reason of owing it to Bryn to hold out. Beyond merely being able to make such a comparison, one might think that it is incumbent on Alys to make it. There may be ways of filling out the example such that Alys has all-things-considered reason to renege on her promise, betray Bryn, and propose a ballot. The difference between such a case, and an alternative situation in which she has most reason to hold out, is not a difference in the kinds of reasons represented by the promise and joint commitment with Bryn, versus the wellbeing of her members. The kinds of reasons that these reasons are is held fixed across the variations. Rather, the difference is in the value of the competing actions available to Alys. It is only as a comparison between such values that her resulting intention could be intelligible as a conclusion of practical reasoning.
5. Conclusion
In this article, I have argued that the version of the irreducibility thesis that is the key unifying plank of the various project in relational ethics is as follows: it is not the case that relational reasons are really just a matter of the value of the actions for which they are reasons. I have offered an argument to the effect that all practical reasons are really just a matter of the value of the actions for which they are reasons.
That argument has hinged on the idea that practical reason is comprised of comparisons between the values of candidate actions. There is a storied debate over whether that is the correct theory of practical reason. I do not suppose to have settled the debate here. Rather, I have drawn out some of the theoretical commitments that the irreducibility thesis must take on, and given voice to a theoretical challenge to which it is answerable. In a word, that challenge is this: if relational reasons are not just a matter of the value of the actions for which they are reasons, how are they to interact with other considerations from the deliberative standpoint?
Joint commitments, individual moral rights, and community-wide moral standards can all ground claim rights that agents hold over one another’s conduct. Such claim rights are conceptually connected with corresponding duties and obligations. The fact that one owes it to another to φ makes it the case that φ-ing has a certain practical value, that it is to be done, to a certain extent. One can acknowledge all of this without endorsing the irreducibility thesis. Indeed, philosophical work by Darwall, Wallace, and Gilbert and the many that have followed their lead has taught us much about the normative dynamics and philosophical underpinnings of reasons that emerge from the deep forms of relatedness that people inhabit with respect to one another.Footnote 12 It would be possible to dwell on these insights, and give them further philosophical attention, even while giving up on the thesis that has been the unifying idea of this field of enquiry.
Acknowledgments
I am very grateful to the editors for inviting me to contribute to this special issue, and to the following friends and colleagues for their comments on earlier versions of the article: Francesco Antilici, James Brown, Dan Khokhar, Anni Räty, Jack Samuel, and the Moral Address Work in Progress group.
James H. P. Lewis is a lecturer in philosophy at Cardiff University.
Open access
