Deontological Desert

Abstract

Although the nature of moral desert has sometimes been examined in axiological terms—focusing on the thought that it is a good thing if people get what they deserve—deontologists typically think desert is more appropriately treated in terms of duties and obligations. They may, for example, prefer to talk in terms of there being a moral duty to give people what they deserve. This essay distinguishes a number of forms such a duty might take, and examines four of them more closely. (In particular, it looks at positive and negative duties with regard to both comparative and noncomparative desert). Questions about the contents of each of these duties are raised, making clearer just how much work would be involved in spelling out the relevant duties more completely. The essay concludes with a brief discussion of the possible implications of such desert-based duties for population ethics.

1. Introduction

If we take seriously the thought that it is a good thing, morally speaking, if people get what they deserve, we are naturally led to the plausible thought that it is a less good thing if someone falls short of getting what they deserve. We are also led to the less obvious, but also plausible, thought that it is also less good (in terms of the value of one’s getting what one deserves) if someone has more than they deserve. Arguably, then, if one is too far from getting what one deserves, perhaps this actually reduces the goodness enough so that in this regard the result is actually a bad state of affairs rather than a good one. One might then wonder: might it be the case that any departure from having what one deserves is intrinsically bad in that regard—no matter how slight the difference—so that the goodness of someone getting what they deserve simply consists in the absence of the intrinsic bad that would occur if one fell short (or had too much)?

If you start pursuing thoughts like these you quickly find yourself developing what we might call an axiology of desert, an attempt to map out the ways in which intrinsic value rises or falls as people get, or fail to get, what they deserve. And armed with such an account, one might then fold it into one’s theory of the good—one’s overall theory of what it is that makes one outcome morally better or worse than another. Given that we have an obligation (perhaps one among others) to bring about better outcomes, we might in this way hope to develop a suitable “desert sensitive” theory of morality. That, at any rate, was what I tried to offer in my book, The Geometry of Desert [1], and as I argued there, developing an adequate account of the intrinsic value of people getting what they deserve turns out to be a surprisingly complex affair.

Conceivably, however, one might think that this sort of axiological approach to desert is misguided. At the very least, it will seem to many to overlook or fail to do justice to the main way in which considerations of desert ought to figure into our moral theory. Those who are deontologically inclined (so that right and wrong isn’t exclusively or even primarily a matter of bringing about better outcomes) may complain that the fundamental role of desert in our moral theory should be cashed out, rather, in terms of a suitable and more focused duty with regard to desert—perhaps a duty to give people what they deserve 1.

In what follows, I want to examine this alternative approach to desert. If we are thinking about desert from this more deontological perspective, aiming to arrive at a duty or set of duties that capture our obligations with regard to desert, what would the resulting principle or principles look like? What would they require or forbid? Here, too, the answer is surprisingly complicated. While it may seem obvious that we have a duty to give people what they deserve, I think it far from clear what the relevant obligation or obligations come to. A deontological theory of desert remains a complicated affair—enough so, so that all I will be able to do here is point to a few of the relevant issues.

Admittedly, in trying to lay out such a theory, I am, I suppose, handicapped, in that I am not myself a deontologist. Accordingly, when I make claims to the effect that a deontologist will want to say such and such, or is likely to believe this or that, it is certainly possible that those more deontologically inclined than I am, will think I have gone astray, that my intuitions are not sufficiently deontological. So be it. If those more naturally sympathetic to deontology than I am eventually offer different answers to the questions that I raise, I may still have done something useful in prompting them to provide those better answers. But be that all as it may, in what follows, I will do my best to offer some tentative answers on behalf of the deontologist.

2. Which Duties?

Before turning directly to thinking about the content of our deontological duties with regard to desert, let me get a few preliminaries out of the way. It is clear that we talk about desert in a variety of contexts, and just what it is that the person deserves varies with the context (this person deserves a good grade, that person deserves a raise, and so on), as does the basis of their being more or less deserving (it might be the quality of the paper they’ve turned in, for example, or the contribution they’ve made to the company’s profits, and so on). But for simplicity, I will focus exclusively on moral desert. I will assume without argument that it is one’s level of virtue or vice that makes one more or less morally deserving, and that it is well-being that one deserves more or less of, depending on one’s level of virtue or vice. What’s more—and more controversially—I will assume that for each person, given their particular overall level of virtue (and vice), there is a particular level of well-being that the person deserves. I will refer to this as the level of well-being that the given person “absolutely” deserves.

(Some may prefer to hold instead that what one deserves is not so much a particular level of well-being but rather a particular alteration in one’s level of well-being, perhaps some particular reward or punishment. Those drawn to this position should feel free to appropriately adjust what I say below; it shouldn’t affect the main points.)

This notion of desert is a noncomparative one: for any given person, what matters is whether or not they are getting what they absolutely deserve. I will assume that this is indeed one important part of our overall notion of (moral) desert as a whole. But I am also going to assume—again, without argument—that there is also a comparative notion of desert, one according to which what directly matters is how well each of us is doing compared to others, in light of how (noncomparatively) deserving we are. Suppose, for example, that you are as virtuous as I am, and so we each absolutely deserve the same level of well-being, but while this is indeed the level of well-being that you are actually at, I am at a somewhat higher level of well-being. Then, from the standpoint of comparative desert, you now deserve to have more well-being as well (even though you are already getting what you noncomparatively deserve): after all, you are just as virtuous as I am, yet you are worse off. While not everyone accepts this notion of comparative desert, most do, and I will take it too to be a part of our notion of what one deserves.

Thus, there are two “partial” desert values—comparative desert and noncomparative desert—and what one deserves on the whole is a function of both of them. As we have just seen, these two partial values can conflict (since, in the example just described considerations of noncomparative desert oppose giving you more than what you absolutely deserve, while considerations of comparative desert nonetheless support doing this), and a complete theory of desert will need to take both of these partial values into account. Thus, a complete theory of desert would include a suitable account of noncomparative desert, a suitable account of comparative desert, and a suitable account of how the two partial values trade off against one another.

Given all of this, however, the deontologist immediately faces a question when they posit a duty to give people what they deserve. Is there, simply, one single duty—a duty to give people what they deserve on the whole? Or is there, rather, a pair of duties—one, a duty to give people what they noncomparatively deserve, and the other, a duty to give people what they comparatively deserve? (If the latter, in cases of conflict, does one of these two duties always outweigh the other? Or does it depend on the details of the case?) Alternatively, might there actually be three duties at play here—a duty to give people what they noncomparatively deserve, a duty to give them what they comparatively deserve, and a duty to give them what they deserve on the whole?

As it happens, the proliferation of potential duties goes considerably further than this. Deontologists typically distinguish between positive and negative duties, where the latter only require agents to refrain from performing various sorts of prohibited acts, while the former require substantive positive action instead. On the face of it, this distinction should be relevant when thinking about desert as well. Thus, on the one hand, the deontologist might want to posit a positive duty to give people what they deserve. On the other hand, they might also want to posit some sort of negative duty as well—perhaps a prohibition against causing people to have less of what they deserve, or some such. (We will worry about details later.) From the logical point of view, since we could conceivably have both a positive and a negative duty with regard to each of noncomparative desert, comparative desert, and desert as a whole, it looks like we may now be up to as many as six possible deontological duties. Even if any given deontologist does not believe in all six of these, they are still going to have to pick and choose from among them.

Indeed, the matter may be more complicated still. When thinking like a deontologist it is natural to suggest that if there is a duty (or set of duties) with regard to desert, this will be a directed duty, one owed to individuals. I owe it to you to give you what you deserve (just as I owe it to Frances to give her what she deserves, and to Larry to give him what he deserves, and so on). And it is often natural to express this thought by talking instead in the language of rights or claims, instead of duties. Perhaps Larry has a right or a claim that I give him what he deserves (if we accept the relevant positive duty), or at least a right or claim that I not bring it about that he is less well off in terms of having what he deserves (if we accept the corresponding negative duty). Although it is controversial how to spell out these ideas precisely, intuitively the thought is that our duties with regard to desert are individualistic in this way. A duty to give Larry what he deserves is a duty owed to Larry, and my doing this is something that he has a claim to.

But recall the axiology of desert that I gestured at in the opening paragraphs: If Larry gets what he deserves, this seems to contribute to the intrinsic goodness of the resulting state of affairs. In contrast, if I give a bit less than she deserves to Frances, this contributes less to the overall value of the outcome. And if I “shortchange” Jeff as well, and do so by a sufficient amount, this may actually be an intrinsically bad state of affairs. Intuitively, all of these individual situations contribute to the overall goodness of the resulting outcome. All of which is just to say, we can assess the values of outcomes in terms of the extent to which people are getting (or failing to get) what they deserve.

To be sure, deontologists typically think that such questions concerning the goodness of outcomes are less important—perhaps significantly less important—than other elements in our moral theory. They typically say that it would be wrong to violate someone’s rights, for example, even if the results would be better if we did. But that is not to say that considerations of the overall goodness of outcomes are irrelevant from the deontological perspective. Other things being equal, most deontologists hold, we have an obligation to make the world a better place. So, if people getting what they deserve is a factor in our theory of good and bad outcomes, then this too should figure in a complete deontological treatment of desert. Of course, here too, the deontologist may well want to distinguish between a positive and a negative duty. There may, for example, be only a weak positive duty to improve the state of affairs with regard to desert. But even if so, there may well be a significantly stronger negative duty, a duty to refrain from making the overall state of affairs worse with regard to desert.

So now we actually have yet another six potential deontological duties! There could, in principle, be a positive duty or a negative duty with regard to the goodness of outcomes evaluated in terms of noncomparative desert, comparative desert, or desert taken as a whole.

This brings us to some twelve possible deontological duties with regard to desert. Duties can be positive or negative, they can be individualistic and directed or they can be concerned, rather, with the goodness of states of affairs, and they can be concerned with either of the two partial values, or with desert taken as a whole. This gives us two times two times three (equals twelve) distinct candidate duties from which to choose.

Presumably, any given deontologist will want to embrace some, but likely not all, of these twelve potential duties. It also seems likely that deontologists will differ among themselves as to which of the twelve are the most plausible ones to adopt. A thorough discussion of the topic, therefore, would examine each in turn, weighing its intuitive plausibility, and outlining its precise contents (or noting controversies about its contents), as well as exploring the implications of accepting various combinations of these duties, so as to better see what any given set, as opposed to another set, would require or forbid. Obviously, however, it won’t be possible to undertake anything remotely like such a systematic survey here.

Instead, I will limit myself to taking an initial look at only a few especially plausible candidate duties. All four duties will be “individualistic” (or “directed”) in the sense already identified (rather than being concerned with the value of states of affairs); they will be duties owed to the relevant individuals. In particular, then, I want to consider a positive duty to give people what they noncomparatively deserve, a positive duty to give people what they comparatively deserve, a negative duty to refrain from making people worse off with regard to getting what they noncomparatively deserve, and a negative duty to refrain from making people worse off with regard to getting what they comparatively deserve. We will turn to the first of these in a moment.

First, though, I want to mention another issue that I won’t have time to pursue, but which would certainly be relevant for any complete discussion: redundancy. Given the logical and conceptual connections between the various potential duties, it seems likely that once one has embraced certain duties, it will turn out that adding certain further duties will be “redundant” in the sense that adding them to one’s moral theory would not directly require or forbid any additional acts beyond those already required or forbidden (by the other duties already embraced). For example, it seems likely that if we already have a negative duty not to make people worse off with regard to what they noncomparatively deserve and a negative duty not to make people worse off with regard to what they comparatively deserve, then nothing further will emerge (in terms of forbidding further acts involving desert) if we also accept a negative duty not to make people worse off with regard to desert as a whole. For how could we make someone worse off with regard to desert as a whole without making them worse off with regard to one or the other of the two partial values? (In contrast, if we start with a negative duty not to leave someone worse off with regard to desert as a whole, it may not be redundant to add, say, the more particular negative duty not to make people worse off noncomparatively. Conceivably, after all, even if you have not made someone worse off with regard to desert on the whole, you may still have made them slightly worse off noncomparatively—provided that you made them significantly better off comparatively).

Of course, even if adding an additional duty is redundant in the sense just described—so that adding it makes no further direct changes to which acts are forbidden or required—there might still be theoretical grounds for embracing sets of duties that include redundancy. It could, for example, be that the existence of multiple duties that coincide in this way provides more reason to perform (or refrain from performing) a given act than we would otherwise have if we accepted only one of them, and that might be relevant in turn with regard to thinking about what we had most reason to do, all things considered. On the other hand, if redundancy can alter the balance of reasons, shouldn’t that potentially make a difference sometimes with regard to what we are required to do concerning desert? As I say, issues of redundancy are not without interest. But I won’t consider them further here except in passing.

One last preliminary. Here, as elsewhere in ethics, special questions arise when we are evaluating acts that involve the creation of “additional” people, people who would not otherwise exist. So, it might be best if we limit ourselves, at least initially, to “same people” cases, where the very same individuals exist regardless of what we do. We can come back to questions about desert and population ethics at the very end.

3. A Positive Duty to Give People What They Noncomparatively Deserve

If, as we are supposing, there is a deontological duty to give people what they deserve, then it is natural to interpret this as involving (at least in part) a positive duty: I must make efforts to bring it about that you have what you deserve. Additionally, you must do the same for me. And if, as I imagine, when we think about desert our thoughts most immediately turn to what any given individual noncomparatively deserves—what they deserve by virtue of their individual level of virtue or vice—then it is natural to think that our positive duties will involve (at least in part) a duty to give people what they noncomparatively deserve. That will be the first duty we examine.

Recall my earlier assumption that for each person, there is a particular level of well-being that this person deserves (by virtue of their level of virtue). Just what that level is may well vary, of course, from person to person (since our levels of virtue vary), but still: for each person there is a particular level of well-being that they absolutely deserve. So according to the idea we are currently entertaining, each of us has a positive duty to give each person what they absolutely deserve. If you deserve 100 units of well-being, and you have only 90, I have a duty to try to bring it about that you get the extra 10.

Of course, the usual qualifications apply. Deontologists typically think that positive duties are limited in various ways; they are “imperfect” duties. Most importantly, of course, is the thought that positive duties will normally be outweighed by any conflicting negative duties. But even when no negative duty is in play, there is a limit to what I am required to do. Special circumstances aside, for example, deontologists typically think that there is a limit to the sacrifices I am required to make, to meet my positive duties. For example, I may have a positive duty to promote equality or fight injustice—but only up to a point, only if the cost to me of doing this is “within reason.” Similarly, then, if it would cost me a lot to provide you with the extra 10 units of well-being that you deserve, I may not be required to do that. But as long as the cost of giving you what you absolutely deserve isn’t excessive, this is something I have a duty to do. In addition, since presumably there are still other values besides desert that I am required to promote as well, it is conceivable that in at least some cases my duty to give someone what they absolutely deserve will be outweighed by one or more of those other values. So, the duty will be only a pro tanto one: I have a reason to give you what you absolutely deserve, but this reason may not be decisive.

We should keep these various qualifications in mind, but for simplicity, we can put them aside. So, in the case just described, where you deserve 10 more units of well-being than you already have, I have a duty (other things being equal) to give you those 10 additional units.

But what if I can’t? What if, say, I only have the ability to give you an additional 8? Surely, I must do that, even though it won’t bring you to the level of well-being that you absolutely deserve. It would be absurd to think that the duty we are considering holds that unless I can give you all of (the rest of) what you absolutely deserve, I have no obligation to give you anything at all. So, the duty must be to bring you as close to having what you absolutely deserve as I can manage. Indeed, even if I can only increase your well-being by 1, you still have a claim to my doing that.

What’s more, it seems plausible to think that your claim to that 1 unit is stronger the farther you are below having what you absolutely deserve. Intuitively, if you deserve to be at 100, and you are only at 70, my duty to improve your well-being by 1 (if that is all I can do) is significantly stronger than if you are already at 95. There is a duty to give you what you deserve in both cases, but the duty is stronger in the former case, since you are that much farther from having what you deserve to have.

This means, presumably, that if there are two people who fall short of having what they deserve, and I cannot help both, then other things being equal I have a stronger duty to help the person who is farthest below having what they absolutely deserve.

But this points, in turn, to a further complication. Since people can vary in terms of how deserving they are (some being more virtuous than others), it also seems plausible to suggest that how strong one’s claim is may also depend on how much they absolutely deserve to have. If a highly virtuous person deserves 10 units more well-being, it may be more important for me to give them 1 additional unit than it is for me to give that 1 unit to a significantly less virtuous person, even if the latter person is even further from having what they absolutely deserve. Intuitively, the strength of someone’s claim will be a function of both how much more they deserve and how deserving they are.

Imagine, next, that you deserve 10 more units of well-being, and I have to choose between two possibilities: I can either give you 5 units more (which will still leave you 5 short of what you absolutely deserve) or I can give you 11 more, which means you will actually end up having 1 unit more than what you absolutely deserve. Presumably, of course, if I could have given you 10 or 11, my duty would be to give you the 10, since, as we agreed above, my duty is to bring you as close to having what you absolutely deserve as I can. (We are not, after all, discussing a general duty to aid; we are discussing, rather, a duty to give people what they deserve.)

Unfortunately, however, in the case we are now considering, I cannot give you exactly what you deserve: I can either leave you still “shortchanged” (by 5) or “overcompensate” you (leaving you with 1 more than you absolutely deserve). Which am I obligated to do? If we really are right to think of the duty in question as a duty to bring you as close to what you absolutely deserve as possible, then I owe it to you to give you the 11 rather than the 5, since that leaves you closer to getting what you deserve. Of course, if my choice instead was between giving you 5 or giving you 25, I would ordinarily be required to give you only the 5, since that leaves you closer to having what you deserve than you would be if I gave you 25. (The duty is to move a person closer to what they deserve, not further away!)

What if I had to choose between giving you 5 (leaving you shortchanged by 5) or giving you 15 (leaving you overcompensated by 5)? Either way, you will be closer to having what you absolutely deserve than you are now. So doing either is better than doing nothing. But either way you will still be 5 “off” from having what you deserve. So, do I fulfill my duty equally well, regardless of what I choose? We might think that, but perhaps we shouldn’t. Perhaps it is better to err on one side rather than the other!

Arguably, when dealing with the virtuous (at least), it is better (other things being equal) to err on the side of giving them more than they deserve rather than less. Indeed, also plausibly, but even more complicatedly, perhaps the extent to which it is better to err on the side of giving too much depends on just how virtuous the person is. Perhaps for the more and more virtuous, it is better and better to err on the side of overcompensating.

But if we do say this, should we also say, in contrast, that when dealing with the vicious, it is actually better to err on the side of their falling short? Or is it, instead, always better to err on the side of overcompensating (other things equal), even with regard to the vicious? However, we answer this question, it does seem as though the following is true: the more virtuous the person in question, the stronger their claim to be overcompensated rather than shortchanged; the more vicious they are, the weaker that claim.

So far, all of this is reasonably straightforward, even if working out the details would quickly become complicated. (All of the issues raised in the last several paragraphs have precise counterparts in the axiology of desert as well, so some of the relevant details—which I cannot pursue here—can be found in The Geometry of Desert. But this in turn raises an interesting question about redundancy: if someone has met their various directed positive duties with regard to noncomparative desert as best they can (meeting stronger claims before weaker ones), would it now be redundant to add a positive duty to improve the overall goodness of outcomes with regard to noncomparative desert? Alternatively, if we already have such a positive duty with regard to the value of outcomes (from the perspective of noncomparative desert), is it redundant to add directed positive duties?)

But we now must face a more difficult—and controversial—question. Notice that all the examples we have considered up to this point involve people who initially have less than what they absolutely deserve. But what should we say about cases where someone already has more than that? Imagine, for example, that someone absolutely deserves to be at 100 but is actually at 110. What does the duty to give people what they noncomparatively deserve ask of me now? Suppose that I can reduce this person’s level of well-being by 10, bringing them down to precisely the level they absolutely deserve to be at. Do I have an obligation to do that?

One might think that anyone who agrees that there is such a duty must be committed to retributivism. But I think that is wrong, at least not as that view is ordinarily understood. We are not here imagining that the person under discussion somehow deserves to suffer or deserves some sort of punishment. On the contrary, they deserve well-being, perhaps a great deal of it (100 units!). It is just that they have somehow ended up with even more well-being than they absolutely deserve. And, our question is, do I have a duty to lower them if I can?

It seems as though if there really is a duty to give people what they deserve, then I do in fact have a duty in this case to lower this person by 10 (Or less than 10, if I cannot lower them the full 10). Indeed, a bit surprisingly, since we are thinking of the duty in question as a directed duty, it seems as though the right thing to say, in fact, is that I owe it to the person to lower their well-being! What’s more, if we really are prepared to think of the duty to give people what they deserve as corresponding to—or perhaps even being grounded in—a right that each of us has to be given what we deserve, then it seems to follow that I will violate this person’s rights (other things being equal) if I fail to lower their well-being! To say the least, that may strike us as an odd thing to say.

Of course, once we put it this way, in terms of rights (or claims) on the part of the person who has more than they deserve, the following idea may suggest itself. Perhaps this person can simply waive their right in the situation in question. That is, although “initially” (as it were) the person does have a claim against me that I lower their well-being, if—as we can easily imagine—they would rather that I not do that, then all they have to do is to waive the claim in question, and now the duty simply vanishes.

Such a proposal helps. But notice that even if the right can be waived, this presumably would not be automatic. So, if, in this particular instance, the right holder fails to waive their right, then it seems that we are indeed left claiming that we owe it to the person to reduce their well-being after all. That still seems odd.

In any event, should the deontologist hold that the right in question can be waived? It is certainly important to recognize that there do seem to be cases where someone can indeed waive their claim to be given what they deserve. Intuitively, after all, if you have less than what you deserve, then even if your claim to be given extra well-being is stronger than that had by someone else, it does seem as though you can waive that claim, making it permissible for me to help that other person instead. But if so, would not it be implausibly asymmetrical to hold that one can waive one’s claim to be given what one deserves only if one has less than what one deserves, not if one has more?

On the other hand, anyone who is truly inclined to accept retributivism—anyone who believes that there are indeed situations where a given individual deserves to be punished in some way—had better hesitate before saying that one can in fact always waive one’s right, and thus eliminate the corresponding duty. For, then, those who deserve to be punished will be able to eliminate any such duty we might have to punish them, simply by waiving away the relevant claim. And that seems implausible as well. (Of course, even if someone can waive their right to have their well-being lowered, and does so, this might still be a case where it is a less good state of affairs if this person really does continue to have more than they deserve. So, this might be a case where adding a positive duty to promote the overall goodness of outcomes with regard to noncomparative desert would not be redundant after all).

Perhaps the deontologist should deal with this issue by holding that there is indeed a duty to lower a person’s well-being if they have more than they deserve, one that the person cannot release me from, but that this positive duty is simply outweighed by a different duty, a negative duty that prohibits harming innocent people. Since lowering the person from 110 to 100 would clearly be harming them, the deontological prohibition against harming the innocent would forbid any such lowering. At the same time, notice that a proposal along these lines would still permit punishing wrongdoers—since such people are not innocent, and so a prohibition against harming the innocent would not come into play.

That seems promising, but even this suggestion is not without its difficulties. On the one hand, if the right to be given what one deserves cannot be waived (as this proposal assumes), how are we to accommodate the intuition that it actually can be waived, at least in those cases where one has less than what one deserves? And on the other hand, should we really be confident that the negative duty concerning harm rules out all cases of harming the innocent? Perhaps it only rules out causing undeserved harm. If so, then since the person in our example deserves to be at a lower level, if we do lower them, we won’t be violating that negative duty after all!

4. A Positive Duty to Give People What They Comparatively Deserve

Let us turn now from the positive duty with regard to noncomparative desert to a positive duty with regard to comparative desert. Recall that when it comes to comparative desert, the question is how well each person is doing in light of what that person absolutely deserves, as compared to how well others are doing (in light of what they deserve).

For example, take a case where both Ari and Becca deserve 100, but in fact each has 110. From the standpoint of noncomparative desert, of course, the situation is objectionable (as each has more than they absolutely deserve). But from the standpoint of comparative desert, in contrast, the situation is perfectly satisfactory. Admittedly, both have more than what they absolutely deserve, but since each has the same amount more, neither has a “relative advantage” compared to the other—neither is doing better than the other, relative to what they noncomparatively deserve. On the other hand, if Ari and Becca both deserve 100, but Ari has only 105 while Becca has 110, that situation is objectionable from the standpoint of comparative desert. Here, Becca has a relative advantage compared to Ari, since Becca has disproportionately more of what she deserves than Ari has of what he deserves.

(Although it won’t be important for the discussion that follows, it may be worth noting in passing that the person with the relative advantage from the standpoint of comparative desert need not be the person with more well-being. For example, if Chris deserves 100 and has 110, but Dahlia deserves 120 and has 122, then even though Dahlia has a higher level of well-being than Chris has, it is actually Chris who has the relative advantage in this example, since Chris is doing better, relative to what he deserves, than Dahlia is, relative to what she deserves.)

It is controversial what the precise metric is for calculating the size of someone’s relative advantage (or disadvantage). Unfortunately, we do not have the space to enter into that debate here (See [1] (Part III)). Happily, however, for present purposes, it will suffice if we agree that—in the case just described—if Becca’s level of well-being cannot be altered (as we can readily imagine), then Ari comparatively deserves to have 5 more units of well-being, so as to eliminate Becca’s relative advantage. And what we are supposing now, as well, is that the deontologist accepts a positive duty to give people what they comparatively deserve.

Of course, if we do give Ari the 5 units of well-being that he comparatively deserves, this will leave him worse off in terms of what he noncomparatively deserves, and conceivably there may be a negative duty that prohibits doing that. But for the moment we are simply focusing on what the positive duty with regard to comparative desert asks of us. And it does seem plausible to think that there is such a duty, even if it can sometimes be outweighed. Suppose, after all, that Ari and Becca both absolutely deserve 100, which Becca has, while Ari has only 90. Then, it does seem, intuitively, as though there are two duties in play here, both supporting giving Ari the extra 10. There is, to be sure, a duty to give the 10 to Ari because he noncomparatively deserves it. But there is also, in addition, a duty to give the 10 to Ari because he comparatively deserves it as well. (There is more reason to give 10 to Ari in this case than there would be if Becca too, like Ari, had only 90. For in that situation, in contrast, the comparative duty would not be engaged.) So let us assume that there is, indeed, a positive duty to give people what they comparatively deserve.

Presumably, the usual qualifications for positive duties apply with regard to this duty as well. For example, ordinarily, one only has to give people what they comparatively deserve if the cost to you of doing this isn’t unreasonably high. And when this duty conflicts with negative duties it may well be outweighed. And so on.

Furthermore, as with the duty to give people what they noncomparatively deserve, if one cannot give someone everything that they comparatively deserve, one has a duty to give as much of what they comparatively deserve as one can. Presumably, the duty is actually to bring people closer to having what they comparatively deserve. Accordingly, in some cases it may even be obligatory to give someone slightly more than they comparatively deserve, if the only alternative is to leave them worse off, comparatively speaking. On the other hand, if I do give Ari more than he comparatively deserves, this now leaves Becca at a relative disadvantage, and that may well be forbidden by a negative duty against creating or worsening relative disadvantages. (Would such a negative duty always outweigh the positive one?) Finally, it also seems plausible to hold that this duty (like the noncomparative one) is correspondingly stronger, the further the person is from having what they (comparatively) deserve.

I assume, of course, that the deontologist wants to think of this second duty as a directed one as well. That is, I owe it to Ari to give him what he comparatively deserves. But now a new question emerges: do I owe it only to Ari to give Ari what he comparatively deserves? Or do I owe that to Becca as well?

Intuitively, there is something unfair about the fact that Becca has a relative advantage compared to Ari. That is why Ari has a claim to be given more; arguably, he has a right to have this relative advantage eliminated or reduced. But does not Becca have a claim to the very same thing? Does not she too have a right to have her relative advantage eliminated (or reduced)? To be sure, in some straightforward sense, Becca is the beneficiary of the unfair situation—but for precisely that reason, does not she too have a right to have that unjust and unearned advantage removed?

So, do I, in fact, owe it to both the disadvantaged and the advantaged to correct the situation? Or is it, rather, only the person with the relative disadvantage who has the relevant right? I won’t try to settle this question on behalf of the deontologist. (But compare the analogous question about inequality: Is it only those with less than others that have a claim to have unjustified inequality removed? Or do those with more than others have a claim to that as well?)

So far, we have been considering an example where Becca’s level of well-being can’t be altered, so our only choice is to leave Ari at a relative disadvantage or to give him 10 units of well-being (thus eliminating Becca’s advantage). But what if the reverse were true? What if we can’t adjust Ari’s level of well-being at all, but we could lower Becca’s, by 10. Are we required to do this?

Obviously, if we do lower Becca from the 110 units she has to the 100 she absolutely deserves, this is better noncomparatively. So, I may already have a duty to do that (although, as we have seen, there are some questions about that as well). But put that aside. What, if anything, does the duty to give each person what they comparatively deserve ask of us now? It is certainly true, after all, that lowering Becca’s well-being by 10 units eliminates her relative advantage over Ari. So does our comparative positive duty direct us to do this?

Consider the question first from Ari’s perspective. If we reduce Becca’s well-being, does this count as giving Ari what he comparatively deserves? One might think not, since lowering Becca’s level of well-being does not actually increase Ari’s. But what of it? Arguably, all that he actually comparatively deserves is to have his relative disadvantage eliminated, and while increasing his well-being by 10 does do that, so does decreasing Becca’s by 10. So, it looks as though we do have a comparative duty to lower Becca’s well-being in the current version of the example. And if this is indeed a directed duty, as we are assuming, do not we owe this to Ari? Indeed, does not Ari have a right that we reduce Becca’s well-being? It does seem as though that might well be what the deontologist should say.

If so, should we also say the same thing about Becca? Are we prepared to say that even the relatively advantaged have a right to have that advantage removed or reduced? Do we now find ourselves needing to say that we have a duty to Becca to lower her well-being? Once again, that may seem at least a somewhat odd thing to say.

When the corresponding question arose with regard to the duty to give people what they noncomparatively deserve, we wondered whether each of us has the ability to waive our right to what we deserve—in particular in cases where being brought closer to what one absolutely deserves involves a loss of well-being. The same question, of course, arises here as well. Arguably, then, even if Becca does have a right to have her well-being lowered, perhaps she can waive that right, eliminating any duty we might otherwise have to her to lower her well-being. But of course, even if she does waive that right (presumably something she need not automatically do), that still leaves the duty we have to Ari to do the same, and there can be no assumption that he too will waive his right in this case. If not, then there is a duty to lower Becca’s well-being after all.

(On the other hand, consider again for a moment the case where we cannot alter the fact that Becca is at 110, but we can give Ari an extra 10, eliminating Becca’s relative advantage. Imagine that this time it is Ari who waives his comparative right, and Becca who refuses to do so. Does Becca have a right that Ari be given 10 units of well-being against his will?)

Here, too, of course, the deontologist might think it important to remember the prohibition against harming the innocent. If, as we can certainly imagine, Becca is innocent (it isn’t as though she has any of her 110 through immoral acts), won’t such a prohibition rule out our lowering her well-being by 10, since that undeniably harms her? But as noted previously, the issue isn’t clear, since the prohibition may only rule out undeserved harm; and since it does seem as though Becca comparatively deserves to have her relative advantage removed, it isn’t obvious whether the prohibition in question so much as comes into play.

5. A Negative Duty with Regard to Noncomparative Desert

The two duties we have considered so far have both been positive duties—requirements to give people what they comparatively or noncomparatively deserve. But intuitively, at least, there are negative duties as well, and deontologists typically think that these are even more important than the positive ones (such that, for example, a negative duty will ordinarily outweigh a positive duty). Since it is plausible to think that the deontologist will also embrace negative duties with regard to desert, let us turn to these next, starting with a negative duty with regard to noncomparative desert.

As a first pass, we might describe such a duty as a prohibition against giving someone something they do not (noncomparatively) deserve. But if we aren’t careful we might find ourselves interpreting this as prohibiting giving anyone anything less than exactly what they (absolutely) deserve. And since—as we have already noted—this is often unavoidable (you might absolutely deserve 10 units more, while I can only provide you with 8), we would then be saddled with the implausible implication that if I cannot give you exactly what you deserve, then I must avoid giving you anything at all.

Happily, a less potentially misleading formulation isn’t hard to find. Just as we have interpreted my positive duty as a duty to bring you closer to having what you absolutely deserve, we can understand the corresponding negative duty as a prohibition against moving you away from (or further away from) having what you absolutely deserve. Thus, if you absolutely deserve 10 more, and I can only provide you with 8 of those, it is permissible to do that. What I can’t do, rather, is reduce your level of well-being even further, leaving you with even less of what you absolutely deserve.

Of course, if I do reduce your well-being in this way, that will be harming you, and we might wonder whether the prohibition against harming the innocent already suffices to rule out this sort of lowering. But as I have previously observed, it might well be that what that prohibition actually rules out is only undeserved harm—and of course if that is indeed the right way to think of this deontological prohibition, then it isn’t really a distinct prohibition at all: it is simply (part of) the deontological prohibition we are currently examining, one which prohibits lowering someone’s well-being when they do not (absolutely) deserve to have less.

Furthermore, even retributivists, who believe that wrongdoers may deserve to suffer or be punished, believe that it can be wrong to punish someone more than what they absolutely deserve. Yet, a mere prohibition against harming the innocent won’t rule that out. What is needed, it seems, is the fuller duty now being entertained: a prohibition against moving someone farther away from having what they absolutely deserve.

If there is such a negative duty with regard to noncomparative desert, then presumably the deontologist believes it is like other negative duties in that it can ordinarily outweigh our various positive duties. It would be wrong, for example, for me to lower Elana’s well-being—if she currently has what she absolutely deserves (or less)—even if this were necessary to bring about a better outcome overall, even if it were necessary to increase the extent to which people on the whole were getting what they deserve (or getting what they noncomparatively deserve), and even if it were necessary to prevent a larger number of violations of this very same duty!

Of course, as with other negative duties, a “moderate” deontologist can consistently believe that despite all of this being true, the prohibition in question has a threshold: if enough good is at stake, perhaps it is permissible to impose certain undeserved harms after all. That raises questions about the precise nature of the threshold, and whether the deontologist should accept a threshold here at all. But I won’t be able to pursue these complications either.

In any event, as with our first two duties, I assume that the deontologist thinks of the current duty as a directed one: I owe it to you not to move you further away from having what you absolutely deserve. Arguably, you have a right not to have undeserved harms imposed on you. And so on.

Notice, however, that if the duty in question is, as it seems it must be, a duty not to move you further from having what you absolutely deserve, then I do not violate that duty if you absolutely deserve 100 and I lower your well-being by 5 units, from 110 to 105. In such a case, after all, I have actually brought you closer to the level you absolutely deserve, rather than further from it; so, the duty has not been infringed. And similarly, I imagine, even if I lower you 11 units, so that you end up at 99—a bit below what you absolutely deserve—I still have not violated the duty, since you still end up closer to having what you absolutely deserve (than you were before). What would violate the duty, of course, would be if I lowered your well-being so significantly that you ended up worse off, with regard to what you noncomparatively deserve, than you were at the outset.

(How, exactly, should your “distance” from what you absolutely deserve be measured? In terms of units of well-being? I doubt that is quite right, for reasons related to a question noted previously—whether it is better to err on the side of shortchanging or overcompensating. I imagine that an adequate metric would also need to take into account both your overall level of virtue and whether you have more or less than what you absolutely deserve.)

So far, we have been examining this third duty only in terms of cases where I lower your well-being. But it is worth noting that I can also violate the duty by raising your well-being instead! Suppose you absolutely deserve 100 and already have it (or more!). If I then increase your well-being even further, then I am actually moving you away from having what you deserve to have, in violation of the negative duty. Indeed, apparently I owe it to you not to increase your well-being in this way; you have a claim against me that I not do this.

One might be tempted to appeal here to a further duty, a general duty to aid people (if one can). Won’t that render my act permissible? But it isn’t clear that it would. On the one hand, it isn’t clear that there is an obligation to give people aid that they do not deserve. And on the other hand, even if there were such a duty, it would clearly be a positive duty; so, shouldn’t the deontologist insist that it is outweighed by the negative duty we are discussing here, one which prohibits moving people further from what they absolutely deserve?

Of course, if this negative duty really is a directed duty, something I owe to you, perhaps you can simply waive your claim that I not move you further from what you deserve. If so, then the duty we are discussing may not oppose my increasing your well-being after all. To be sure, one might worry that there still seems to be something morally objectionable about my giving you this undeserved increase in well-being. Arguably, however, this is just the fact that even though you have agreed to the increase, the resulting state of affairs is nonetheless less intrinsically valuable in this regard. (If so, would this show that we need to add yet another duty, a duty not to worsen outcomes with regard to desert? This is yet another question that a fuller discussion would need to examine.)

6. A Negative Duty with Regard to Comparative Desert

The remaining duty of the four I intend to discuss here is also a negative duty. But unlike the last, which concerned noncomparative desert, this one concerns comparative desert.

By this point, I imagine, the most important ideas we need to explicitly note will already be familiar. Intuitively, of course, what we are after is a prohibition against making things worse for people with regard to comparative desert. Imagine, for example, that Ari and Becca both deserve 100, and Ari is at 105, while Becca is at 110. In this situation Becca has a relative advantage compared to Ari: she is doing better, relative to what each absolutely deserves, than he is. (True, both have more than what each absolutely deserves; but Becca is even further beyond having what she absolutely deserves than Ari is.) So, if, in a situation like this, I were to increase Becca’s well-being, this would increase Becca’s relative advantage as well (and increase Ari’s disadvantage). Doing this is forbidden by our final duty.

To be sure, increasing Becca’s well-being moves her even further away from what she absolutely deserves. So doing this is already forbidden by the negative duty against making people worse off with regard to noncomparative desert. But the point right now is that this act is also forbidden by a negative duty that prohibits making people worse off with regard to comparative desert as well. There are, it seems, two prohibitions at work here, both opposed to raising Becca’s level of well-being. (And that is just as well, since if there were only the first of the two, then Becca might be tempted to get around it by waiving her right not to be moved further from what she absolutely deserves.)

Obviously enough, the negative comparative duty also prohibits me from lowering the level of well-being that Ari is at. For this too increases Becca’s comparative advantage. Admittedly, if I only reduce Ari’s well-being by a few units (say, from 105 to 102), this may bring him closer to getting what he noncomparatively deserves (namely, 100). So, this may well be required by the positive duty to bring people closer to having what they absolutely deserve. But for all that, lowering Ari’s well-being even by a small amount makes his situation worse in terms of comparative desert, so doing this is prohibited by the negative duty we have just introduced. And given the standard deontological view that negative duties outweigh positive ones, it seems that the prohibition against lowering Ari’s well-being will outweigh the positive duty that might otherwise require our moving him.

Presumably, this new duty, like the others we have looked at, is a directed duty. At a minimum, and most obviously, I owe it to Ari not to do something that will increase his relative disadvantage. But beyond that, I may owe it to Becca as well, that I refrain from increasing Ari’s disadvantage. After all, Ari’s having a relative disadvantage (compared to Becca) is the very same fact as Becca’s having a relative advantage (compared to Ari). So, it seems conceivable—though the idea is more controversial—that I may owe it to her that I not increase her unjust and unearned comparative advantage. (Admittedly, Becca might waive her claim not to have her relative advantage increased. But even if she did, Ari need not do so. So, the negative duty might still be in play.)

It may also be worth noting that if we imagine instead a case where both Ari and Becca start out with exactly what they absolutely deserve, then moving either one of them, in any direction at all, will be prohibited by the negative comparative duty. For if both have exactly what they absolutely deserve, neither has an advantage compared to the other, so altering either person’s level of well-being (while leaving the other where they are) will create a comparative advantage where none existed before. The negative comparative duty will rule that out. On the other hand, if both are given increases (or decreases) then the negative comparative duty may be okay with that: depending on the details, it could be that neither gains a relative advantage compared to the other.

(Of course, if—as seems inevitable—there are other people who exist as well, things are more complicated. If both Ari and Becca are given comparable increases, for example, but Elana is left behind, then even if neither Ari nor Becca gains a relative advantage compared to the other, Elana may now have a relative disadvantage compared to both, one that was not there before. If so, the negative comparative duty will rule out aiding Ari and Becca after all.)

Intuitively, I think, both of the negative duties we have discussed should seem attractive to the deontologist. But if we do accept both of them, there can be cases where we seem to be obligated to do nothing. Suppose, for example, that Ari and Becca each absolutely deserve 100, but neither has this: Ari is at 80 and Becca is at 90. Imagine as well that I cannot alter Ari’s well-being at all, but I can either raise or lower Becca’s level by 10. What should I do?

Obviously, raising Becca’s well-being brings her closer to having what she absolutely deserves, so doing this is required by the positive duty with regard to noncomparative desert. Unfortunately, however, it also increases the size of Ari’s comparative disadvantage, so it is forbidden by the negative comparative duty. Additionally, since deontologists think negative duties outweigh positive ones, it seems that I am forbidden to increase Becca’s well-being. So, should we lower Becca’s well-being instead? Doing this eliminates her relative advantage, so it is required by the positive duty with regard to comparative desert. But it moves Becca further away from having what she absolutely deserves, so it is forbidden by the negative noncomparative duty. And since deontologists think negative duties outweigh positive ones, I am forbidden to reduce her well-being as well. It thus turns out that I am permitted neither to increase her well-being nor to reduce it. I am required to do nothing at all.

How often will cases like this arise? That is unclear. But I worry that in the real world—where people vary wildly in terms of how close they are to getting what they absolutely deserve and we can’t always help the people who deserve it the most—we might well find ourselves in this situation almost all the time!

7. Creating People Who Would Not Otherwise Exist

Let me close with a very brief discussion of what happens if we relax the “same people” assumption I invoked in Section 2. Suppose we find ourselves in a situation where we have the option of bringing it about that someone will come into existence who would not otherwise exist. What do our various desert-based duties say about such cases?

Ideally, of course, we would create someone who will end up having exactly what they absolutely deserve, where doing this neither creates nor increases relative disadvantages for anyone. Then, neither negative duty is in play.

But are we required to create such a person if we can? Some may want to say so, interpreting the positive noncomparative duty as one which requires us to act in such a way that we cause people to get what they deserve, even if the person in question would not otherwise exist. Others, however, will insist that the obligation is only relevant with regard to people who already exist (or who will exist) independently of our action, in which case there is no duty to create the extra person at all. (But what if creating this person would actually reduce the comparative disadvantages of some people who already exist? Would that give us a reason to do it?)

Of course, as we all know, more often than not the people we may add to the world will not actually end up having exactly what they absolutely deserve, but will instead fall short of this (or, sometimes, end up with more). But we have embraced a negative duty with regard to noncomparative desert. Does creating such a person violate it?

Here are three possible positions:

(1)

The easy line says that the negative duty is only relevant with regard to people who exist independently of our actions. So, when it is a matter of adding extra people, the negative duty does not come into play at all. Thus, as far as noncomparative desert is concerned, we can create a new person no matter how far they end up falling short (or long) of having what they deserve. After all, this distance can’t be compared to what it would have been had they never existed. So, I can’t possibly make it worse.

(2)

The hard line says that the negative duty is indeed relevant even for people I cause to come into existence. So, unless the person will end up having exactly what they deserve, it is forbidden to create such a person. After all, had I not brought them into existence, they would not be in the situation where they have a complaint concerning the fact that they do not have what they deserve to have. But if I create them, and they fall short (or long) of having what they absolutely deserve, they do have a complaint, a complaint that would not exist but for my action. So, in such cases, it is forbidden to cause the existence of the person in question. If I do it, I have wronged them.

(3)

The moderate line agrees that the negative duty is relevant even for people I cause to exist, but it is more forgiving than the hard line. It asks us to return to a question raised in the opening paragraph: are all cases where someone fails to get what they deserve intrinsically bad? Or is this only true if the person is sufficiently far from having what they deserve? Admittedly, any divergence here at all is something that is less desirable than when the person does have exactly what they deserve, but perhaps relatively minor divergences are still intrinsically good nonetheless, rather than intrinsically bad. If so, then perhaps all that the negative duty rules out is creating a person when they will be so far from having what they absolutely deserve, that their being that far has created a situation that is, in that regard, intrinsically bad, rather than good. If we adopt the moderate line, then we are permitted to create people who will not have what they deserve after all; but only if the divergence is sufficiently small.

As might be expected, similar issues also arise with regard to the negative comparative duty. Suppose that I cause someone to exist who will have exactly what they absolutely deserve. Unfortunately, since there will almost inevitably be other people who do not have what they deserve, this results in my saddling these other people with new comparative disadvantages: they will now be relatively disadvantaged compared to the person I created. So, have I violated the negative comparative duty?

Once again, there are three main options:

(1)

The easy line will insist this is a nonissue. Since those who are now disadvantaged would have stood in no relation at all to the person had he not been created, it cannot be claimed that their relative relation has been made worse. So, no matter how great the new relative disadvantage, the negative comparative duty simply is not relevant.

(2)

In contrast, the hard line will insist that new comparative relations are indeed relevant, and so unless adding the new person creates no additional comparative disadvantages at all, it is forbidden. After all, if I had not brought the new person into existence, those who are now disadvantaged compared to the new person would not have this additional complaint. So, in creating this new person, I have wronged them.

(3)

Finally, the moderate line will agree that the new comparative relations are relevant, but will insist nonetheless that here too we need to ask whether all instances of comparative differences are sufficiently great to constitute situations that are intrinsically bad. Even if the ideal is to have no one relatively disadvantaged at all, perhaps if the disadvantages are sufficiently small—if people are sufficiently close to being comparable (in light of what each deserves)—this is still good overall. If so, then maybe it can be permissible to add someone new, even though this creates new relative disadvantages, provided that these new disadvantages are small enough.

No doubt, deontologists will differ among themselves as to which of these various alternative views are the most plausible ones. It is, after all, often difficult to know how best to extend normative principles designed for “same people” cases so as to deal plausibly with “different people” cases as well. That the same may also prove true for desert-based principles should hardly surprise us.

What may well surprise us, however, is the fact that finding suitable deontological principles for desert—and working out their details—is less than straightforward even when we limit ourselves to same person cases. It is easy enough for the deontologist to insist that desert should be incorporated into our moral theory in a deontological, rather than a merely axiological, fashion. But anyone who hopes to state the relevant principles for a deontological account of desert will have their work cut out for them. We have done nothing more here than note a few of the relevant issues.