Rights, Utility, and Today’s Most Surprising Sentence

For my sins I was recently re-reading Ronald Dworkin’s Taking Rights Seriously (1978), in particular its chapter 9,  “Reverse Discrimination.” The book has not aged well, and this chapter in particular today sounds even more like a dated period piece than the construction of a philosophical argument than it did when I first read it in 1994.

In any case, in a book that invites us to ‘take rights seriously,’ one may be surprised to find this sentence: “An individual’s right to be treated as an equal means that his potential loss must be treated as a matter of concern, but that loss may nevertheless be outweighed by the gain to the community as a whole” (p. 227).

Thus infringing an individual’s “rights” is indeed “a matter of concern,” but if the utilitarian calculus indicates that that infringement leads to greater overall utility, then we might not only be justly allowed to proceed with the infringement, but we may even be morally required to do so. The issue at stake in this chapter is the claim of some whites that the fact that they were being held to different and higher standards for admission to things like law and medical schools than were members of other races was an instance of unequal—and therefore possibly unconstitutional or otherwise illegal—treatment. Dworkin argues that no Constitutional right is violated by this “reverse discrimination,” and he goes on to argue that no moral right is violated either. The alleged moral right is the right all individuals have to be treated equally, for which Dworkin argues elsewhere in the book (see esp. chap. 6, “Justice and Rights”). In the case of reverse discrimination, however, Dworkin argues, as he states in the sentence above, that that right may be infringed when the public good requires it.

All this is quite surprising, given that Dworkin claims in the book’s “Introduction” that the whole point of his book is to argue that “Individual rights are political trumps held by individuals,” and that “a collective goal is not a sufficient justification for denying [individuals] what they wish, as individuals, to have or to do” and “not a sufficient justification for imposing some loss or injury upon them” (p. xi). Indeed, the book’s chapter “Justice and Rights,” argues, as Dworkin explains, “that our intuitions about justice presuppose not only that people have rights but that one right among these is fundamental and even axiomatic. This most fundamental of rights is a distinct conception of the right to equality, which I [Dworkin] call the right to equal concern and respect” (p. xii; italics supplied).  Dworkin specifically and repeatedly argues that his conception of individual rights is incompatible with “the theory of utilitarianism” (p. vii and passim).

It is difficult to reconcile Dworkin’s position on rights and utility, on the one hand, with his position on reverse discrimination, on the other. But perhaps that goes to show only that it is not ultimately possible to subscribe both to a robust theory of individual rights and to utlitarianism. The two will inevitably conflict, at which point one will have to choose which wins. One might wish to have both, but one’s moral and political position can serve, finally, only one master.

17 thoughts on “Rights, Utility, and Today’s Most Surprising Sentence

  1. Jim, I think you are right about that, but I think there might be a further point too, which is that you could take the most adamant inviolate-right stance and utterly castrate it with the wrong conception of what it is a right to. For example, even the most straightforward, brain-dead hedonist act utilitarian could say, “I have an absolutely inviolable rights-based view. Each individual is absolutely entitled to have his or her interests count as of exactly equal weight as those of anyone else, and such a right may never be overpowered by any other consideration.” Now we’ve got a strong rights view, a clear form of egalitarianism, and we still have a maximizing view with all the defects of such views.

  2. not so. There must always be tension between the two. I assert the right to murder anyone who gets in my way on the road. Why do you deny me this right?

    Mostly we don’t face questions like this because society has formed a consensus approach that we all sign on to; but there is always a gray zone where it comes up, either for society of individuals.

    1. Ozzie, where is the tension, exactly? Of course there can be tensions between claims that people have rights (or the rights that they stake claims to); that’s as sure as sausages. But it’s not clear to me what you think is not so.

      1. “one’s moral and political position can serve, finally, only one master”

        Not so. Ones Moral and political position must always balance between the two. Only one sometimes doesn’t realise that

      2. There surely have been people who have thought that there can be multiple masters. But I don’t know anybody who has modeled a way in which we can think we have have reason to serve two multiple masters. In the end, such models invariably require a breakdown in responses to the demands of reason, simply because if reason can’t arbitrate between the claims of the masters, no rational resolution is possible. I myself doubt that there’s any reason to accept the “two masters” view.

  3. Tricky issue. The complaint of “reverse discrimination” is obviously false… no school official wishes to avoid white applicants. (The point is “reverse discrimination” is a mens rea type of crime… wanting to promote a minority != wanting to demote the majority.)

    The real complaint is one of distributive justice. Any system of remedying past injustices must be borne equally by all members of the majority (arguendo). Instead, the first white applicants refused bear all of the cost. If perhaps every white applicant had an equal chance of being rejected in a lottery to make room for one minority, I might be on board.

    Unfortunately, analyzing distributive justice is one of the things racists and bigots are really bad at, so they can’t formulate the issue very well!

    1. Verbal, the task of remedying past injustices is not an issue of distributive justice. It is a task of compensatory justice, or restitution. Those forms of justice make different demands, and part of the problem with reverse discrimination is that it is not obvious that it meets those demands. If I wrong you, compensatory justice requires that I make good your losses. If the aim is to correct cases in which A has been harmed by B, imposing a cost on C to make A good is not obviously doing what compensatory justice demands. The tricky issues here surround attempts to make reverse discrimination fit the model of compensatory justice, for example by identifying A and C. Skeptics, I think, think that simply doesn’t work.

      1. No one is complaining about remedying past injustices. By definition, remedying past injustices has appeal. (In fact, few people will even argue in public about whether affirmative action targets are deserving of restitution.)

        Yes, I agree “compensatory” justice is involved, but affirmative action does not have a compensatory PROBLEM; if anything it overcompensates. It does have a problem that the cost is borne unequally as it attempts to restore justice, and that is a distributive problem.

        Anyway, whatever… we can slice our own personal flavors of justice until the cows come home… the important point is that “reverse discrimination” is obviously false, and people who talk/think that way lack clarity of moral thought and should not be relied upon.

      2. Verbal, I don’t understand you here. The law says that discrimination on the basis of race is illegal; presumably that is itself a reflection of the injustice (of some form) of doing so. Now you have a policy which treats white applicants differently than other applicants. What justifies doing something that is prima facie unjust? You have your answer if you can show that it is required as a matter of compensatory justice. But you agree that’s not on. So now either you claim that the practice needs no justification (in which case you have to show why it is not, despite appearances, a matter of discrimination on the basis of race; good luck with that), or that although it needs justification, it is justified. What’s the justification? And how does it trump the prima facie demand of justice that one not engage in discrimination on the basis of race? That’s the question, I’d think.

  4. The “conflict” you refer to is between the moral theories. Utilitarian and deontological theories are not consistent with one another–which was my point. But, alone, either theory leads to absurd choices being justified as moral.

    One on hand you can have trivial gains to the majority being used to deny important rights to the minority. On the other, you can have duty-bound individuals living in complete misery.

    So, the moral choice is to appropriately balance liberty and utility, as determined by the context. That this balancing cannot be simply obtained through philosophical reasoning is what I was arguing, namely that the standard brands of moral reasoning are of limited value in making moral choices.

    1. Sven, two things.

      1. Part of the point of my reply to Jim was that the formal structure of utilitarian and deontological theories does not suffice to differentiate them. You can make either into a species of the other with the right ancillary assumptions. True, with those assumptions they are distinct. But it is the ancillary assumptions that make them so.

      2. I am not very optimistic about the balancing story. Why? It provides no conflict resolution, since conflicts about those balances (maybe better: what constitutes the right balance in particular cases) are at least as deep and intractable as the original conflicts the balancing is supposed to adjudicate. Context really doesn’t determine anything by itself, because two people can reach exactly opposing conclusions in precisely the same context.

      1. 1. Interesting. This depth of philosophy you are referring to is way above my pay grade (as Obama would say), so I’ll have to take your word for it. Aren’t there, however, lots of adherents to the utilitarian and deontological approaches that would disagree that their formal structures can’t be differentiated? Just wondering.

        2. I don’t think value balancing is very promising either, as a moral philosophy that is (though I think there are people trying to do that sort of thing). But that is OK. As I’ve said, I’m highly skeptical of moral philosophy leading to morality anyway. There are, of course, other pathways to making moral decisions (or even having moral principles) than philosophy. I do, however, very much admire moral philosophy as a worthy intellectual pursuit.

      2. I’m replying to my own post to reply to Sven, since I can’t seem to do it otherwise.

        1. First, on formal structures. Of course there are natural and intuitive ways to capture the distinction between maximizing views like utilitarianism and consequentialism and deontological views. But it turns out to be (perhaps not surprisingly) very difficult to capture what they are. This is what makes it plausible to think that both utilitarians and deontologists think they are capturing our strong moral intuitions. They are, but they are doing so in ways that, to differentiate, takes more work than we might naturally suppose.

        2. I agree with you that moral philosophy doesn’t tell us much about how to arrive at sound decisions. In fact, I think it tells us very little. But I think it’s a mistake to expect it to. I think Aristotle had that one about pegged 2500 years ago.

  5. I am unable to reply to Mark LeBar’s April 29th post up above, so I am forced to append it here…

    Mark LeBar says:
    “What justifies doing something that is prima facie unjust? You have your answer if you can show that it is required as a matter of compensatory justice.”

    Mark, we have definitely miscommunicated. I am not defending affirmative action. If I were, yes, I would use “compensatory” justice as the centerpiece of my rationale.

    I am criticizing what some people think is a FLAW of affirmative action. Many people like to assert it is reverse discrimination. These people are wrong. The flaw they are attempting to describe is merely a violation of distributive justice. Not reverse discrimination. AA lacks the mens rea required to consider it to be reverse discrimination. (If there were infinite potential admission spots, no other races would be displaced.)

    1. Highlyverbal, I think I follow you, but I’m still not sure, and I’m intrigued by this last point. So do you think that wrongful discrimination has, as a necessary condition, the thought that even when there is no scarcity one aims to harm or deny benefit to somebody? That’s an interesting idea, and I agree it wouldn’t be met in most of the notorious cases. But I’m inclined to think it is too strong. I’d think what would be necessary would be a different intent, something like this: When scarce goods have to be allocated, take into account a feature of the recipients the irrelevant question of their race. After all, the beginning of moral concern about racial discrimination is the thought that race is simply irrelevant as a ground for most kinds of decision-making.

  6. Yes. I like to start the debate from conditions of no scarcity because it shows promoting one race != discriminating against another. (The etymologies suggest it; ‘pro-‘ means towards but “dis-” means apart.)

    Perhaps we might consider some more examples:

    Example 1:
    I attended college in the late 80s when affirmative action was a politicized issue. Sometimes, when it would arise in late night bull sessions (at a university with an AA program at that time), participants in the discussion who were black claimed they “didn’t check the box.” In other words, they wished to avoid any stigma or doubts about their merits by asserting they didn’t need AA. Now, there was always some degree of skepticism about these claims, but we could imagine that some fraction might be telling the truth. This creates the following interesting possibility: a black person benefiting from AA might very well knock out another black person, who didn’t want to check the box. Don’t we want our definition of discrimination to be robust enough to exclude this case? (Somehow, when it is a white person who gets bumped, scarcity and luck of the draw add up to discrimination?)

    Example 2:
    Imagine an admissions office at Hypothetical University where the savants in this office have created an affirmative action policy where every applicant universally receives an accurate bonus based on how disadvantaged they are in society. Now, this Compensation Index (CI) includes many things and is weighted based on decades of sociological admissions data, like economic status, education level of parents, etc. But it also includes one component for race, too. Now, it is very reasonable to imagine that there might be cases where a black applicant with moderate grades and test scores might get a slight bump from the race component of the CI and displace a white applicant with a huge CI from the other parts… meaning poorer grades and test scores but from a very troubled background. With no AA policy, the black person would be admitted first. With the CI AA policy, the black person is also admitted first. If we consider promoting one race the equivalent of discrimination, the white person might litigate to remove the race component while preserving the rest of the CI and demand admission over someone more qualified? The word “irrelevant” in your post is very problematic… once we are considering any compensation, what might sneak in with it?

    Example 3
    Imagine that at Hypothetical University, they amend the admissions policy to include a Retention Index that augments the CI for admissions decisions. The RI measures how good a student someone is likely to be (how likely to matriculate, how likely be famous, etc), and again it is based on demographics including race. We might imagine this time middle-class whites get a bonus but upper- and lower-class whites get a reduction. And, upper- and middle-class blacks get a bonus, and lower-class ones get a severe reduction. So now you have upper-class whites claiming they are discriminated against compared to upper-class blacks; the middle-class is treated completely equally; and blacks are suing over the setting for the lower-class. Heck, it’s MULTI-discriminatory. But then a clever intern runs the numbers that year without the RI part, and the exact same number of each race is being admitted, it’s just different individuals … so do results matter? Or intent? Or is it the process?

    My point here is, the policy of adding bonuses is being applied universally. When the index or weighting can make the claims of discrimination flip-flop, we need a better definition.


    Enough examples. I am not an advocate of AA, but I do know I would be remiss if I failed to point out: people who support AA believe that grades + test scores are not the final word in academic merit. If there is some societal imbalance with respect to race, some resulting taint of grades + test scores seems inevitable. So they would question whether there is truly some “objective” evaluation of students that you are accusing them of deviating from… perhaps they would suggest that you re-evaluate the “discrimination-neutral” parts of your model that you are trying to preserve.

    1. Well, you might guess that I’m not satisfied. That’s largely because what seems to have occurred in most of these cases is that we’ve lost track of where the moral problem is in the first place. If the thought is, what’s wrong is disproportionate representation in some population (say, too few of one race in a student body), my response is: go ahead and make a case that that’s wrong. I think the task of showing that there is a moral problem with disproportionate representation per se is insuperable; there just is no view that can sustain that outcome.

      But we don’t ask that because we think that typically such outcomes represent pernicious processes leading to them. The salient moral harm would occur when one used race as a criterion for conferring some benefit or acknowledgement or the like when it is irrelevant to doing so. (You claim that the idea of ‘irrelevancy’ here is “problematic.” I disagree. I think it is contentious and controversial, but claims as to relevancy always are. And it sure looks like sometimes race is relevant and sometimes not.) For example, one might think (I would certainly think) that race would be irrelevant to the decision whether or not some educational opportunity ought to be offered to someone. If that is so, then there is a moral problem with including race as a criterion for admission. And that is true whether the criterion is intended to work for or against disadvantaged minorities. (And that means, as I think is the right result, that the “non-box-checking” black is wronged by the process which excludes him.)

      Now, you can abandon or weaken that claim, and hold that race is relevant across a wider variety of choice situations than one might think. For example, one can claim that race is relevant to college admissions. But then one has a problem in explaining what is wrong with processes that for decades excluded students in virtue of their race. The obvious moral problem can no longer be the moral problem. And I don’t think there’s a good Plan B for making a moral case against discrimination in any form. Ultimately, it’s the arbitrariness of such discrimination that makes it a failure of rationality and a way in which one person can wrong another. That’s a story about our reasons for what we do, not the way the world turns out.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s