Utilitarianism, long the dominant moral framework for political philosophy, especially in America, has fallen from grace in recent years. The arguments for the inadequacy of utilitarian theory have been well rehearsed, and while some of those presented below are my own, I would be surprised to hear that any of them had not been previously advanced by someone else. My goal here is not to break new ground, but simply to summarize the reasons for rejecting utilitarianism which I find most compelling.
1) Separateness of Persons
John Rawls resurrected rights-talk in political philosophy with his claim that the “separateness of persons” is one of the most basic moral facts about human beings. Utilitarianism ignores this central fact by applying the “principle of decision for an individual to the society as a whole.” Robert Nozick sums this argument up nicely: “The moral side constraints upon what we may do, I claim, reflect the fact of our separate existences. They reflect the fact that no moral balancing act can take place among us; there is no moral oughtweighing of one of our lives by others so as to lead to a greater overall social good. There is no justified sacrifice of some of us for others.”
To see the force of this objection, consider the following two cases: A) Two people, both moderately content. B) Two people, one in utter bliss, the other suffering terrible agony. As long as the sum of utilities and disutilities is constant across the two cases, the utilitarian must be indifferent between them. If the blissful person in B is a bit more blissful, raising the total utility of scenario B ever so slightly above that of A, then the utilitarian must find B morally preferable. Most people, by contrast, would say that not just the total utility, but the distribution across persons matters. The pain of one person is not “canceled out” by the happiness of another.
For both Nozick and Rawls, this priority of the autonomous individual leads to the liberal priority of the (deontic) right over the (consequentialist) good. Of course, this in itself constitutes a positive claim about what kind of beings we are. Ultimately, I think this claim is at least partly wrong if, as some have charged in Rawls’s case, it’s based on a Kantian conception of persons, and by extension on Kant’s problematic metaphysics. Still, I think this problem just points the way for fruitful inquiry, rather than thoroughly undermining liberal rights. I think this objection becomes stronger in light of the second & third.
2) Utility as Derivative
“Happiness” and “utility” are abstract concepts subsuming a broad range of particular (sometimes qualitatively very different) emotional and mental states. It may be useful, for purposes of analysis, to model an individual’s preferential choices as corresponding to numerical value assignments, but this is just a useful tool. I do not prefer action A to B *because* A has more “utility” any more than Florida is warmer than New York *because* it has a higher temperature in degrees. Preference is prior to utility, not explained by it. The act of choice does not consist of a maximization function for some single homogeneous quantity: it relies on an individual having an integrated system of desires and preferences with external objects. By this I mean that my objective when I, say, decide to write something, is not (primarily or directly) the pleasure of writing, but rather the production of a good piece of work. Assume this is not the case: that my goal really is the pleasure. In that case, we must ask why it is that I get such pleasure. Only because I think there’s some value in producing good writing. If this value is grounded only on utility, our explanation is puzzlingly circular.
The point of this line of argument is that “utility” is not primary: if happiness matters, this is because *people* matter, and each person is concerned with her own particular goals and projects. The principle of utility gets things backwards by making utility the object of moral value, and persons convenient receptacles of it.
3) Non-Comparability
If the notion of a single homogenous quantity maximized by personal choice is problematic, how much more so is the familiar question of how to compare utilites interpersonally? This is not just a reiteration of the “separateness” objection: there the problem was whether such comparisons were morally appropriate. Here it is whether the comparison is technically possible, even if appropriate. Again, our ability to compare two alternatives and rank one as more pleasurable (or what-have-you) is based on our integrated system of preferences. Only this makes possible a model of our preferences which uses a consistent numerical scale. But there is no “social” system of preferences, no superperson who feels all our pains and pleasures “equally”. If our mental experience is essentially private, it is not logically possible for me to be “inside your head”. That is, even if there were a social superperson, its preferences would just constitute one more system. Arrow’s Theorem, which showed how voting systems must fail in one of several ways in aggregating preferences, may be relevant here.
4) The Missing Superstructure
If utility is what matters, whose matters? Human beings’? Plants? Animals? If we include animals, which almost certainly can feel pleasure and pain in something similar to the way we do, we must accept a seriously revisionist moral theory, even if animal utilities are weighed less heavily than those of persons. If we limit our concern to human beings, or even weigh their satisfaction more heavily, we need some additional moral justification for selecting humans. If there is some such justification, however, it seems that utility must share the stage with a separate value. If certain considerations are morally weighty enough to define the scope of the utility principle, it is not clear why they are not weighty enough to be incorporated into our first-order moral theory.
5) Integrity and Reasons for Action
The “integrity” critique of utilitarianism is usually associated with Bernard Williams (see Williams & J.J.C. Smart, Utilitarianism: For and Against. Cambridge Universiy Press, 1973). Williams argues that utilitarianism is unreasonable because it asks us to be neutral towards (and therefore in some sense distanced from) our own strongest commitments. Here the communitarian critique of the liberal self, which focuses on the problematic commitment to Kantian autonomy alluded to above, can be brought to bear on utilitarianism. If we are what Michael Sandel would call “situated selves,” then the utilitarian demand of impartiality is literally inhuman. Instead of upholding the most imprtant human values, the search for an impersonal “moral point of view” culminates in their annihilation.
This is not merely the (almost certainly true) claim that people who turn themselves into strict utilitarian calculating machines will be unhappy. This would merely make the theory “self-effacing,” as Derek Parfit claims all consequentialist theories are likely to be. In that case, utilitarian reasoning will simply tell us to be rule-followers, rather than employing utility reasoning directly.
Rather, the claim is that we are constituted by our aims and ends in a way that makes utilitarian reasoning unacceptable at the highest level, not merely as a practical decision principle. It might make us unhappy to think like utilitarians in our daily lives, but this is not the point. The point is that we cannot accept an impersonal utilitarian framework, even for the construction of rules for action. The contradiction between our “encumbered” selves and the radically impersonal conception of “the moral point of view” remains: the latter is too alien to provide us with a source of reasons for action. This problem is less obvious when we’re talking about desirable rules for institutions, but it’s still there. The utilitarian considerations which indicate such-and-such a legal or social order must also motivate my compliance with that order. If I consider a regime justified on utilitarian grounds to be legitimate, I must believe that utility is the standard of moral goodness generally.
It’s true, we can consider our “encumbered” commitments as reasons for crafting rules in a certain way. We might follow a rule to help our friends and loved ones, rather than promote utility generally, because we recognize a general human tendency to form personal bonds, and because this will leave us specially suited to promote these people’s goods, and so on. But this just shows that rule utilitarianism can sometimes give us the right answers for the wrong reasons.
These are my major problems with utilitarian theory. I will not argue here for any particular alternative, but I do think these objections, taken together, provide a good prima facie reason for rejecting utilitarianism.
