Mises Daily Articles
Is Utilitarianism Viable?
Ethics as Social Science: The Moral Philosophy of Social Cooperation by Leland B. Yeager (Edward Elgar, 2001. viii + 299 pgs.
Reviewed by Robert P. Murphy
Hi, my name’s Bob, and I’m a recovering utilitarian. It’s been a few months now since I admitted the truth to myself. You see, I used to believe that all moral issues could be reduced to a simple maximization of utility. Many of my closest friends told me I had a problem, but I refused to listen. They meant well, but they just didn’t get it. Or so I thought.
I find myself in an odd position. Had I read Professor Yeager’s book a mere six months ago, I would have heralded it as a fine contribution to the clarification of a sophisticated (and correct) utilitarianism. True, there would have remained (in my opinion) one yawning chasm yet to be satisfactorily explained by the utilitarians, but this wouldn’t have bothered me too much; after all, science never rests, and can only push back the boundaries of our ignorance.
I have since changed my mind, and now consider this previously minor shortcoming in the theory (explained below) to be in fact a devastating weakness in utilitarianism. Yeager’s book ingeniously deals with many criticisms (since uncharitable critics force the utilitarian to devote most of his energies to explaining what he does not mean), yet it fails in this vital regard. As with the Ptolemaic models of the solar system (which were helpful in many respects), I believe it is time to admit that utilitarianism must be discarded in favor of a new approach to ethics.
* * *
Yeager advocates a utilitarianism of the variety articulated by Ludwig von Mises in Human Action. The normative terms good and bad make sense only in the context of society, and something—be it an action, law, or ideology—is morally good only insofar as it promotes human happiness or satisfaction. Because of the nature of human interaction—a nature discovered by the positive economist as well as other scientists—social cooperation is the means by which individuals satisfy their diverse, subjective ends, and thus ethical issues can be reduced to the expected strengthening or weakening of social bonds. In short, what promotes cooperation is good, what hinders it is bad.
I am nothing but an amateur in the field of ethics, so I do not want to comment on the complicated web of arguments and counterarguments that Yeager referees in his book. (Indeed, I often found myself unsure whether Yeager agreed or not with his paraphrased remarks of a given author.) My relative ignorance notwithstanding, I am confident that Ethics as Social Science is the single best reference for the serious student—whereas Henry Hazlitt’s The Foundations of Morality is still a better introductory book.
After this deserved praise, and in light of my recent "conversion," I shall devote the rest of this review to the serious problems I perceive in Yeager’s book. Assuming that the readers of this Web site are familiar with the basic ethical philosophy of modern utilitarianism as expounded by Mises and Hazlitt, I will get right to the heart of the matter as I see it.
* * *
The fundamental problem with utilitarianism is this: Despite a succession of ingenious proponents, its advocates have yet to explain why the individual should behave morally. The fact that we are all better off if we all behave morally is utterly true and utterly irrelevant. (Such an argument violates the cherished Austrian precepts of marginalism and individualism.) The truly difficult moral issues resemble the familiar Prisoner’s Dilemma; regardless of everyone else’s behavior, the individual does better by exploiting others. It is true that a society suffering from widespread theft would be intolerable, even from a thief’s point of view, but any individual robbery has very little impact on the overall level of crime.
Yeager does the best he can to address this problem. He explains (pp. 169–172) that work in economics and biology shows that cooperation can yield a higher lifetime utility—or evolutionary fitness—since exploiters can be punished in the future for current selfish behavior. He points out (pp. 164–167) that psychological reasons may also explain why "honesty is the best policy." These clever and interesting results mitigate the apparent conflict between egoism and altruism.
Nonetheless, they merely show that the standard examples of critics need to be picked more carefully. No appeal to "tit-for-tat" strategies can evade the dilemma of someone called upon to sacrifice his life. Yeager has the honesty to acknowledge this:
We can understand the general utility of a rule requiring a person, in rare extreme cases, to sacrifice his interests and even his life for others’ sake. Such a duty might be part of the terms of employment of a . . . soldier. . . . Morality can require fulfilling responsibilities even when, in exceptional extreme cases, doing so is ruinous to oneself. Holding oneself and others to the rules is in the long-run interest of everyone. Once we tolerate anyone’s making an exception in his own favor, we subvert the purpose that the rules serve. . . .
But suppose that one of these extreme cases had actually arisen and that you, a superior officer, were in radio contact with the person whose self-sacrifice you now required. How could you persuade him to perform his duty? You could hardly argue that his doing so would be in his own interest as well as in the general interest, for his own interest would be to save himself. You might appeal to his interest in being remembered as a hero rather than surviving as a coward despised by himself and others. But that argument is hardly compelling: it amounts to denying, after all, the postulated exceptional but genuine clash between self-interest and social interest.
. . . If [the soldier] had a deeply ingrained moral character, he would perform his duty anyway. Furthermore, it would have been in his own interest throughout his life up to the time in question to have that sort of character, as opposed to the character of someone ready to shirk his duty whenever shirking seemed expedient.
The rules calling for self-sacrifice in the exceptional case are directed to persons in the abstract who might conceivably some day find themselves in the postulated exigent roles and situations. They are not especially directed to specific persons whose lives are already at stake (p. 186).
In other words, if you pause to wonder why you should ever die for a cause, then the honest utilitarian must admit: You should not.
This candid admission, in my opinion, is fatal to utilitarianism. There are all sorts of situations—i.e., not simply the soldier being asked to take a hill—where conventional morality requires an individual to forego genuine (i.e. long-run) gain. References to the benefits of a virtuous character will not convince someone who is lacking such a character in the first place.
Moreover, if everyone agreed with Yeager and other utilitarians that it were foolish to sacrifice oneself in these rare instances, an element of doubt would arise in all social interactions. Although pangs of conscience might be a wonderful evolutionary byproduct, it would be in the interest of everyone to steel himself against such "irrational" feelings (while still behaving in accordance with them under normal circumstances). One’s very life might one day depend on it.
It doesn’t really matter whether my conjecture is empirically true. The decisive issue is that, if it were true—that is, if the level of conventionally moral behavior did in fact deteriorate over time, until everyone viewed each other as a potentially deadly enemy—the utilitarian would have nothing much to say. He might lament the trend, but only in the way an astronomer would lament a comet hurtling toward Earth. Throughout the process, the utilitarian could not condemn anyone’s actions as immoral.
* * *
Utilitarianism seems to rob the words good and bad of their specifically ethical character. The utilitarian cannot make a distinction between guilt and simple error. The person who robs a bank to achieve happiness has made a mistake in qualitatively the same sense as a person who overcooks a steak. (I do not believe Yeager specifically addresses this type of objection.)
In fact, we can go further. Is it really true, for example, that Josef Stalin acted against his interests, even in the long run? Does the utilitarian really concede that our possible condemnation of Stalin is purely an empirical matter? (It might be true that had every other Soviet acted in his true interests, dictatorship would have been impossible. But this is dodging the issue.)
* * *
To get around the last example and others like it, Yeager would presumably bring up his emphatic emphasis on rules versus act utilitarianism—his arguments for which I find very weak indeed—and on the generalization principle. In a discussion arguing for the importance of such a principle, Yeager somewhat whimsically asks what objections the utilitarian could raise against an asserted ethical criterion of satisfying the desires of economist Gordon Tullock (at the expense of everyone else):
An ethical doctrine could not be beneficial unless it could be expected to guide behavior, by and large. A doctrine, along with the practices and institutions it recommends, should be capable of becoming transparent, of being publicly and seriously espoused. People should be able to accept it even after reflecting on its meaning and grounding. . . . Its credibility would suffer . . . if it were patently rigged to the differential advantage of specific individuals or groups at the expense of people in general (p. 190).
I do not see that this rescues Yeager in the case of Stalin. Stalin's absolute rule was well-publicized and was nothing if not serious. Although we would all agree that it was patently rigged to the differential advantage of party members, its credibility survived for Stalin’s entire life. (And the possible objection that what is immoral in private life remains so in public life is not available to Yeager, who makes quite clear that he is not an anarchist.)
* * *
Another problem with Yeager’s utilitarianism is its vacuity. Although he points out (p. 230) that this is not true in a formal sense—since, after all, we could offer all sorts of ultimate criteria, rather than human happiness—I still think Yeager’s system comes close to being nonfalsifiable. (Yeager, who reminds his reader in several places (e.g., p. 143) that he is more "scientific" than many of his critics, will certainly be concerned with such a charge.)
To demonstrate this, let us consider Yeager’s treatment of a sophisticated argument against utilitarianism. (Note too Yeager’s less than disinterested tone.)
An example contrived by J.A. Mirrlees . . . affords insight into what the critics may have in mind when complaining about aggregation (and if they are not just mindlessly echoing one another). Social welfare is the arithmetic sum of the utilities of society’s two members, Tom and Dick. Both have the same utility function with plausible properties. . . . Intuitively [and as a result of the specified properties], working a lot leaves a person less time and capacity to enjoy income, and having a large income reduces one’s willingness to work. The production function is very simple: 1 unit of Tom’s labor produces 2 units of real income, while 1 unit of Dick’s labor produces only 1 unit of income. Only in their productivities do the two men differ.
On these radically simple assumptions, straightforward calculus shows that maximizing social welfare requires Tom both to work more hours and also to receive less income than Dick. Tom, though more productive, is unequivocally the worse off of the two. . . . Tom has both absolute and comparative advantages over Dick in transforming labor into income. In transforming income into utility and welfare, Dick has neither an absolute advantage nor an absolute disadvantage, but he does have a comparative advantage. Tom should therefore specialize in producing income and Dick specialize in consuming it. . . .
Any (imaginary) utilitarian who remained content with considerations of this sort would be committing grave oversights. Most obviously . . . he would be overlooking incentives. . . . Considerations of fairness must also enter into any even halfway sophisticated version of utilitarianism. An ethical code cannot promote the welfares [sic] of individuals unless it commands wide adherence, which it cannot do if seen as grossly unfair (pp. 113-114, emphasis added.)
It is this latter counterargument which troubles me. What more can the critic do than to show that utilitarianism, while obviously plausible in certain cases, leads to absurdities in other cases? Yeager seems to come very close to the following defense:
- The critic argues that utilitarianism would imply monstrous consequence X.
- People do not like monstrosities.
- Therefore, by its very definition, utilitarianism must not imply X after all.
(Before leaving the topic of aggregation of utilities, I should mention that I was amazed to find no mention of Arrow’s Impossibility Theorem—which rules out any ordinal social welfare function—despite the fact that Yeager discusses at some length various welfare economists and their often technical arguments.)
* * *
Besides these major issues, there was another aspect of Ethics as Social Science that troubled me. As alluded to above, Yeager at times appears rather partisan. He voices actual "outrage" over the (admittedly) unsympathetic treatment of utilitarianism by certain critics, yet he himself yields to such caricatures of his opponents. For example, he accuses Walter Block of "an unintended, unrecognized, and paradoxical statism—the unarticulated idea that the state is responsible for suppressing all evil and promoting all good and that something the state should not suppress is by that very token not evil and perhaps even good after all" (p. 276).
Although I agree that Block, in his Defending the Undefendable, at times overstepped—for example, in the case of counterfeiters—this possible error in no way implies statism. (The intentionally shocking sense in which Block uses the term "hero" is explained by Rothbard in the book’s Foreword.) I cannot see how Yeager reached his conclusion except for its utter irony in light of the professed antistatism of his opponent.
In discussing Murray Rothbard’s natural-law defense of blackmail, on the grounds that what is legal and what is moral need not coincide, Yeager (a) accuses Rothbard of the same subtle statism attributed to Block, and (b) incredibly wonders whether this distinction is valid after all (p. 279). Surely Yeager must be familiar with the standard arguments for drug legalization; his criticisms of Rothbard in this regard are simply incomprehensible to me.
(To address a certain ambiguity: In this context, natural-law arguments attempt to deduce a person’s rights, which then may be justly enforced by a legal apparatus. I cannot speak for other writers in this tradition, but Murray Rothbard certainly never intended to deduce the entire body of private morality from the nonaggression axiom. For Rothbard, what is illegal is thus a subset of what is immoral.)
* * *
Although providing a thoughtful tour of previous work and offering a few novel arguments, Leland Yeager’s Ethics as Social Science ultimately fails in its attempt to rescue utilitarianism from its many flaws. Precisely because thinkers of the caliber of Yeager, Mises, and Hazlitt were unable to expound the doctrine in a satisfactory way, I have come to conclude that such an exercise is impossible.