Law & Society Blog

/// What follows is the final installment of a series of examinations of utilitarianism from Harwood’s seminal essay, “Eleven Objections to Utilitarianism”, along with my demonstrations of how these objections are unsound. Previous installments address such topics as integrity, justice, promise-keeping, supererogation, average and total utilitarianisms, rule utilitarianism, and hedonism. ///

10. Utilitarianism makes interpersonal comparisons of utility.
What utilitarianism demands, implicitly, is that the happiness of persons can be compared to one another (usually using some abstract idea of a “hedon”, or unit of pleasure). However, this in itself seems suspect, since the experience of happiness is entirely a) idiosyncratic, b) subjective, and c) variant between persons. Thus, it seems bizarre, or outright impossible, to say that one may compare happiness across persons.

Harwood rightly dismisses this objection for a variety of reasons.

First, he stresses that utilitarianism allows for different means by which satisfaction may be achieved. So the fact that needles make me miserable without doing the same to you, doesn’t indicate anything about pain itself. Thus, the idiosyncratic nature of pleasure and pain can be dealt with by utilitarians.

Second, he argues that we regularly and commonsensically make interpersonal comparisons of utility. To illustrate, he writes that “we build freeways even though we know that it is just a matter of time before and innocent baby who would not have died nearly so soon had the freeway never been built gets crushed in an automobile accident on the freeway. But the great convenience of the freeway and the other lives saved by allowing ambulances and other emergency vehicles to use the new freeway … outweighs the harm caused to the crushed baby”. This seems to defuse some of the motivation behind the objection, although, as an appeal to this mysterious thing we call “common sense”, it falls short of being persuasive.

Third, he argues that interpersonal comparisons of utility have just as many problems as intrapersonal comparisons of utility; all the same arguments made against the former can be made against the latter.

This isn’t true, however. I may compare my imagined happiness in the future to that of the present, or the present to the past, and my act of doing so does not interfere with the objection that both are subjective experiences. But if I were to try to compare my experiences to another person’s, I would run into problems, because experience is always unique to the subject. The point — which is worth taking seriously — is that you can’t just lay out the contents of one person’s conscious mind, lay out those of another person’s, and compare the two. Minds just aren’t like that.

Still, I think that, for all intents and purposes, we may make comparisons between persons by taking our analyses with a few grains of salt. But even if we were to allow that such comparisons could be made in principle, we can rightfully ask, “Well, how? Put it in empirical terms, already”.

It is said that utilitarianism has served to be an inspiration to economic theorists, but one thing that economists have grown to accept is that happiness cannot be measured en masse. Rather, they tend to rely upon observing the preferences of persons, postulating through the use of models of human action that there is a connection between desire satisfaction and happiness. An economic sociologist and utilitarian could then argue that certain models of social action allow us to infer levels of happiness, at least in part, on the basis of satisfying desires.

I won’t begrudge the general epistemic point, although the calculation depends upon a plausible model of human action. The favorite of classical economists and vonMiseans, rational choice theory, is something which we now know is implausible, and has rightfully fallen out of grace across the social sciences. Still, I mention these things only to point out yet another place where utilitarianism appears to be in desperate need for coherant and plausible analytic social philosophy to help provide genuine moral guidance.

[Edit: So those are Harwood’s first few rejoinders. But he also considers a second kind of thought-experiment. Let’s say that there are two kinds of people: those who are disposed towards dourness, and those disposed towards contentment. The same effort that it would take to make the dour person happier by 2 hedons could also make the contented person happier by 4. This experiment tells us, then, that utilitarianism councils us to expend more effort on those who are easily amused, since they will produce greater hedonic dividends.]

Harwood’s objection to this experiment is fairly brilliant and deserves to be quoted at length. He accuses a thesis of variance between interpersonal happiness as leading to bad statistical mojo. “Even if satisfaction was as wildly unpredictable from person to person… would it not mizimize our margin of error if we assumed that each person’s satisfactions were comparable? It seems so. For if we started giving preference or extra weight to persons whose satisfaction was assumed to be weightier [i.e., someone who is easily and intensely delighted], then the wildly unpredictable nature of satisfaction … would imply that we are as apt to be preferring and weighting the satisfactions of the right persons… as those of the wrong persons. If we chose the wrong person and thus gave extra weight… to the satisfactions of a [dour person]… then we have compounded any mistake we would have made by considering all satisfactions comparable, and we have extended the margin of error further than it was.”

[Edit: I really haven’t got much to add to this point, because I think Harwood offers a persuasive argument relative to what he’s arguing against. My only caveat would be to point out that the objection need not state that the differences in temperament are wildly unpredictable in order to be morally salient. It might be obvious that some people get more happiness than others as a matter of character; it is not at all unpredictable. And so it really would seem worthwhile to help these persons over those of more phlegmatic temperament, which seems wrong.

The trouble with all this is the habit which it forces us to endorse a habit of arbitrarily helping certain persons over others, which (if taken to such extremes where moral intuitions would come into play) would violate the optimizing spirit of utility.]

11. Utilitarianism forces us to violate the publicity condition.
Ethics, to put it roughly, is the guidance towards right thought, action, and life. This rough definition uses the word “guidance”, which seems to imply a necessary social element. Based on these axiomatic truths, we can understand that ethics, in order for it to be what it is, must be a public phenomenon. It can’t be secretive.

But some utilitarians have argued that “it would be a mistake to let most or all people directly pursue maximizing satisfaction, since too many will show bias or incompetence in calculating what will maximize satisfaction”. I can see very good sense in this objection, because (as we’ve seen in this series) a number of unsound objections are based upon bizarre, unrealistic worldviews can use utilitarianism to create nasty and evil results. Utilitarianism is dangerous when used improperly.

This has caused some to remark that utilitarianism is “elitist” or “secretive”. But that’s not getting to the core of the problem. To put it another way, the moratorium on publicity can be invoked by admitting the sheer impracticability of making utilitarian calculations in the real world. We have already dismissed objective utilitarianism because it provides no succor except to those who can see the future — that is to say, none of us. Meanwhile, subjective utilitarianism gives us a wide range of possibilities and risks, which requires some cool-headed and informed deliberation which we often don’t have the opportunity to engage in.

This objection concerns me more than any of the others because it seems to be quite strong. The publicity condition must be defended. The intuition behind the doctrine arises out of an inescapable conviction that unwritten laws must be either exposed and treated with rational scrutiny, or be ignored. Human social life is largely a mystery for me, but mindless conformity to (and rationalization of) norms strikes me as one component of Arendt’s banality of evil, mentioned earlier. An attack upon unwritten laws strikes me as an attack upon that very thing. When we take it seriously, we (rightfully) understand that for any abstract system of laws which direct action, thought, or life, and which is incommunicable, or (just as well) are based upon mere thoughts or beliefs instead of speech acts, then those systems are amoral by default.

Moreover, the publicity condition seems to be necessary because it is supported by the most plausible doctrines of accountability. We may formulate a person as accountable for some act if they could not possibly be blameworthy (or praiseworthy) by a moral agent. And a person cannot be blamed for a law to which they had no way of knowing. Thus, the laws which a person are held to, must be made public, so that persons may be held accountable to them.

My argument is in favor of examining utilitarianism in a slightly different way. It should no longer be considered a fundamental norm; instead, it is a kind of supportive groundwork for any fundamental norms at all. What I would like to say is that utilitarianism is a meta-ethic. This kind of utilitarianism (or “meta-utilitarianism”) provides guidance to the ethicist, but not guidance in general, simply because when taken alone it wouldn’t be very good at getting the job done. Its purpose for the ethicist is to provide clarity and guidance when reasoning concerning moral quandries, and to spur forward more nuanced analyses of social science. Ultimately, however, guidance must be intelligible in short form, either in the form of rules or narratives, and these things will be more helpful in everyday life. Since its audience is limited to ethicists, it does not need to satisfy the publicity condition.

I can imagine the reader getting impatient, and yelling: “that’s still elitist!”. And certainly, meta-utilitarianism (MU) seems more aloof than classical utilitarianism in some way. The way I think of MU is as if it shared a relationship with ethics like that between geometry and soccer. [Special thanks to Mitchell Langbert (CUNY) for his emphasis on this fact in discussion]. Zinedine Zidane didn’t need to do complex calculations during the World Cup, but every single move he made was still compatible with geometric truths. Similarly, the virtuous person doesn’t need to perform utilitarian calculations in every ethical dilemma; but for any plausible ethical reasoning, it will have to be compatible with utility. I don’t think it is elitist for that fact, any more than geometry is elitist.

I come to this conclusion on the basis of examining the evidence in the harder cases presented in the series. None of them seem to be decisive, many often missing the point. But all of them conclusively show that more explicit decision procedures towards judgments are needed than mere postulation of the principle of utility. Indeed, utilitarianism does not seem to exhaust all that there is to say on the subject of moral guidance; to use just one example, it is not usually obvious when, and to what extent, considerations of the social system are to be counterweighed against considerations about individual happiness, and these things make the project difficult to keep up in practice.

Some theorists may interpret this formulation as the last cries of the theory itself. If we are speaking in terms of the Benthamian calculus, and of archetypical utilitarians who are so skilled as to compute their way into infallible judgments at every turn, then in indeed, the theory is dead, has long been dead, and has only the scarcest hope of resurrection. But what if we take these dismissals more broadly, to include a rejection of the entire family of utilitarian theories (and especially MU)? No, I do not think such accusations can succeed. In the way I have presented it, meta-utilitarianism is, and should be, forever a part of the landscape of moral thinking. And even if there are more objections (as there always are), my more modest goal here has been to corall the efforts into tasks with a hope of success. If the theory at large, or MU in the small, are ever to be defeated, it won’t be through these unsound examinations of utilitarianism.

—
Summary of the important working parts of the argument

i. Agency is an incorrigible value.
ii. Enjoyment is a product of choice, and is unique to beings with agency.
iii. If supererogation is a significant concept, it derives out of successful moral risks. If it isn’t significant, then virtue serves as a replacement concept.
iv. Virtue is, in part, the fulfillment of agentic duties: the obligations to know and learn, and to empower oneself.
v. Counterfactual cases are significant only when they represent plausible or intelligible universes.
vi. Trust compels promise-keeping.
vii. Truth is not an incorrigible value. It is sometimes wrong to tell the truth. This counters proposals by thinkers like W.K. Clifford.
viii. Rights demand commitment at the risk of making them powerless.
ix. All “rules” are really just principles with extra weight. Rule-following is associated with positive duties more than negative ones. Nevertheless, the demands of utilitarianism apply across all situations.
x. The experience of waking from the experience machine serves as a reason to doubt that the machine is reliable. Also, the experience of waking from the machine rightly inspires moral horror. These are reasons why a person might not be advised to enter the machine.
xi. Hedonism is a thesis about the intrinsically valuable. It is not a thesis about right and wrong.
xii. Utilitarianism seems to violate the publicity condition. Until peripheral questions are answered regarding values and social systems, it must be understood only as a meta-ethic.

License

This work is published under a Creative Commons Attribution-Noncommercial 2.5 License.