Utilitarianism is concerned with maximisation of expected utility, and is not concerned with distribution of utility. One unit of pleasure in Jones and five units in McTavish is neither better nor worse than three in Jones and three in McTavish. A utilitarian has to think of people rather as buckets into which happiness can be poured. (The analogy is imperfect, because buckets cannot have negative amounts of liquid, whereas people can be positively unhappy.) Indeed, properly speaking, utilitarianism, as I conceive it, should be universalistic: the happiness of non-human animals, and of extra-terrestrial beings too, should we have any dealing with or effects on them, should count equally. Of course sentient beings can have varying capacities for pleasure: I doubt whether a lizard is really conscious, while sheep almost certainly are. The buckets can vary in size. In practice the very small buckets will not affect our calculations, but medium sized ones will. If there should be scepticism about the consciousness of non-human animals I would remark that in practice we deal not with utility but with expected utility: as Bishop Butler said, probability is the very guide of life. So a 50% probability of consciousness in two million sheep is equivalent, for our purposes, of [sic] certainty of consciousness in a million sheep. Our concern for non-human animals cannot be shrugged off so cheaply.
There would be a technical problem if we wished to maximise a combination of amount of utility and distribution of it. We would have to have some common measure of the two things. One cannot simultaneously maximise the values of two independent variables. (This particular objection would not apply to an ethical theory that made considerations of fairness rigid constraints within which maximisation could then occur. However this rigidity of constraint would provide other difficulties for such a theory.) The main consideration is the following one. The utilitarian compares two or more different persons (or animals, etc.) with two different temporal stages of one person. Thus I do not mind sacrificing my happiness this year for greater happiness in some future year. So what is wrong with sacrificing Jones' happiness for the greater happiness of McTavish? Hence the bucket analogy. In practice, however, the tendency of utilitarianism may well be egalitarian. Equal, or at least not too vastly unequal, distribution of money (or other wealth) will probably increase the total happiness. This is an empirical matter, not one of ethical principle.
(J. J. C. Smart, "Utilitarianism and Punishment," Israel Law Review 25 [1991]: 360-75, at 361-2 [italics in original])
Note from KBJ: I'm sorry, but I'm not a bucket, and any normative ethical theory that thinks of me as a bucket is a nonstarter.
Keith Burgess-Jackson, J.D., Ph.D. 