Utilitarianism and rationality

Utilitarianism is an attempt to choose our moral standards by rational criteria and, more specifically, by asking what moral standards would maximize expected social utility. Here the adjective “expected” is short for “the expected value of” whereas social utility itself is defined either as the sum or as the arithmetic mean of all individuals' utility levels in the relevant society. (Some utilitarians prefer the first definition, others prefer the second. But, in actual fact, in analyzing specific moral problems, other than those of population policy, the total number of individuals can be regarded as given so that maximizing the sum of individual utilities will be mathematically equivalent to maximizing their arithmetic mean.) These definitions, of course, presuppose that individual utilities are cardinally measurable and admit of meaningful interpersonal comparisons (cf. Harsanyi 1955, 1976, Chapter 4).

No doubt, nonutilitarians will deny that social utility is the right criterion for morality. But it seems to me that, even for nonutilitarian social scientists and philosophers, it should be a question of some interest what kind of social standards would in fact maximize social utility.

As is well known, there are two alternative interpretations of utilitarian theory. One is act utilitarianism, which defines a morally right action as one that would maximize expected social utility in the current situation. The other is rule utilitarianism, which makes the choice of a morally right action into a two-step procedure.