right, and others absolutely always wrong, do, I think, generally hold that the former do also, as a matter of fact, absolutely always produce the best results, and the latter never. But, for the reasons given at the beginning of Chapter III, it is, I think, very unlikely that this belief can be justified. The total results of an action always depend, not merely on the specific nature of the action, but on the circumstances in which it is done; and the circumstances vary so greatly that it is, in most cases, extremely unlikely that any particular kind of action will absolutely always, in absolutely all circumstances, either produce or fail to produce the best possible results. For this reason, if we do take the view that right and wrong depend upon consequences, we must, I think, be prepared to doubt whether any particular kind of action whatever is absolutely always right or absolutely always wrong. For instance, however we define “murder,” it is unlikely that absolutely no case will ever occur in which it would be right to commit a murder; and, however we define “justice,” it is unlikely that no case will ever occur
