[350] This passage, which in the second and subsequent editions occurred in chap. ii. of Book i., was omitted by Professor Sidgwick from that chapter in the sixth edition, with the intention of incorporating it in Book iv., which he did not live to revise.
[351] Cf. J. S. Mill, Utilitarianism, chap. ii. Mill, however, only affirms that the “rules of morality for the multitude” are to be accepted by the philosopher provisionally, until he has got something better.
[352] I refer to the abstract principles of Prudence, Justice, and Rational Benevolence as defined in chap. [xiii.] of the preceding Book.
[353] Theory of Moral Sentiments, Book i.
[354] This operation of sympathy is strikingly illustrated in the penal codes of primitive communities, both by the mildness of the punishments inflicted for homicide, and by the startling differences between the penalties allotted to the same crime according as the criminal was taken in the act or not. “It is curious to observe,” says Sir H. Maine (Ancient Law, chap. x.), “how completely the men of primitive times were persuaded that the impulses of the injured person were the proper measure of the vengeance he was entitled to exact, and how literally they imitated the probable rise and fall of his passions in fixing the scale of punishment.” And even in more civilised societies there is a very common feeling of uncertainty as to the propriety of inflicting punishment for crimes committed long ago, which seems traceable to the same source.
[355] No doubt this influence is confined within strict limits: no authority can permanently impose on men regulations flagrantly infelicific: and the most practically originative of religious teachers have produced their effect chiefly by giving new force and vividness to sentiments already existing (and recognised as properly authoritive) in the society upon which they acted. Still, it might have made a great difference to the human race if (e.g.) Mohammed had been fond of wine, and indifferent to women.
[356] On this point I shall have occasion to speak further in the next section.
[357] I refer especially to the views put forward by Mr. Spencer in the concluding chapters of his Data of Ethics.
[358] This definition, however, does not seem to me admissible, from a utilitarian point of view: since a society in this sense perfect might not realise the maximum of possible happiness; it might still be capable of a material increase of happiness through pleasures involving a slight alloy of pain, such as Mr. Spencer’s view of perfection would exclude.
[359] See especially chap. ix. Pars. 12-15.