This is a favourite commonplace with eighteenth and early nineteenth century writers.

But it is very difficult to prove that we can best secure our own happiness by habitually regarding the happiness of others as equally important to us with our own. In point of fact, as has been said, the utilitarian is obliged to assume that a certain derivative and practical egoism is necessary. And it is quite impossible to prove that the interests of oneself and of others can never clash. It is mere abuse of terms to say that self-sacrifice may be the greatest pleasure. Whatever may be the pleasures of martyrdom, which of late have probably been somewhat overrated, they are always very "impure" (in the Benthamic sense. of the word), and usually very evanescent.

Failing the attempted reconciliation put forward by Dr. Sidgwick, we are driven either (1) to accept the ultimate reasonableness of egoism, or (2) to give up hedonism as a complete and independent system. If happiness is the summum bonum at all, it must be the happiness of each for each. This conclusion, with which Butler was content,' is not in itself so shocking as might be supposed. Practically, it would probably make no great difference in the action of the vast majority of those who are seeking to lead moral and reasonable lives. And although morality would suffer by a lowering of its ideals, and self-sacrifice would

See his sermon upon the Love of our Neighbour (XI.), which practically admits that the duty of self-love takes precedence over that of love to others.

lose one of its motives (the conviction that it is reason, able as well as natural), yet some compensation might be found in the promulgation of a short and easy method of being good, a royal road to virtue.

§ 8. Objections to Utilitarianism with regard to Distribution.

We have spoken at length of the difficulties common to all hedonistic methods. There are in the case of utilitarianism special difficulties relating to distribution. Bentham and his school put forward the Greatest Happiness of the Greatest Number as the one ultimate and necessary starting point of ethics. They asserted that a complete ethical system could be built up on it. As a matter of fact the maxim simply declares what is desirable, and requires to be supplemented by further statements, (1) as to the persons who are to be counted in the " number," and (2) as to how we are to divide the pleasure between them.

In the "greatest number" are we to include (1), all our own countrymen; (2), all white men; (3), all civilized persons; (4), all men now alive? Are we to include (5) posterity? And if so, how far in advance is our regard to go? The usual answer is that we are to regard all men, including all posterity, in so far as our action can influence their happiness. But this leaves us in a hopeless difficulty. We cannot tell how far the effects of our action may extend. And we shall somehow find that what benefits our own country

men or our own age, may be harmful to other men living or to posterity. And as foreigners, or at any rate posterity, will be an infinitely larger total than our own countrymen now alive, we may be driven to the conclusion that we must not regard the interests of the latter. We must not use up our coal for fear posterity may suffer. We must not add to our state debts for the same reason.

The difficulty becomes more if, as modern feeling seems to demand, we include the lower animals in our "greatest number." Anti-vivisectionists already demand that each dog or cat is to count for one, and to declare that remedies purchased at the cost of sufferings to a few scores of rabbits and puppies are in their nature immoral. We must no more cure hydrophobia by Pasteurism than cure skin diseases by bathing in the blood of children, like the sultan in the oriental story. Is every animal to count for one? or shall we have a table of fractions? However small the hedonic fraction assigned, say to herrings, the unbounded consumption of roe will in time extinguish as much happiness as even an epicure is capable of. The extension of the herring fishery, regarded as praiseworthy at Yarmouth, becomes a criminal matter when viewed from the point of view of the herring or the vegetarian.

On what principle are we to distribute a given amount of pleasure? Bentham lays down that "everybody is to count for one, and nobody for more than one;" and this principle has apparently been accepted by all utilitarians. It cannot be evolved from the

It

axiom that the Greatest Happiness of the Greatest Number is the ultimate aim of all virtuous action. is a new axiom, and involves the assumption of an extrinsic standard of justice; in other words, its adoption is a tacit confession that we cannot bring all virtues under the Greatest Happiness formula.

But what does the new axiom mean? It seems to assume that all men have the same capacity for pleasure and pain. This is notoriously untrue. There are certain kinds of enjoyment for which most men have very little or no capacity. The pleasures of art and of the pursuit of truth are confined to a small fraction of the civilized races. How many millions of people in the world derive any appreciable pleasure from hearing the best music or seeing the best pictures? We cannot distribute equal amounts of this sort of pleasure to a refined and educated man, the average artizan, and the average Kaffir, any more than we can distribute equal amounts of light to a clear-sighted person, a patient with cataract, and a blind man. The only rational plan will be to exclude the incapable from the distribution. To put it in another way. Some men have a greater quantitative capacity for pleasure than others. Suppose we have 4 units of pleasure to give away, and that Prospero's capacity is denoted by 3x, Caliban's by x. If we give 2a units to each, Caliban will simply waste a units; while Prospero will be still far from completely happy. Surely it will be better to give Prospero 3x and Caliban x. And as the capacity for pleasure varies, it would seem that, if we are to aim

at the production of as much pleasure as possible, we ought to take in hand chiefly those persons who are most capable of happiness, otherwise our trouble will be to some extent wasted.

The truth is, men are no more equal in their capacity for enjoyment than in anything else. And the axiom will have to be accepted in a very special and nonnatural way, if it is to be accepted at all. All are to count for one, unless we can show cause against it. The presumption is to be in favour of equality. But this equality is only in the abstract; when we come to concrete facts we shall have to acknowledge that the pleasure of a bushman is not equally desirable with that of a bishop, quantitatively as well as qualitatively.

But we are entirely without guidance as to what will be the ratio between them. Instead of the delightful simplicity of "each to count for one," we are likely to find ourselves requiring an elaborate schedule of hedonic constants.

A further difficulty of distribution arises, when we remember that "Greatest Happiness of the Greatest Number" may mean greatest total happiness or greatest average happiness. If the former, the end of universalistic hedonism will be to increase population as much as possible. No matter how small the pleasure in the lot of the individual, as long as any pleasure is left at all, we can make the aggregate sum of pleasure indefinitely great by increasing the number of persons who possess it. If, as seems more