§3. Other thinkers of the evolutionist school suggest that the difficulties of Utilitarian method might be avoided, in a way more simple than Mr. Spencer's, by adopting, as the practically ultimate end and criterion of morality, ``health'' or ``efficiency'' of the social organism, instead of happiness. This view is maintained, for instance, in Mr. Leslie Stephen's Science of Ethics; [1] and deserves careful examination. As I understand Mr. Stephen, he means by ``health'' that state of the social organism which tends to its preservation under the conditions of its existence, as they are known or capable of being predicted; and he means the same by ``efficiency'';---since the work for which, in his view, the social organism has to be ``efficient'' is simply the work of living, the function of ``going on''. I say this because ``efficiency'' might be understood to imply some `task of humanity' which the social organism has to execute, beyond the task of merely living; and similarly ``health'' might be taken to mean a state tending to the preservation not of existence merely, but of desirable existence---desirability being interpreted in some non-hedonistic manner: and in this case an examination of either term would lead us again over the ground traversed in the discussion on Ultimate Good (in chap. xiv. of the preceding Book).[2] But I do not understand that any such implications were in Mr. Stephen's mind; and they certainly would not be in harmony with the general drift of his argument. The question, therefore, is whether, if General Happiness be admitted to be the really ultimate end in a system of morality, it is nevertheless reasonable to take Preservation of the social organism as the practically ultimate ``scientific criterion'' of moral rules.
My reasons for answering this question in the negative are two-fold. In the first place I know no adequate grounds for supposing that if we aim exclusively at the preservation of the social organism we shall secure the maximum attainable happiness of its individual members: indeed, so far as I know, of two social states which equally tend to be preserved one may be indefinitely happier than the other. As has been before observed, a large part of the pleasures which cultivated persons value most highly---æsthetic pleasures---are derived from acts and processes that have no material tendency to preserve the individual's life: [3] and the statement remains true if we substitute the social organism for the individual. And I may add that much refined morality is concerned with the prevention of pains which have no demonstrable tendency to the destruction of the individual or of society. Hence, while I quite admit that the maintenance of preservative habits and sentiments is the most indispensable function of utilitarian morality---and perhaps almost its sole function in the earlier stages of moral development, when to live at all was a difficult task for human communities---I do not therefore think it reasonable that we should be content with the mere securing of existence for humanity generally, and should confine our efforts to promoting the increase of this security, instead of seeking to make the secured existence more desirable.
But, secondly, I do not see on what grounds Mr. Stephen holds that the criterion of `tendency to the preservation of the social organism' is necessarily capable of being applied with greater precision than that of `tendency to general happiness', even so far as the two ends are coincident: and that the former ``satisfies the conditions of a scientific criterion''. I should admit that this would probably be the case, if the Sociology that we know were a science actually constructed, and not merely the sketch of a possible future science: but Mr. Stephen has himself told us that sociology at present ``consists of nothing more than a collection of unverified guesses and vague generalisations, disguised under a more or less pretentious apparatus of quasi-scientific terminology''. This language is stronger than I should have ventured to use; but I agree generally with the view that it expresses; and it appears to me difficult for a writer who holds this view to maintain that the conception of ``social health'', regarded as a criterion and standard of right conduct, is in any important degree more ``scientific'' than the conception of ``general happiness''.
Holding this estimate of the present condition of Sociology, I consider that, from the utilitarian point of view, there are equally decisive reasons against the adoption of any such notion as ``development'' of the social organism---instead of mere preservation---as the practically ultimate end and criterion of morality. On the one hand, if by ``development'' is meant an increase in ``efficiency'' or preservative qualities, this notion is only an optimistic specialisation of that just discussed (involving the---I fear---unwarranted assumption that the social organism tends to become continually more efficient); so that no fresh arguments need be urged against it. If, however, something different is meant by development---as (e.g.) a disciple of Mr. Spencer might mean an increase in ``definite coherent heterogeneity'': whether or not such increase was preservative---then I know no scientific grounds for concluding that we shall best promote general happiness by concentrating our efforts on the attainment of this increase. I do not affirm it to be impossible that every increase in the definite coherent heterogeneity of a society of human beings may be accompanied or followed by an increase in the aggregate happiness of the members of the society: but I do not perceive that Mr. Spencer, or any one else, has even attempted to furnish the kind of proof which this proposition requires.[4]
To sum up: I hold that the utilitarian, in the existing state of our knowledge, cannot possibly construct a morality de novo either for man as he is (abstracting his morality), or for man as he ought to be and will be. He must start, speaking broadly, with the existing social order, and the existing morality as a part of that order: and in deciding the question whether any divergence from this code is to be recommended, must consider chiefly the immediate consequences of such divergence, upon a society in which such a code is conceived generally to subsist. No doubt a thoughtful and well-instructed Utilitarian may see dimly a certain way ahead, and his attitude towards existing morality may be to some extent modified by what he sees. He may discern in the future certain evils impending, which can only be effectually warded off by the adoption of new and more stringent views of duty in certain departments: while, on the other band, he may see a prospect of social changes which will render a relaxation of other parts of the moral code expedient or inevitable. But if he keeps within the limits that separate scientific prevision from fanciful Utopian conjecture, the form of society to which his practical conclusions relate will be one varying but little from the actual, with its actually established code of moral rules and customary judgments concerning virtue and vice.