REVIEWS 841
which are derived the rule of exogamy and also the peculiar avoidances between relatives required by customary law among savages. Although purporting to be an explanation of social origins, the work is really little more than a conjecture how exogamy and avoidance originated.
Mr. Atkinson's treatment of the subject is not clear, and even with the help of Mr. Lang's annotations it is sometimes difficult to follow the course of his argument. He died before he was able to give his manuscript final revision, and to this may be attributed much of the vagueness which appears. Mr. Atkinson did not claim to have estab- lished his theory, but presented it as a plausible hypothesis which he believed further research would verify. The chief merit of the essay is that it is an attempt to fill a serious gap now existing in the theory of development; its chief defect is an inadequate conception of the problem, which defect is common to most speculations based upon purely anthropological data.
This point must be made clear in order to show the sociological importance of the matter, which is now so obscured that works of this character are apt to be regarded as having merely an antiquarian value of no special interest to the sociologist. It should be borne in mind that the theory of the descent of man, as formulated by Darwin and worked out in detail by Haeckel and 'others, accounts for man as an animal and not as a social being. Darwin, with his habitual candor, pointed out this gap in the theory. He ascribed to natural selection the formation of the human species, and to sexual selection the forma- tion of the subspecies usually designated as the different races of man- kind. But he admitted (sec. 214) that the development of social and moral qualities is not accounted for by natural selection, and he was careful to say (sec. 1006) that in order to obtain from sexual selection the effects he attributes to it, its operation must be referred to a pre- social period, when man was on a par with other animals in habits of associations. The difficulties in the way of any theory tracing society as an outgrowth of such gregariousness as is found among species nearest to man in physical structure, are so great that Lester F. Ward 1 holds "that man is not naturally a social being, that he has descended from an animal that was not even gregarious by instinct." Sociologists, as a rule, regard society as being the product of human intelligence, accounting for it through the operation of sympathy, imitation, and other emotional states, together with rational appreciation of the advantages of social life. But this only brings up the problem of
1 Outlines of Sociology, p. 90.
