30 Presidential Address.
are ever to get at grips with such a question if psycho- logical considerations are altogether ruled out on a priori grounds of method. Gomme at any rate was not such a pedant as to reject a useful hint, though it come from any quarter. Nor does Dr. Rivers show himself pedantic^ inasmuch as he has passed on from sociology to ethnology, and from ethnology to psychology, with a progressive enlargement of outlook which makes his book a classic for all those who wish to study method in the making.
Dr. Rivers, indeed, allows in so many words that " there is one department of sociology in which . . . psychological assumptions become indispensable," namely, when the pur- pose is " to show how social institutions come into existence as the result of the contact and blending of peoples." - Such assumptions, however, he insists, are not to be treated as " laws." They must be tested by the study of social processes ere ever we can so regard them.-^ With this we must all agree. After all, as folklorists and anthropologists, we are not interested in psychology or sociology as such, but in the science of human culture, a far more concrete and comprehensive study, which makes use of these dis- ciplines, and of others as well, just in so far as they throw light on the subject of culture from this side or from that. Or again, we are not interested as ethnologists in the history of any particular culture-area in itself A so-called " law " is no law, a demonstration of tendency is not a real demonstration, so long as it holds good only for the British Isles, or for Melanesia. Our science is concerned with the general conditions of culture-contact ; and to this end, and to nothing short of it, must our sociological and psycho- logical studies be conjointly directed.
Dr. Rivers is, of course, fully aware of this. Indeed,
though his treatise on the history of Melanesian society
has primarily an ethnographical scope, he has managed, in
a 'iQw pregnant pages, to formulate such general conditions
"~ Sociological Rcvieu\\^. (\fj\()),'i. "^ lb. g.
