Page:Folk-lore - A Quarterly Review. Volume 25, 1914.djvu/27

 Rh sociology pure and simple. Now I have nothing to do with that. I am not addressing a society of sociologists to-night, but a society of folklorists. But if, as I am inclined to suspect, Dr. Rivers is ready to identify the interest of the folklorist with that of the sociologist, and to lay it down for all those who study survivals of human culture of whatever kind that, until they receive further orders, they are to keep psychological considerations at arm's length, and confine themselves to a purely sociological or exterior view of the movement of history, then it is high time that someone should utter an emphatic protest in the name of all that this Society holds to be most sacred.

Dr. Rivers supports his case by an analogy. Now an analogy, unless it is capable of being turned into homology, is no more than a literary device, and one that is somewhat out of place in a scientific argument. Because two things have some superficial feature in common, you cannot, without assuming some underlying identity between them, go on to expect that in some further respect also the one will turn out to be as the other. Dr. Rivers compares sociology with geology on the strength of the fact that each in some sense,—a very loose sense, to be sure,—is primarily concerned with problems of stratification. Just as the geologist has hitherto usually begun by trying to plot out the actual order observable in the layers composing the earth's crust, so the sociologist. Dr. Rivers maintains, should begin by trying to make out the actual order in which one deposit of culture has been superimposed on another in the formation of social custom. Until the stratigraphical sequences are sufficiently established, he argues, it is more or less futile in the case of geology to attempt explanations in terms of physics and chemistry; whence, by analogy, it must be equally futile to attempt explanations in terms of psychology in the case of sociology. For sociology has not yet succeeded in determining the seriation of past conditions, at any rate whenever a people lacks written records,