5 1 6 THE AMERICAN JOURNAL OF SOCIOLOGY
other hand, it cannot be asserted too emphatically that neither anthropologist nor ethnologist does or can classify associations when he classifies peculiarities of individuals in association. The associations are diversified by physiological and racial traits, but these do not exhaust the qualities of the associations.
Not to involve ourselves in discussions of anthropological technicalities, let us assume that the anthropologists have agreed upon a system of classification which carries out the general zoological scheme. Let us suppose that they have divided the human race into types designated as A, B, C, and D, Suppose that these types include classes indicated by the symbols a, b, c, and d. Suppose that orders of each class are represented by the terms a 1 , 6 l , c 1 , d 1 . Suppose that subspecies, varieties, etc., are clearly made out and designated by a", ", a , d" ; a, 6, c, d, etc., etc. Now, our point is that when the anthropo-sociologist has applied these categories to a given population, as Demolins has done to the population of France, 1 or as Lapouge has indicated an ambition to do in a comparative way to populations in general, 2 he has not classified the state. He has merely classified certain details of more or less importance abstracted from the whole body of conditions and activities which constitute the state.
The same propositions hold true of the ethnologists. With- out carrying out our illustration in detail in their case, let us suppose a hierarchy parallel with the foregoing, designated by the symbols E, F, G, H ; e,f,g, h; e { , /',*, ft, etc. These, as before, are symbols of abstractions from associations. They are not subdivisions of real associations.
The sociologist, on the other hand, does and should try to classify societies. He must decide for himself which groups of anthropological and ethnological results are of value as data for sociological elaboration.
An approach to formal indication of the sorts of judg- ments likely to be passed may be made by considering the relation of anthropological and ethnological categories to eco- nomic classification, i. e. t classification and analysis of economic states, and of all states subsequent to that phase considered
1 Les Frartfais d 'aujoura' 'Aui. * Vid. LAPOUGE, L'Aryen.
