THE AMERICAN JOURNAL OF SOCIOLOGY
occasion to press back beyond the evident fact that human associations have exhibited these distinct phases. We need not even attempt to formulate the order in which the different motives have predominated. It is certainly not a constant order, though the variations from the order given may be excep- tions easily explained.
In proposing these general divisions of social conditions we have had in mind the activities of the human race as a whole, though it is evident that neither periods of civilization nor geographical limits could be designated in which these distinc- tions would precisely classify the whole human family. If soci- ological theory is ever complete, its generalizations must cover the largest circuit of actual association. Up to date, however, the conditions of human contacts have been such that the exten- sion and intention of the associational process have to be learned chiefly by observation of associations ranging from the family at one extreme to the state at the other. It is by no means certain whether the most fruitful method of settling upon a mor- phological and functional and teleological system of classifying associations will be found by those who start with the family and move outward through association after association until they reach the state, or by those who begin with the state and analyze down through association after association until they reach the family, and thence the individual. Perhaps there would be more superficial consistency with our introduc- tory references to the individual assumption (chap, iv, pp. 60-62; chap, v, pp. 17799) if we adopted the former course. For reasons which need not be detailed, however, we find the latter alternative more immediately promising, though by no means necessarily likely to acomplish so much that it will be final without verification and expansion through independent elaboration of the opposite method. We therefore propose a system which may be called a basis for national sociology. The theory so called will be related to the more inclusive sociology very much as the orthodox political economy, or, as the Ger- mans have more appropriately phrased it, " national economy," has been related to the more general theory of utilities striven
