510 THE AMERICAN JOURNAL OF SOCIOLOGY
Classification of human associations must therefore make a place for genetic, for morphological, and for functional catego- ries, and the morphological and functional phases of the scheme must evidently at last fall into a genetic series parallel with the evolution of the associations themselves. The morphological and the functional categories in turn will have to be worked out by a process of adjusting perceptions of workings to conceptions of ends toward which the workings tend.
It might be held that such classifications of societary activi- ties as those of Spencer, Schaeffle, and De Greef are functional, and that a functional criterion of activities is implicitly teleologi- cal. The reply is that, whatever may be the logical implications and correlations of the notion " function," it by no means in actual use draws with it a working notion of end. Indeed, it is more than probable that sociological criticism will at last fix upon the absence of an adequate concept of the end, or the " universal," indicated in the visible parts of the social process, as the factor which limits the availability of the functional classi- fications of which those named are types. The best of them, such as that of Schaeffle, appear to be but unorganized cata- logues of activities after all, so soon as we confront the final task of correlating the activities analyzed and scheduled. We dis- cover incidentally that the idea of function treated as an abstraction, and held apart by itself, is as sterile as any other purely formal conception. The idea of a " sustaining system," a " distributing system," and a " regulating system," for instance, is of itself something as far from touch with the realities of human association as the vulgar idea of "perpetual motion " is from embodiment of the realities of mechanics. To make terms of mechanism real it must be mechanism that functions in a real process. We must have companion-conceptions, in the case cited, first of what is sustained, what is distributed, what is con- trolled ; and, second, of why the sustaining, distributing, and controlling, i. e., of the end to which the mechanism and the process are incidental. This is the setting in which we find most use for the formula to which we have constant recourse, viz.: Society, or human association, must be considered as a
