COMPARATIVE STUDY OF ASSOCIATION 667
solely to the descriptive and analytic consideration of the char- acteristics of some one phase of the entire phenomenal complex. It is unnecessary to discuss here the relative vulnerability of these two points of attack or to compare at length the merits of the different methods of approach appropriate to each. It is probable that each point of attack is of high significance, and that each method is destined to play a part of great import to the general development of the science of sociology.
Were this a comparative study of sociologies, rather than of the subject-matter of sociology, it would be of interest at this point to make an analysis of the works of the various writers upon subjects social, to the end of determining the motives, par- ticular phases treated, points of attack, methods of procedure, and substantial results of each. After this preliminary step had been taken and each work had been, as it were, "reduced to its elements," it would then be in order to make the further attempt, through a still more searching analysis and comparison, to trace back all these elements to certain fundamental factors, and, having found these factors, to exhibit synthetically the coordination and correlation of the whole about them as it has actually occurred thus far in the process of growth. To be sure, the relation to the entire age environment would have to be kept constantly in view at each step, but when the whole task had been completed, it would be surprising, indeed, were it not discovered that in large degree these different works form parts — widely sundered, pos- sibly — -of a developing unity whose symmetries, naturally, can be but vaguely foreshadowed in the early evolution of a com- paratively small number of its constituent factors. Such a com- prehensive survey of the field, however, would of itself require many pages, and must, therefore, be left to another time and place. It will answer our purpose here to anticipate one of its possible results in calling attention to a trait which, in varying degree, seems to be characteristic of the majority of the works thus far appearing under titles which may be classed as socio- logical.
Influenced partly by the prevailing trend of thought, which is wont to consider each fact as a term in a genetic series,
