Page:American Journal of Sociology Volume 4.djvu/135

 METHODOLOGY OF THE SOCIAL PROBLEM I 19

representation of the different elements, which must be simultaneously or successively perceived, of which the fact consists. This analysis of facts takes place, however, in a definite order of succession, within which in gen- eral three stages may be distinguished. Of these only the first is the neces- sary preparation for all the rest in the sense just indicated. The others may combine with synthetic methods. In this combination they form parts of induction and deduction.

The first of these steps or stages is elementary analysis. It consists of separating a phenomenon into the visible parts composing it, without any concern about the relation of these parts to each other — e. g., the movement of a tree in a gale, into the fluttering of the leaves, the bending of the limbs, and the swaying of the trunk. Such is the analysis which furnishes the uncriticised, or partially organized, raw material of social knowledge which I call descriptive sociology. To be sure, this material is accumulated partly by means of scientific processes in which there is a high degree of refine- ment beyond the primary processes to which I am referring. Relatively, however, accounts of the component activities of associated men, even when they are descriptions in turn of industries, of politics, of domestic, artistic, intellectual, social, religious institutions, are merely elementary analyses. Examinations of the relations between the distinguished parts are not prop- erly begun. This elementary analysis, of whatever degree its precision, serves in the first place merely descriptive purposes. Description consists essentially in pointing out things that e.xist side by side, and one after the other. Such description prepares the way for search into causal relations, although when the subject-matter is ver)- complicated we have to be a long time content with description alone. Causes are too obscure for immediate detection within the described facts. Moreover, this elementary analysis may make use of various sorts of aids. Thus the different senses may in turn be brought to bear on the thing observed. The ears may report the sounds made by the tree and the wind. The nose may report the odors pressed from the leaves, etc. Or the subjective feelings created — fear, awe, pleasure — may be made parts of the analysis and description. All the time the logical character of the process remains one, and even when artificial helps are brought to bear on the facts — field glasses, instruments for measuring the velocity of the wind, etc. — the analysis is still the same. So is it when the description is made up from the testimony of many witnesses, from various historical documents, from statistical tables, etc. Even when the ultimate purpose of analysis is the discovery of causes, so long as causes are not discovered, and the search merely succeeds in making the details of the facts more clear, the process is, after all, only elementary analysis — as I remarked in the case of descriptive sociology.

A case in point is chemical analysis, although it is a process so much more complex psychologically than the process of describing the movements of a tree. The result of qualitative chemical analysis is merely a knowledge