132 THE AMERICAN JOURNAL OF SOCIOLOGY
biology and psychology. The sociologist must depend upon the biologist for instruction about the borderland between the world of things and the world of people. The sociologist must begin to know how to discover for himself among the phases of reality which appear to belong distinctly in the world of people. Hence he must be at least an initiate, if not an expert, in the methods of psychological analysis.
Now, in order to give another indication that all this very remote preliminary discussion has direct bearings upon every general and special problem of social science, the whole course, from this point, will be made to rally around a single question. It includes innumer- able lesser questions that might be proposed in making a catalogue of unexplored social relationships. Let us suppose, for the sake of simplicity, that the social problem is concerned with this one question only, instead of with the thousand and one which are involved, viz.: What factors prevent, permit, or promote variations in types of correlation among people, and what are the formulas of these influences ?
It will be seen that this question calls for research among social correlations of all orders. What influences preserve or destroy the corre- lation which any family type presents ? Or we may propose the same form of question with respect to international alliances or to the present social equilibrium in any country.
It will be obvious, too, that we must have a working concept corre- sponding with the expression " variation in types of correlation among people." I mean by this phrase any change which results in a new ratio of valuation or in a new form of functional importance among the individ- uals affected.
The question proposed is a concrete expression of the proposition that, in presuming to study " the social problem," we come under the necessity of finding out what are the most general influences that operate in the world of people. It may be worth while to point out that the terms of the question suggest at once the categories which will be needed later ; viz., the category of social order, in which the statical laws prevail ("prevent"); and the category of social progress, in which the dynamic laws prevail ("permit or promote ").
It is clear, too, that the answer to our question must involve at last a synthesis of answers to the same question in its application to different life spheres. For instance, comparative constitutional his- tory collects cases of variations in structure of the governmental system, economic history cases of variations in structure of the industrial system,
