46 THE AMERICAN JOURNAL OF SOCIOLOGY
hand, or to some form of evolutionism on the other. Whether this is a chance world, an arbitrary world, an orderly world, a mechanical world, a spiritual world, or a combination of these and other conceivable types of world, is a question to which every sociology, if it is formally complete, must assume an answer. Sociology cannot furnish an answer of itself, but it is a trun- cated structure of thought, or, worse, a vapor hurtled by the winds, if it does not fit into presumptive answers to these questions.
Sociology cannot be a system of deductions from the assumptions which it may adopt about antecedent relations, nor will it utterly fall with the failure of those assumptions, but it will inevitably be of a piece with some system, or un-system, of organizing presuppositions. Accordingly, sociologists like Comte have had a fundamental philosophy that in terms rejected not only a theistic but a spiritual conception of the universe. They have nevertheless been recognized as helpful in developing certain parts of sociology by men who as strongly contended for both spiritualism and theism. Again, men have held a fundamental philosophy which excluded the evolutionary idea. They have nevertheless developed sociologies which had a high degree of coherence, whatever might have been their degree of plausibility.
It does not follow that sociology may be equally true and complete, whether attached to a credible or to an incredible philosophy. The bearing of what has been said is that soci- ology may not have reached a stage in which it can demonstrate that one theory of reality as a whole is compatible and another incompatible with the system of relations which it is the special task of sociology to formulate. Meanwhile it is a necessity of complete mental action to have a place for a theory of the things that are fundamental to the social relations. Otherwise there is no sufficient check upon the play of our imagination in constructing an artificial world out of that social material which it is our task to observe and describe and analyze and proxi- mately interpret. It is a veto upon hasty conclusions in sociology to be obliged to make them fit into some general
