796 THE AMERICAN JOURNAL OF SOCIOLOGY
men in association the datum that discontinuity of some sort and some degree is universal among men in association. He sets this datum down in the list of things that must be known more completely in all its bearings upon the actions of men in contact with each other.
So we might go through a list which we may name "incidents " of association. They are data of sociology : deposits of much observation of the world of people from many points of view, but raw material with which we begin a study of men from the point of view of the sociologist, i. e., when we want to correlate all that we can learn about the world of people into accounts of the laws of human association in general. In other words, there are larger truths in the laws of human associa- tion than emerge when we study in turn particular kinds of association. Those studies of particular kinds of association are incomplete, therefore, until they are merged into knowledge of these larger truths. The task of finding out precisely what these larger truths are and how they are related to each other furnishes the primary problems of sociology.
Our survey up to this point suffices to sharpen a simple per- ception which must presently afford much needed light on soci- ology. There has been endless perplexity among sociologists about the concept "society." It has been asserted, on the one hand, that if there is to be a science of society, there must first be a definition of society. By others it has been urged with equal confidence that the definition of society must of necessity be a product of a science of society, and cannot be had until the science is relatively complete. There is an element of truth in both these contentions, and both may be urged with somewhat similar force in connection with the reality "association."
The perception that should resolve the difficulty, however, is that the universal fact of association in the world of people is not to be taken as a closed concept, containing consequences to be drawn out by deduction as a system of sociology.' The fact of association is rather an open world to be inductively described and explained. It is a fact of indefinitely varied forms, kinds, degrees, extents. Wherever there are two men there is association.
