198 THE AMERICAN JOURNAL OF SOCIOLOGY
We cannot take the individual as our unit unless we rob anthropology of its unit. Only a part of man the spiritual part is molded by association. He gets hungry, tired, or sleepy as a man, not as a socius. Many of his instincts, cravings, and thoughts are pre-social, or, if you prefer, extra-social. Like the walls of old castles that have weathered into oneness with the cliff, the socialized part of us is so weathered that you can hardly tell where it leaves off and temperament or individu- ality begins. It is certain, nevertheless, we cannot reduce the whole man to a "cell" in a "social organism." Some of us do get printed with the full design of our time and tribe. But most of us take the pattern only in spots, and there are, moreover, eccentrics and recalcitrants who utterly refuse to be drawn in between the social rollers.
Nor can we take as our unit the "social organ," meaning thereby the functional group. So long as division of labor was regarded as the leading feature of society, it was natural to be chiefly interested in the co-ordinated groups of workers, fighters, or directors. But it has come to be perceived that there are many groups which can in no sense be said to fulfil in society an office analogous to that of an organ in a living body. Along- side of their functional groupings, men are found associated into guilds, corporations and parties, bound together by a community of aims, and striving each to gain an advantage at the expense of the rest. Nor is this all. Besides these interest groups, we recognize in classes, castes, and sects likeness groups, held together by the consciousness of kind. Beyond them we may distinguish natural groups, such as family and neighborhood, and fortuitous groups, such as crowd or public.
In truth, people are ever clasping and unclasping hands, uniting now for a day, now for life. Could we run history through a kinetoscope, we should see groups forming, dissolv- ing, and re-forming, like the figures of dancers on the floor of a ball-room. What, then, is more natural than to conclude: "The group is the true unit of investigation in sociology"?
Now, whoever will acquaint us with the genesis, development, and maintenance of all kinds of groups will lead us far, very far,
