Page:American Journal of Sociology Volume 11.djvu/643

 SOCIOLOGICAL CONSTRUCTION LINES 627

nomena are dependent on the inferred similarity of the experience of different individuals. But it is no more dependent upon it than all " description." " Red " is the name of a subjective experience (referred to an objective cause). Descriptive words like "long" and "true" are as really names of subjective experience as words like " angry " and " afraid." Concepts and propositions exist only in consciousness, and all description, indeed all language, is based upon the supposed similarity of human experience similarity of perception and conception in case of the objective, similarity of conscious states in case of the emotional or appreciative. If it were still objected that the attempt to communicate knowledge of emotions is more liable to misunderstanding than other descrip- tion, because we differ more in our feelings than in our cognitive processes, we could afford to admit, if necessary, that there is a difference of degree; but any difference on this account is only in degree, and may not be even that. Color-blindness that invali- dates the universality and publicity of sense-perception may be as common as any equally wide departure from the normal in the great common human emotions. To assume uniformity of con- ception with reference to the supposedly public, scientific, and purely cognitive, even among experts in description and argu- ment, may occasion serious misunderstanding as often as it does to attribute to men emotional similarity. Indeed, when men are looked at in broad classes in a way to suit the purposes of soci- ology, individual emotional idiosyncrasies may become negligible. It is not necessary that emotions of different individuals should be identical, any more than it is elsewhere essential to the pur- poses of science that specimens of the same species should be identical. The botanist does not despair because specimens of the same variety of plant are not identical, though no two leaves in all June be quite alike. And in interpreting the observable evidence, even of nice emotional differences between individual associates, we have acquired amazing skill that warns us how far to trust our inference of the emotional similarity of man, by virtue of which the motives and emotions that characterize social classes become public, as revealed in overt signs. The metaphysicians do not claim that we are especially liable to error when we take the