Page:American Journal of Sociology Volume 15.djvu/736

 722 THE AMERICAN JOURNAL OF SOCIOLOGY

conduct. The marking-scale must not be arbitrary, a mere in- genious invention. It must be a simple and natural expression of observed relations. If possible, its elements should be derived from those common-sense gradings which the world uses in rough and ready fashion for the purposes of everyday life, and which therefore have the warrant of experience.

A familiar example of empirical gradings is the scale of con- sanguinity. The children of brothers or sisters we call first cousins, and describe them as one degree less nearly related than brothers or sisters. Children of first cousins we call second cousins, and describe them as two degrees less nearly related than brothers or sisters, and so on. These degrees, as everybody recognizes, are nothing more than successive positions in a scheme. No absolute numerical value, or distance from zero, attaches to any one of them. A' kindred but less familiar ex- ample is offered in Lewis H. Morgan's ordinal arrangement of the successively wider and wider groupings composing tribal society. The arrangement is, namely : first grouping, the family ; second grouping, the totem kin or clan; third grouping, the phratry or brotherhood of clans ; fourth grouping, the tribe ; fifth grouping, the federation of tribes.

One intent of this paper is to call attention to other rough marking-scales which we daily make use of in describing social phenomena, and to show how they may be extended and made sufficiently precise for scientific purposes. A second intent is to indicate how, by distributing a considerable mass of existing numerical data in accordance with the marking-scales to be described, we may, through the usual methods of statistical analysis, arrive at sociological conclusions that are interesting and possibly important.

To a great extent the phenomena of population are described by statistical measures from zero. Density, excess of births over deaths, emigration and immigration, are expressed in absolute numbers, of approximate accuracy. But such measures are not forthcoming when we ask a question like the following, namely. Is a population made up of native white Americans, foreign-born Irishmen and Germans, foreign-born Italians, Hungarians and