Page:American Journal of Sociology Volume 11.djvu/31

This page needs to be proofread.

STUDIES IN EUGENICS I 5

earliest, and perhaps the most successful, of all attempts to deci- pher prehistoric customs by means of those now existing among barbarians, and by the marks they have left on the traditional practices of civilized nations, including ourselves. Before his time those customs were regarded as foolish, and fitted only for antiquarian trifling. In small fighting communities of barbari- ans, daughters are a burden ; they are usually killed while infants, so there are few women to be found in a tribe who were born in it. It may sometimes happen that the community has been recently formed by warriors who have brought no women, and who, like the Romans in the old story, can supply themselves only by capturing those of neighboring tribes. The custom of capture grows; it becomes glorified because each wife is a living trophy of the captor's heroism; so marriage within the tribe comes to be considered an unmanly, and at last a shameful, act. The modern instances of this among barbarians are very numerous.

4. Australian marriages. The following is a brief clue, and apparently a true one, to the complicated marriage restrictions among Australian bushmen, which are enforced by the penalty of death, and which seem to be partly endogamous in origin and partly otherwise. The example is typical of those of many other tribes that differ in detail.

A and B are two tribal classes; i and 2 are two other and independent divisions of the tribe (which are probably by totems). Any person taken at random is equally likely to have either letter or either numeral, and his or her numeral and letter are well known to all the community. Hence the members of the tribe are subclassed into four subdivisions: Ai, A2, Bi, B2. The rule is that a man may marry those women only whose letter and numeral are both different from his own. Thus, Ai can marry only B2, the other three subdivisions, Ai, A2, and Bi, being absolutely barred to him. As to the children, there is a difference of practice in different parts : in the cases most often described, the child takes its father's letter and its mother's numeral, which determines class by paternal descent. In other cases the arrangement runs in the contrary way, or by maternal descent.