Hirotoshi Hiwaki

Did patterns of endogamy exist in Roman kin groups? This issue has recently been raised in the context of arguments relating to Roman social structure. J. Goody suggests that the Christian Church in the western Roman empire fundamentally altered the structure of the Roman family by restricting close-kin marriage, adoption, remarriage, and concubinage. Contrary to Goody's thesis, R.P.Saller and B.D.Shaw insist that parallel and cross-cousin marriages were rare among aristocrats, and parallel-cousin marriages rare among the mob during the first three centuries after Christ. Consequently, the Christian ban on close-kin marriage had little impact on society. In addition, they point out that Roman dowries were relatively modest and aristocrats normally owned landed property dispersedly, so there was no reason to develop marriage strategies like close-kin marriage to protect the unity of the landed property of a family. Even today, however, scholars have not come to much agreement concerning this issue mainly due to a shortage of related historical materials. This article first attempts to investigate the theories regarding this issue, and comes to the following conclusions. (1)Those who look on Roman society as endogamous can not find strong evidence for their argument. (2)Similarly, those who look on Roman society as exogamous do not have a leg to stand on, especially Saller and Shaw who misinterpret their data based on an examination of funerary inscriptions, which lead them to the wrong conclusion. The author concludes from the above that ancient Rome was an "in-marriage" society from the Late Roman Republic to the establishment of Christianity as the state religion. To these conclusions, the following points have been added. (3)The explanation for the enactment of imperial legislation on close-kin marriage did not lie only in the realm of Christian ideology of family and sexuality. These laws did reflect real and current practice of close-kin marriage among the inhabitants in the western Roman empire. (4)There is strong evidence for the practice of endogamy in female marriage. In such cases, close-kin marriage makes sense as a strategy to prevent the fragmentation of family estates.

R. P. Sailer recently criticized lexica failing to indicate what circle of relatives is included in their definition of "family" for familia or domus. According to his investigation, familia chiefly included agnates joined by a male line. Domus, on the other hand, included both agnates and cognates connected by a female linking relative. Accepting Sailer's suggestion, Makoto Shimada attempted to explain the change of family type in Roman society. He advanced a new theory that the Roman elite shifted from an emphasis on the agnaticfamilia to an emphasis on the cognatic domus in 90/ 89B.C. In this paper, by examining the meanings of familia and domus, the author attempts to answer the following two questions : first, when and how did Roman kin groups change, and, what type of kin groups appeared in Roman society. The results of the examination are as follows. 1)Contrary to the suggestion of Sailer, there is strong evidence for a parallel structure between familia and domus after 200 B.C. Along with the decline of patriarchy after 200 B.C., the category of familia also changed from patrilineal to quasi-bilateral and hence, included cognatic descent, as did domus after the first century B.C. Consequently, the date when Roman kin groups changed was about 200 B.C. and the structure of new kin groups were quasi-bilateral. 2)Around 200 B.C., marriage patterns changed drastically(i.e., the spread of sine manu, divorce and close-kin marriage). This change was the critical point of substantiation of changes in kin groups in that peried. 3)A quasi-bilateral structure was suitable for the political reality, but it also had a defect in that it was hard to maintain unity in the kin group. The spread of close-kin marriage could be interpreted as part of an effort to maintain the unity of the kin group.