Agnatic or Cognatic? A Re-Evaluation of Ondo Descent

Abstract

Both the Ijebu and Ondo kingdoms have been described by Lloyd as having a cognatic descent system (I962; I966; I968). The cognatic descent groups of these southern Yoruba peoples,' he has indicated, contrast with the agnatic descent groups characteristic of the more northerly Yoruba. This has been said to be one of the ways in which the Yoruba differ significantly among themselves in some aspects of social structure while being relatively homogeneous culturally. Lloyd has attempted to discover other differences in social and political structure which appear to be correlated with this contrast in descent organisation. It is more accurate, I feel, to describe the Ondo descent system as agnatic, and I will indicate here my basis for making this judgement. I must emphasise first, however, that differences between Lloyd and myself are not the result of our working from different definitional assumptions, since we seem to be in general agreement here. Thus the basis for our opposed conclusions depends on the nature of the data collected and on our interpretation of it. I agree with Lloyd's (I966) expressed definitions. Cognatic descent groups, he states, are ancestor-oriented, with descent being traced through ascendants of either sex; while agnatic descent groups are ancestor-oriented, with descent being traced through male ascendants only. As he points out, both types of descent system are structurally favourable to the formation of corporate groupings. Cognatic descent is unrestricted in the sense that an individual can theoretically be a member of as many groups as are formed on the basis of the descendants of the original ancestor. Agnatic descent groups, by contrast, are restricted in the sense that an individual can belong only to groups composed of persons descended from a common ancestor through male links. In distinguishing between agnatic and cognatic descent groups, only one definitional criterion is used by Lloyd: the rules for membership in these groups. Any other social phenomena which are frequently considered when discussing descent groups, such as residence rules, jural inheritance and concepts of biological relatedness have, correctly, been excluded from his definitions, since they can vary independently of the type of descent.