ABSTRACT
Ever since Morgan, the study of kinship terminologies by anthropologists has been principally concerned with principles of social classification. The ethnographer observes that a set of individuals, Al, A2, A3 … An are all lumped together by speaker X as belonging to a single category A, and the argument then centres upon the principles of logic which could lead to this classification. The emphasis is almost entirely algebraic rather than linguistic. The words employed are usually reported but then quickly laid on one side. It is assumed (without any obvious justification) that the prime meanings of the terms in question are genealogical referents and that any supplementary ‘meaning’ which is not genealogically derived is a metaphorical extension of an ‘original’ kinship meaning. With rare exceptions, such as Radcliffe-Brown's famous observation that the Ba Thonga address their mother's brother as ‘male mother’ (Radcliffe-Brown 1923), there has been very little attempt by professional anthropologists to analyse the constituent elements of kinship language as such.
