ABSTRACT
This paper is mainly concerned with the meaning of the word «kinship» 1 in anthropological usage. I seek to show that the word is notably imprecise and ambiguous, and therefore obscures and confuses some important distinctions; and I go on to suggest a terminology by means of which the distinctions could be kept clear. In order to anchor this potentially abstruse discussion in empirical reality, I begin by considering the meanings of kinship terms among the Baganda. My difficulties with kinship in fact began with my doubting whether some of these terms really meant what they had to mean if they were kinship terms; and these doubts became systematic after I read Leach's paper on Trobriand terminology (Leach 1958).
