Abstract

Studies have shown that passivization in Bantu languages is effected by means of a derivational suffix attached to the verb. This passivization pattern is residual and no longer productive in Lunda in that only a very small set of verbs allows the passive morphological suffix. It has been replaced by a new type of a passive construction which at face value appears as an impersonal construction because of the presence of the nonreferential class 2 third person plural subject prefix and the non-promotion of the non-agent to subjecthood in a clause. In this paper, I will show that the Lunda construction meets some of the characteristics of a canonical passive and that passivization also serves as one of the tests for transitivity despite the failure of the non-agent to become the subject of the clause.