Subject and Object transpositions in Kiswahili Bantu

Whiteley (1968: 10) states that, "It is the property of items participating in an object-relationship that they may also participate in a subject-relationship, and one way of exposing differences of transitivity is to transpose the item(s) in the object-relationship with those in the subject-relationship while retaining the same lexical items." He calls the operation "entailment" and the predications "an affiliation-set". Modern linguists call it an inversion operation. (1-2) are operations of entailment that define an affiliation-set: 1) M-gema a-li-pand-a mnazi-ni 'the palm wine tapster has climbed up the coconut tree'; 2) Mnazi-ni ku-li-pand-a m-gema 'up the coconut tree climbed a/the palm wine tapster'. (3-4) are ungrammatical constructions: *3) M-gema a-li-pand-a 'the palm wine tapster has climbed'; *4) Mnazi-ni ku-li-pand-a 'up the coconut tree climbed'. (3-4) reveal that the predicate verb -panda is a transitive verb that requires an obligatory complement for its clause to be grammatical. The lecture then introduces two new distinctions. Namely, it points out that transposition implies either symmetric entailment in which the transposed items occupy the exact positions vacated by each other or asymmetric entailment in which at least one transposed constituent does not occupy the exact position vacated by the other. (1-2) involve symmetric entailment. The lecture further focuses on what happens to the demoted subject mgema 'tapster'. It is not interested in what happens to the promoted object mnazini 'in the coconut tree'. We argue that since mgema 'tapster' in (1) is the agent and it goes to occupy the object position vacated by mnazini in (2), it must be an object AGENT since its omission would render (2) ungrammatical, as in (4), just like the omission of mnazini in (1) would render it ungrammatical, as in (3). The lecture concludes by noting that there are many subject-object transpositions in Kiswahili Bantu that have been recorded since the beginning of its grammatical descriptions but they have been discarded in modern grammatical works due to the influence of so-called modern linguistic theories like Chomsky (1981), Baker (1988), Wechsler (1995: 5), and others. Modern theories claim that in no language of the world do agents grammaticalize as objects. Thus Weschler (1995: 5) says 5) 'Cooper ate the doughnut' will not be realized as 6) 'The doughnut ate Cooper' with Cooper in (6) as object and AGENT in any language. In Kiswahili Bantu, these patterns exist. Johnson (1939) and Sacleux (1939) record 7) N-jia zi-na-zo-pit-a wa-tu 'lit. roads that are passing people' as oppsed to 8) Wa-tu wa-na-o-pit-a n-jia 'people who are passing roads'. We conclude that, as Bantuists, we need to place emphasis more and more on the internal idiom of Kiswahili Bantu rather than discard them in favour of Indo-European grammatical theories that have no relevance for the language. If we fail to do so, the valuable Bantu idiomatic structures of the language will be lost for good and its grammar impoverised.