Pronoun behaviour in Ekegusii, a Bantu language of south-west Kenya

Ekegusii is a member of the Bantu family of Niger-Congo languages and is spoken in the Kisii County of south-west Kenya. It is generally regarded as under-researched, with only a small number of linguistic papers written on it, including a basic grammar and some studies of its tonology. This talk will discuss morphosyntactic data collected on a fieldwork project I undertook in July – August last summer, focussing on developing a hypothesis to account for some puzzling pronoun behaviour. Like most Bantu languages, Ekegusii allows pro-drop and has a relatively flexible word order. From the data collected, I hypothesise that there are two different types of overt pronouns, appearing in different syntactic configurations - one type (type A) appears in initial position (1-2), whereas the other (type B) appears after the verb (3-4). In some instances, both pronouns can be used grammatically (e.g. (5-6)); however, this is not always the case (cf (7)). This talk will discuss hypotheses about the distributional criteria conditioning this surprising morphosyntactic difference, considering work about information structure such as Mira Ariel’s Accessibility Theory (e.g. Ariel, 2001) as well as making comparisons with other Bantu languages. There will be also the chance to discuss the fieldwork process in general and my own personal experience. (1) inche nkorokwa Elisabeth 1PS.SG.A ‘I-am-called’ ‘Elisabeth’ I am called Elisabeth. (2) *nkorokwa inche Elisabeth ‘I-am-called’ 1PS.SG.A ‘Elisabeth’ (3) nkomanya ore bwango ‘you-learn’ 2PS.SG.B ‘quickly’ You learn quickly. (4) *ore nkomanya bwango 2PS.SG.B ‘you-learn’ ‘quickly’ (5) inche noyo Gordon 1PS.SG.A ‘belong’ Gordon My father is Gordon (lit. “I belong to Gordon”) (6) noyo Gordon inde ‘belong’ ‘Gordon’ 1PS.SG.B (7) *nkorokwa Elisabeth inde ‘I-am-called’ ‘Elisabeth’ 1PS.SG.B