Several languages, including Amharic (Semitic; Leslau 1995), Engenni (Kwa; Thomas 1978) and possibly Aghem (Bantu; Hyman 1979) display a curious pattern in which an indexical pronoun such as ‘I’ or ‘you’ can refer to the speaker or hearer of a reported speech act rather than of the actual discourse (see also Hale & Platero 1996 and Speas 1999 for related facts in Navajo). The following is an ex...

The data were collected from various hospitals and nursing homes in Durban, Pietermaritzburg, and Capetown, as shown below: 3,165 European babies from two private nursing homes in Durban. 1,057 coloured (i.e. half-caste)t babies from a mission hospital in Durban, and two provincial hospitals, one in Capetown and one in Pietermaritzburg. 2,188 Bantu babies from a mission hospital in Durban. 1,39...

This paper describes a proof-of-the-principle experiment in which maximum entropy learning is used for the automatic induction of shallow morphological features for the resourcescarce Bantu language of Gı̃kũyũ. This novel approach circumvents the limitations of typical unsupervised morphological induction methods that employ minimum-edit distance metrics to establish morphological similarity bet...

We describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which includes data from nine Southern Bantu languages. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and we report on our investigation of the stability of the ASR models derived from the corpus. We also repor...