Abstract

In recent fieldwork in Uganda, we found that the Bantu language Luganda displays not just one, but three different morphosyntactic strategies to express focus. First, there is a positional effect for wh questions and corresponding answers to appear Immediately After the Verb, as is familiar from other Bantu languages (Watters 1979, Buell 2006, van der Wal 2009). Second, constituents can be placed in a preverbal position with an agreeing focus morpheme (gye) preceding the verb, as in (1). Third, nouns can appear with or without an initial vowel, where the absence of this vowel indicates focus (2).

(1) M-mése o-mu-sóta gy-e gw-a-kuttê.

9PX-rat    3A-3PX-snake    9-FOC    3SM-PAST-catch.PERF

‘It’s a rat that the snake caught.’

(2) Y-a-yéra o-lú-ggya / lú-ggyâ.

1SM-PAST-sweep    11A-11PX-yard    / 11PX-yard

‘She swept the yard / the YARD.’

This perhaps surprising abundance of strategies to encode focus triggers the following initial research questions: 1) what is the difference in semantic/pragmatic interpretation of each strategy? do they encode they same type of focus? 2) how can we tell? what are the diagnostics for focus? In this talk I apply various tests for different types of focus, in order to support an analysis where the preverbal focus construction expresses identificational focus, and the absence of the initial vowel is shown to express exclusive focus (refining Hyman & Katamba 1993). This not only provides us with a more detailed analysis of the focus constructions in Luganda, but also explains the interaction between the different strategies, and gives insights into the interface between morphosyntax and information structure.

This research was conducted in a small project funded by the Alborada Fund and carried out together with Saudah Namyalo.