focus marking in bura: semantic uniformity matches syntactic heterogeneity

the present article introduces a theory of (morpho-)syntactic focus marking on nominal categories in bura, a central chadic svo language spoken in the northeast of nigeria. our central claim is that the particle an plays a crucial role in the marking of subject and non-subject focus. we put forward a uniform analysis of an as a focus copula that selects for syntactic predicates of type <e,t> and a focused constituent of type <e>. this uniform semantic representation is transparently mapped onto different syntactic structures: in a clause with a focused subject, the focus copula appears between the subject in spectp and the predicative vp. on the other hand, syntactically focused non-subjects are fronted and appear in a bi-clausal cleft structure that contains the focus copula and a relative cleft-remnant. the non-uniform analysis of focus marking is further supported by the structure of predicative constructions, in which the focus copula separates the focused subject and the adjectival or nominal predicate. it is also shown that alternative unified analyses fail to account for the full range of bura data. the latter part of the article provides an analysis of the bura cleft construction. based on syntactic and semantic evidence, we come to the conclusion that the clefted constituent is base-generated in its initial surface position, and that an empty operator moves within the relative clause. the article concludes with a brief discussion of the potential conceptual reasons behind the observed subject/non-subject asymmetry in bura.