The Accusative Case parameter

Tanya Reinhart and Eric Reuland

Project description / Abstract

A central question in Case theory since the eighties has been the relation between Case and the thematic system. Purely structural Case (Nominative, Accusative) was contrasted with Case dependent on the verb's thematic structure (e.g. Instrumental and Oblique). Structural Case was determined by syntactic relations such as government, or later Spec-head/checking, while inherent Case was determined thematically (Chomsky 1981, 1986). The current project investigates a novel perspective on the relations of Case and Theta. It is not that the Cases themselves are divided into structural and thematic, but rather all Cases have these two components. Case thus encodes two different relations: that of a θ-argument, and that of a syntactic complement. This requires finding and investigating languages and contexts differentiating between the two relations. While the thematic component is expected to be universal (being the implementation of the θ-criterion), structural Case may well be parameterized. The projects envisages three main testing grounds: i) exceptional case marking; ii) internal theta-role reduction (Reinhart 2003, Reinhart & Siloni 2003); iii) HAVE versus BE auxiliary selection. The investigation will focus on cross-linguistic correlations between the various options in this domain.