| Case is described in terms of its function and meaning, i.e. syntax and semantics. Theorists, however, have tried to reduce case to either syntactic or semantic phenomena. Given that case has both function and meaning, the null hypothesis is that case is both syntactic and semantic. In this dissertation, I explore the syntactic and semantic bases of case assignment. The key idea is that an NP may get nominative/accusative because its syntax and semantics collaborate to mark it so. Syntax and semantics have independent principles to determine case. Surface case is the result of resolution between syntactic and semantic case. In most cases, syntactic case matches its corresponding semantic case, and complete redundancy occurs. Sometimes, syntactic case conflicts with semantic case; languages may differ in choosing syntactic case over semantic case, semantic case over syntactic case, either syntactic or semantic case, both syntactic and semantic cases, etc. I develop formal machinery under conceptual semantics (Jackendoff 1990, 1997, 2002) and the case-in-tiers theory (Yip, Maling & Jackendoff 1987; Maling 1993; Maling, Jun & Kim 2001) to express the fundamental idea that case is both syntactic and semantic. Empirical support for this theory comes from the aspectual nominal construction (=ANC) in Korean and Japanese, the light verb construction (=LVC) in Korean/Japanese and Hindi, and the dative subject construction in Old English. A detailed cross-linguistic investigation of these constructions uncovers a number of problems that are hard to deal with in either a purely syntax-based or a purely semantics-based case theory. Once we try to understand case in terms of both function and meaning, however, most, if not all, difficulties in these constructions find a natural explanation. Based on my investigation of case within conceptual semantics, I conclude that the role of case in grammar is to establish correspondence between syntax and semantics. |
