Semantic parameters and the unaccusative split in the Austronesian language family

A fundamental issue facing all models of grammar is the mapping problem: how to link the semantic arguments defined by the lexical verb in a clause to their formal realizations in morphosyntax. This paper investigates this problem across a range of intransitive verb types, often grouped into unergative and unaccusative classes. It demonstrates that the criteria which determine this split vary across the Austronesian languages investigated, but that a strict hierarchy, putatively universal, of semantic parameters will predict exactly where the various splits can be drawn.