Some complications with neurolinguistic research on unaccusativity

Abstract

Two classes of intransitive verbs involve different syntactic-semantic linking patterns. Unergative verbs follow the canonical pattern because the subject noun-phrase (NP) expresses the actor of the described event (e.g., The boy ran), whereas unaccusative verbs have a noncanonical pattern because the subject NP expresses the undergoer of the described event (e.g., The boy fell). Many neurolinguistic studies suggest that, compared to unergative verbs, unaccusative ones are harder to process. After summarizing this literature, I point out two complications with it. First, a large proportion of the researchers unjustifiably assume a Chomskyan analysis of unaccusative verbs whereby the NP linked with the undergoer role is an underlying direct object that must be moved to the subject position. Second, most of the studies ignore several important aspects of unaccusativity, including problems with putative syntactic diagnostics and extensive crosslinguistic diversity that is nonetheless semantically constrained.