Linking Form to Meaning: Reevaluating the Evidence for the Unaccusative Hypothesis

Abstract

This dissertation investigates the linking between meaning and form of intransitive verbs by examining a popular linguistic hypothesis that postulates a one-to-one correspondence between the syntactic position of the argument and the semantic role it takes (the Unaccusative Hypothesis). It shows that the empirical data do not support the hypothesis that there is a binary syntactic distinction. In three eye-tracking experiments, I demonstrate that the processing patterns that were previously argued to support the Unaccusative Hypothesis are not replicable or reliable. Next, I conduct a close investigation of three unaccusative diagnostics and find that these diagnostics do not clearly distinguish unaccusative verbs from unergative verbs. This dissertation argues that the movement account of unaccusativity is neither compatible with the empirical data nor theoretically meaningful in contemporary syntactic theories. It further questions the notion that there is a dichotomy in intransitive verbs and suggests that multiple categories of verbs may better explain the data pattern. This claim needs further investigation which is beyond the scope of this dissertation.