Gradable nature of semantic compatibility and coercion: A usage-based approach

This study investigates the gradable nature of semantic compatibility between constructions and the lexical items that occur in them and also of the related concept of linguistic coercion, in relation to actual language use, on the basis of the usage-based model, proposed by Langacker (1987). The study shows that semantic compatibility between linguistic elements is a gradient phenomenon, and that speakers' knowledge about the degree of semantic compatibility in a given case is closely correlated with language use, specifically language processing and frequency of usage. This study also shows that coercion, which is the resolution of semantic incompatibility between a construction and a lexical item occurring in it, is also a gradient phenomenon related to usage. To do so, this study investigates linguistic knowledge of semantic compatibility between the English sentential complement construction and various verbs that occur in it, and compares this semantic compatibility with empirical data obtained from acceptability judgments of various sentences, a corpus, and an experiment on sentence processing. My findings show that the more compatible a verb is with the construction, the faster their co-occurrence is processed and the more frequently it is used. On the basis of this correlation between gradable semantic compatibility and usage, this study suggests that the study of coercion be expanded to investigate its linguistic and extra-linguistic contexts and to determine what kind of interactions lead to a better or easier resolution of incompatibility, by incorporating empirical language use data.