A Probabilistic Account of Logical Metonymy

In this paper we investigate logical metonymy, i.e., constructions where the argument of a word in syntax appears to be different from that argument in logical form (e.g., enjoy the book means enjoy reading the book, and easy problem means a problem that is easy to solve). The systematic variation in the interpretation of such constructions suggests a rich and complex theory of composition on the syntax/semantics interface (Pustejovsky, 1995). Linguistic accounts of logical metonymy typically fail to exhaustively describe all the possible interpretations, or they don’t rank those interpretations in terms of their likelihood. In view of this, we acquire the meanings of metonymic verbs and adjectives from a large corpus and propose a probabilistic model which provides a ranking on the set of possible interpretations. We identify the interpretations automatically by exploiting the consistent correspondences between surface syntactic cues and meaning. We evaluate our results against paraphrase judgements elicited experimentally from humans, and show that the model’s ranking of meanings correlates reliably with human intuitions.