IRCS Technical Reports Series

Document Type

Technical Report

Date of this Version

July 2002


University of Pennsylvania Institute for Research in Cognitive Science Technical Report No. IRCS-02-01


This paper explores the learning of raising verbs (e.g. seem), verbs which present particular problems for the language learner. In addition to having highly abstract lexical meanings, these verbs fail to provide some of the cues that guide learners to the meanings of other verbs. The central problem explored here is that discovering the syntactic structure of a raising expression (in particular, discovering that the main clause subject is not an argument of the raising verb) is not straightforward. Raising sentences like John seems to be happy are string-identical to control sentences, such as John wants to be happy, but the two have very different structures. The focus of this paper is the question of how a learner could determine the syntactic structure of raising expressions, and thus determine the syntactic and semantic properties of raising verbs. The results of a series of experiments with English-speaking adults are presented, as well as preliminary evidence from two on-going experiments with children. The experiments suggest that good cues to raising verbs or a raising structure come from expletive subjects (it, there) and from the pairing of an inanimate subject with a stative lower predicate (the rock to remain . . . ).


raising verbs, learning, syntactic structure



Date Posted: 14 October 2004