Tracking Verbs Across Languages

Beth Levin

Current theoretical linguistic research on the lexical knowledge of native speakers of a language shows that the human lexicon is highly structured; verbs fall into classes on the basis of shared meaning, and the members of these classes have in common a variety of properties concerning the expression and interpretation of their arguments, as well as the extended meanings that they show. What is of particular importance to work in machine translation is what light linguistic research can shed on the space of cross-linguistic similarities and divergences in the organization of the verb lexicon. In this talk I will discuss some insights into these questions that have emerged from my recent work with Malka Rappaport Hovav on intransitive verbs in English and other languages. My goal will be to illustrate parallels between the syntactically-relevant semantically-defined subclasses of the intransitive verb class across languages, as well as striking, though systematic, differences in the properties that some of these subclasses exhibit across languages.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.