Home Research Publications Courses Workshops Tutorials Blog

-----

Compositionality and the Theory of Argument Selection


LOT Winter School
Nijmegen, The Netherlands

Date: January 8-12, 2007



James Pustejovsky
Computer Science Department Brandeis University
Waltham, MA 02454 USA
jamesp@cs.brandeis.edu


Course Abstract


In this course, I will address the question of how words encode context. More specifically, I examine the formal mechanisms responsible for argument and adjunct selection in language. I outline a theory of lexically type-driven semantic selection and composition that permits restricted polymorphisms, while explaining the creative use of words in novel contexts. Building explicitly on Generative Lexicon's richer notion of compositionality, I explore the type language and logic necessary to model local context in natural language. Words encode local context as typing information. The compositional rules of the language refer to these types. The operations possible during composition for these types are: i. selection, ii. accommodation, and iii. coercion (exploitation and introduction). As an illustration of this theory, I explore the phenomena of nominal and verbal polysemy, and verbal alternations.

Recently, there has emerged an appreciation of how complex this problem is, as well as a new understanding of the parameters at play in the interpretation of polysemous expressions. Two classes of parameters have been broadly identified as contributing to the interpretation of polysemous words: more complex lexical representations, and a means of incorporating local context compositionally. In this course, I formalize this distinction as that of inherent versus selectional polysemy, and demonstrate that polysemy cannot be modeled adequately without enriching the compositional mechanisms available to the language. In particular, lexically driven operations of coercion and type selection provide for contextualized interpretations of expressions, which would otherwise not exhibit polysemy. I contrast this with Cruse's view (and others) that it is not possible to maintain a distinction between semantic and pragmatic ambiguity. I will argue that a strong distinction between pragmatic and semantic modes of interpretation can be maintained, and is in fact desirable, if we wish to model the complexity of contributing factors in compositionality.

Readings for the Course