Roles, Co-Descriptors, and the Formal Representation of Quantified English Expressions

摘要

In representing the semantics of English sentences it is traditional to distinguish logical form from semantic content. The logical form is represented by some sort of predicate calculus or lambda calculus. In computational linguistics this predicate calculus or lambda calculus notation is usually carried over: directly; by replacing parenthesized scope with 'contexts'; and by replacing universal quantification with dynamically scoped iteration procedures. This paper proposes another possibility for representing logical form. It is based on five main ideas: the use of roles in a semantic net; the referential/attribute distinction; the distributive/collective distinction; the use of predicates taking sets, kinds, or prototypical individuals as arguments; and the use of two levels of representation for quantified expressions.