A conceptual semantics account of negative polarity licensing: A study of the any expression.

Abstract

This thesis, in the spirit of work done in generative grammar, focuses its attention on the crucial link between a theory of meaning and a theory of understanding and how this link may be expressed in the form of a linguistic theory for it to enjoy psychological plausibility. The current approach adopts Jackendoff's (1983; 1990) model of grammar which primarily conflates "semantics" and "pragmatics" into an autonomous level called Conceptual Structure (CS). Another aspect of the CS model which will be emphasized in this thesis is the correspondence between phonological structure and conceptual structure and its effect on syntactic structure. In essence. the model underscores the creative capacity of all three levels while de-emphasizing the "syntactocentrism" (Jackendoff 1990) of previous models of grammar. This thesis uses the CS model to account for the phenomenon of negative polarity licensing in general and to provide an explanation, in particular, for examples such as the following: (1) (a) $\sp*$John didn't see anyone. He saw Bill. (b) John didn't see ANYONE. He saw Bill. (2) (a) $\sp*$Anyone cannot enter. (b) (Just) ANYONE cannot enter. The fact that metalinguistic negation is significantly different from truth-conditional negation and that it is, in many cases, coerced by contrastive stress which, in turn, affects grammaticality reflects the "correspondence" between the various levels of grammar. Besides linking various levels of grammar by correspondence rules rather than hierarchically arranged derivational rules, this model allows us to propose a conceptual notion of negativity in order to unify the class of negative polarity licensers. The advantage of using such an approach rests on the fact that the level of conceptual structure, and the process of conceptualization in itself, is not completely language-dependent. Hence, this means that some of the hypotheses proposed in the thesis are universally applicable. Overall, this thesis suggests that conceptually relevant semantic theories can indeed give us what we need from a theory of interpretation--which is, general rules of interpretation that are able to go beyond the "formal pyrotechnics" (Hornstein 1984; p.149) of "semantic and pragmatic approaches to meaning".