This paper argues that the definiteness effect (DE) of possessive have sentences in English is indicative of a discourse-pragmatic function of presentational focus, drawing on analogous earlier proposals for the DE in existential there sentences. The DE is proposed to be encoded via a lexical specification of possessive have for an indefinite complement, supported by interpretational differences between have sentences arising from the (in)definiteness of the have complement. That is, have is argued to be polysemous, and there is one particular sense of have that licenses only indefinite complements. This sense of have (i.e., possessive have) encodes an underspecified possessive relation, determined at the sentence level by the meaning of its complement NP. The DE of possessive have is argued to reflect a presentational function based on the informational lightness of have, a property characteristic of verbs in presentational constructions such as locative inversion. Thus, by hypothesis, the meaning of possessive have consists of two components: it encodes (1) an underspecified possessive relation, and (2) a presentational function. This proposal is supported with an in-depth discussion of definite complement have sentences. It is argued that there are actually two separate senses of have that license definite complements, and these separately reflect the possessive and presentational components of possessive have.
Proceedings of the 2004 Texas Linguistics Society Conference: Issues at the Semantics-Pragmatics Interface
edited by Pascal Denis, Eric McCready, Alexis Palmer, and Brian Reese
Table of contents