An Experimental Investigation of the Semantics and Pragmatics of Specificity

There are many different theories of specificity in the semantic literature. These include specificity as scope (Ioup 1977, among others), specificity as referentiality (Fodor and Sag 1982), specificity as a choice function reading (Reinhart 1997, Winter 1997, Kratzer 1998, among many others), and specificity as speaker identifiability (Groenendijk & Stokhof 1980). Importantly, theories of specificity often rely on fairly subtle judgments. This is particularly true when it comes to a certain-indefinites, a type of specific indefinite that has been much investigated in the theoretical literature (e.g., Hintikka 1986, Kratzer 1998, Schwarz 2001, Farkas 2002, among many others). The goal of this paper is to examine the behavior of a certain-indefinites experimentally, in order to provide evidence for or against particular semantic theories. The main study reported in this paper examines the semantics of a certain-indefinites, taking as its starting point is the proposal of Schwarz (2001) that a certain-indefinites and a-indefinites are derived through different semantic mechanisms. A secondary, pilot study reported in the Appendix examines the pragmatics of a certain-indefinites, probing the requirement of identifiability (cf. Farkas 2002). Both studies use context-based interpretation tasks with linguistically naïve informants. The findings provide support for existing theories, but also raise further questions concerning the interpretation of English indefinites.