On denoting descriptions
Paul J. E. Dekker · 1997
this paper we want to show that the notions of information that have become fashionable in various recently employed systems of dynamic semantics, are very useful for the characterization of the referential interpretation of definite descriptions, and the specific interpretation of indefinite ones. With this more or less trivial observation, we hope to serve a more general goal. To our opinion, current linguistic theorizing fails a clear and generally accepted picture concerning the delineation of the semantic and pragmatic components of a comprising theory of interpretation. In addition, we think, we nowadays fail a convincing conception of the interaction of these two components of the whole theory, even where we were to have consensus about their delineation in certain areas of work. With the work reported in this paper we want to bring back the issue of the semantics -- pragmatics interface back on the agenda again. Employing tools from dynamic semantics, viz., its notions of information, we present a separate, but combined, semantic -- pragmatic treatment of definite and indefinite noun phrases. Basically, we adopt a Russellian treatment of these descriptions on the semantic side, thereby (i) assuming some form of context dependent quantification and (ii) adopting a dynamic-style constructive reading of the (implicit) existential quantifiers. A modification of Russells theory along the first lines is obviously needed in order to get rid of too strong uniqueness requirements in the case of definites (and of too weak readings of indefinites). The modification along the second lines is key to this paper. Precisely the use of dynamic semantic `discourse referents'---or `subjects' as we call them here---provides us with a hook to attach the pragmatics of these noun phra...