Abstract

The topic of this chapter is the semantics of incomplete definite descriptions or sentences of the form ‘the F is G’ where more than one object satisfies the nominal F. The traditional Russellian analysis holds that definite descriptions are quantifiers which contribute uniqueness conditions to propositions. This view renders all tokens of incomplete definite descriptions false. There are two general strategies for maintaining the quantificational analysis; one is to treat definite descriptions as containing hidden indexicals and the other is to concede their non-denoting nature (and thus the falsehood of their tokens). It is argued that the former ‘semantic’ strategies lead to overgeneration while the latter ‘pragmatic’ strategies share with their semantic counterparts the mistaken claims that when someone speaks, something unique, and exclusively determined by context of utterance, is expressed. Once these errors are revealed, incomplete definite descriptions are shown to pose no problem for the quantificational analysis.