Abstract

In Chapter III of Principia Mathematica and several later writings, Russell gave a proof to show that definite descriptions are “incomplete symbols” that have “no meaning in isolation”. Despite the importance Russell seems to attach to the proof, many commentators have regarded it as obviously fallacious. Perkins has offered an interpretation that aims to rehabilitate the proof, but does not, I suggest, vindicate the conclusion about meaning in isolation in a sufficiently robust sense. This paper aims to close that gap by offering an alternative interpretation according to which the proof relies on Russellian ideas about propositional structure. It concludes by revisiting the frequently debated question of whether Russell could have accepted a generalized-quantifier treatment of definite descriptions on which they denote second-order properties.