ABSTRACT
That definite description is inherently a pragmatic notion, one which depends on context-derived assumption the speaker makes about what the hearer is likely to know, was not by any means always taken for granted in the logical-philosophical tradition. Thus Russell (1905, 1919) treats definite description as somehow a semantic matter of unique identification. And it is far from clear how such a notion differs from mere semantic reference. When other philosophers departed from such a tradition, as in Strawson (1950), 1 they resorted to the notion of presupposition, i.e. the celebrated three-valued logic. In such a logic, a proposition could be asserted as either True or False; or else it can be Presupposed, meaning that it is taken for granted to be true, as precondition for other expressions to be either true or false (see Keenan, 1969, 1971). Within such a framework, the contrast between indefinite and definite reference is still ascribed to speaker-related features such as unique reference, more specifically to whether that unique reference is asserted (for indefinites) or presupposed (for definites). Thus consider:
Indefinite reference:
A man died
Asserted:
‘There exists a man’
‘That man died’
Definite reference:
The man died
Presupposed:
‘There exists a man’
Asserted:
‘That man died’
