Negation, ambiguity, and presupposition

In this paper I argue for the Atlas-Kempson Thesis that sentences of the form ⌜The A is not B⌝ are not ambiguous but rather semantically ‘general’ (Quine), ‘non-specific’ (Zwicky and Sadock), or ‘vague’ (G. Lakoff). This observation refutes the 1970 Davidson-Harman hypothesis that underlying structures, as ‘full semantic representations’, are logical forms. It undermines the conception of semantical presupposition, removes a support for the existence of truth-value gaps for presuppositional sentences (the remaining arguments for which are viciously circular), and lifts the Russell-Strawson dispute of 1950–1964 from stalemate to a formulation in which a resolution is possible for the first time. Suggestions of Davidson, Montague, Stalnaker, Kaplan and H. P. Grice are shown to be inadequate semantic descriptions of negative, presuppositional sentences. I briefly discuss the radical Pragmatics view of my 1975 publications and suggest that it too fails to do justice to the linguistic data. I speculate that Semantic Representations should be given the form (more or less) of computer programs, describable in Dana Scott's mathematical semantics for programming languages.