Internal and External Negations

Abstract

Russell' distinguished primary from secondary occurrences of definite descriptions: in 'The king of France is not bald' read as 'There is one and only one king of France, and he is not bald', 'The king of France' has a primary occurrence; in 'The king of France is not bald' read as 'It is not the case that there is one and only one king of France who is bald', 'The king of France' had a secondary occurrence. Later2 what he had to say about definite descriptions Russell also took to hold for namesordinary proper names. Ordinary proper names were just 'disguised descriptions'. But the distinction between primary and secondary occurrences which Russell applied to definite descriptions he never applied to pronouns. While 'The king of France is not bald' could be read in two ways, there was never any suggestion that 'He is not bald' had this ambiguity, and this got Russell into trouble. The differences between Russell's treatment of definite descriptions and his treatment of pronouns have been shown, by Strawson,3 to be the source of Russell's demand for 'logically proper names'. 'Logically proper names' would have to be those names which functioned as Russell's pronouns were taken to function, and there are no such. But Strawson, while he sees that to avoid the demand for impossible names we must allow pronouns to have a logic parallel to that for names does not reach the, surely obvious, conclusion that we must apply the distinction between primary and secondary occurrences to pronouns.4 It is this conclusion I want to spell out in this paper, along with some of its consequences.