The class of statements mentioned at the end of the last chapter is the class of perception statements. The linguistic facts again provide prima facie grounds for saying that such statements admit both of an extensional and of an intensional interpretation. However, these linguistic facts can be reconciled with the claim that taken in their standard meaning perception statements are extensional. We can explain, consistently with this claim, how it is sometimes legitimate to understand these statements intensionally.

It is a contentious question whether a statement of the form ‘Smith sees an F’ entails that there exists an F which Smith sees. Without doubt many examples can be given of situations where the utterance of such a statement seems to lack this implication. For instance, talking of a drunk man, or of a man suffering an hallucination, one might say ‘He sees pink rats’, and not imply that there are pink rats which he sees. Again, one talks of seeing spots in front of one's eyes when there are no spots in front of one's eyes. Further examples are given by Miss Anscombe in her article ‘The Intentionality of Sensation’.

Confronted with such linguistic phenomena, some philosophers have argued that in the primary sense of ‘see’, ‘Smith sees an F’ does not carry an implication of existence, while others have claimed that there are two senses of ‘see’, in one of which statements of this kind carry an implication of existence, and in the other of which they do not.