ABSTRACT
It would be natural to say that the names ‘Maurice’ and ‘London’, as they occur in an utterance (or inscription) of the sentence ‘Maurice’s mother lived near London’ refer to things (a person and a place, respectively), and that someone using these words would be so referring in normal circumstances. It would also be natural to say that the occurrence of ‘Maurice’s mother’ in an utterance (or inscription) of the sentence refers to something (although wrong philosophically if Russell is right), and that someone using these words would be so referring. In sharp contrast, it would be decidedly unnatural to say that any of ‘lived’, ‘near’, ‘near London’, ‘lived near London’, or ‘Maurice’s mother lived near London’ referred to things, or that someone using these words would be so referring. But of course this will not prevent philosophers and linguists from arguing that verbs, adjectives, prepositions, prepositional phrases, verb phrases and even sentences do refer to things. A theoretical notion of reference may have to replace our ordinary notion as far as a theory of meaning is concerned. And it is no news that the recent history of semantics is littered with theories according to which functions, sets, properties, types, propositions, states of affairs, situations and facts serve as the references of all manner of expression.
