Empirical methods in the study of semantics Summary

A too short summary of this paper might be that language is what it is because we use it to say things. The capacity to say something - to affirm or deny some comment about some topic - may not be uniquely limited to man, but certainly he is better at it than any other animal. Saying something requires that we have certain grammatical machinery in our languages, ways to combine topics and comments, ways to make one sentence the topic for another comment. Saying something also requires that we have certain semantic machinery, so that what we say can be interpreted by our listener on the basis of what he knows about the meanings of its constituent parts. These are problems that linguists and psychologists share, and that form the kernel of the young science called psycholinguistics. If and when we are able to achieve a deeper understanding of what men do when they say something, we should be able to use that understanding to improve the communication of meaning.