Linguistic Metaphors and Collocation in Nonliterary Corpus Data

Abstract

The description of linguistic metaphors has often been restricted to intuitively generated examples or the analysis of short texts. The development of large corpora of computer-searchable nonliterary texts enables researchers to make more accurate statements about the use of linguistic metaphors. In the studies described in this article, a computerized corpus has been used to determine the extent to which clusters of words from particular source domains are used in the same target domain. It seems that there may be several ways in which clusters of words are metaphorically mapped. The studies further showed that some constraints on target domain language use are not predictable from models of metaphorical mapping but can be explained by pragmatic factors.