| Abstract: | A Natural Semantic Metalanguage (Wiezbicka 1972) approach which takes advantage of discourse-analytical methods as a source of data can be used not only to precisely define lexical items, but also to answer questions as to word choice and distribution. Cumming & Ono (1997: 132) point out that ‘the attention that has been devoted to the lexicon has focused primarily on explanations for lexical category distinctions… rather than to choice within lexical sets.’ An NSM approach which takes advantage of discourse data is well-positioned to address the issue of choice within lexical sets, and a model for how speakers do so is outlined in this paper. Such an approach can best be understood in terms of Optimality Theory (Prince & Smolensky 1993), and would therefore have the added advantage of providing a single possible framework for the understanding of semantics, phonology, and syntax. |
