ABSTRACT
This chapter describes a generic semantic grammar that can be used to encode themes and theme relations in every clause within randomly sampled texts. The preceding illustrations demonstrate that in applying a semantic grammar to a sample of texts, the researcher assumes that the texts are structured according to the semantic grammar and that the phenomenon of interest is related to this structure in a specific way. In a semantic text analysis, the researcher encodes only those parts of the text that fit into the syntactic components of the semantic grammar being applied. A generic semantic grammar is required to encode interrelations among themes within a domain of relatively unstructured texts. The argument here is that in ordinary discourse a speech act's meaning consists of an unintentional, taken-for-granted component plus an intentional, asserted component. The ensuing discussion reveals a structure of linguistic ambiguity within ordinary discourse by showing that descriptive utterances admit of semantic opposites.
