Open Access

Stefan Engelberg

• The paper argues that lexical-semantic representations within event semantics often suffer from a severe lack of empirical content. To demonstrate this, some main tenets of truth conditional semantics and its consequences for the lexical-semantic metalanguage are discussed. It will be shown that lexical semantic meaning postulates have to be improved in the following ways: i) the formulation of empirically verifiable definitions of the metalinguistic predicates, ii) the clarification of the logical type of these predicates, iii) the establishment of a procedure by which the predication can be verified, iv) the careful observation of the semantics of the logical operators, and v) the establishment of identity criteria for the basic ontological sorts. Furthermore, the strategies many lexical-semantic theories develop in order to immunize themselves against empirical falsification will be identified in several examples. Finally, it will be argued that an empircally sound lexical-semantic theory within truth-conditional event semantics has to be based on empirical research in cognitive science