In recent work, tableau-based model generation calculi have been used as computational models of the reasoning processes involved in utterance interpretation. In this linguistic application of an inference technique that was originally developed for automated theorem proving, natural language understanding is treated as a process of generating Herbrand models for the logical form of an utterance in a discourse. This approach captures anbiguity by generating multiple models for input logical forms.
In this paper we apply the model generation approach to a particular case of ambiguity: the interpretation of negated sentences. Using model generation, we will demonstrate how the various possible readings of simple negated sentences are generated, and by what criteria an interpreter chooses among these possibilities. Our investigation of negated sentences will lead us to propose constraints on the model generation system which, we will suggest, represent broadly applicable principles of interpretation.