Cecile T. Balkanski
Utterances that include rationale clauses and means clauses display a variety of features that affect their interpretation, as well as the subsequent discourse. Of particular importance is the information that is conveyed about agents’ beliefs and intentions with respect to the actions they talk about or perform. Hence, for a language interpretation system to handle these utterances, it must identify the relevant features of each construction and draw appropriate inferences about the agents’ mental states with respect to the actions and action relations that are involved. This paper describes an interpretation model that satisfies this need by providing a set of interpretation rules and showing how these rules allow for the derivation of the appropriate set of beliefs and intentions associated with each construction.