Representation of inference in the natural language

The purpose of this work is to investigate how processes of inference are reflected in the grammar of the natural language. I consider a range of phenomena which call for a representational theory of mind and thought. These constructions display a certain regularity in their truth conditions, but the regularity does not extend to closure under arbitrary logical entailment. I develop a logic that allows me to speak formally about classes of inferences. This logic is then applied to analysis of indirect speech, belief reports, evidentials (with special attention to Bulgarian) and clarity assertions.