2012. Paper presented at the IMI workshop in Paris, January 30-31, 2012Paper abstract bibtex
Interrogative inquiry refers to the process of knowledge-seeking by questioning. In this paper, we investigate the process of interrogative inquiry in the context of conversations. To this end, we develop a formalization of interrogative inquiry based on inquisitive semantics and pragmatics. This is motivated by the capacity of inquisitive semantics to provide a semantic account of questions and answers in natural language, and the capacity of inquisitive pragmatics to provide a pragmatic account of the behavior of questions and answers in conversations. The paper begins with a presentation of the modelling of questions and answers in the inquisitive framework. Then, after a brief comparison of the inquisitive account of questions with Hintikka's treatment of questions in the Interrogative Model of Inquiry (IMI), we discuss and define the notion of interrogative rule which aims to characterize the question-answer steps that one can make in an interrogative inquiry. We then put the interrogative rule into a temporal perspective, by introducing the notion of interrogative protocol, which aims to govern interrogative inquiry as a temporal process. The notion of interrogative protocol enables us to reach formal definitions of the notion of interrogative inquiry and the associated logical notion of interrogative consequence, that we illustrate with some concrete examples. Our framework thus defined allows then for a formal logical and computational study of the process of interrogative inquiry. On the logical side, we relate the notion of interrogative consequence with the ones of distributed information and yes-no question. On the computational side, we shape the bases of a computational investigation of interrogative inquiry in our framework, and we present first computational results. From this computational perspective, we propose to revisit the so-called strategic aspects of inquiry, one of the main themes of Hintikka's IMI, from an algorithmic point of view. We end this paper with some concluding remarks and suggestions for further works.