In this project, we propose an extension of the theory of properties we have developed in terms of tropes (spatio-temporal manifestations of properties or particularized properties) to the analysis of event structures in language. This will allow us (i) to solve a number of difficulties faced in treating conditionals and modals with presently existing models (in particular the model recently developed by the Japanese team); (ii) to broaden de the range of applications of our ontological theory grounded on the notion of trope to new linguistic data, in particular by confronting it to the data recorded by the Japanese team.

Properties, and especially tropes, have long been the focus of philosophical studies, dating back to ancient and medieval philosophy (Aristotle). Since the recent study of these issues from a philosophical perspective has been very productive, our aim is now to explore possible extensions of our approach to events. In particular, the development of an integration method of spatiotemporal manifestations (tropes) of events (expressed by propositions) in a single event will allow us to avoid problems of ontological over-generation that affect most of the presently existing models for modals and conditionals. Indeed, conditionals and modals, which are very close to one another, have been the focus of a number of studies in the past decades. It has been shown that conditionals could not be properly treated within classical logical systems, and since then, a number of alternatives, such as the ‘modal bases’ approach, have been proposed. They basically rest on a model-theory in which the notion of similarity between worlds plays a central role, or at least use the notion of sets of possible worlds. Methods that allow handling sets of worlds or propositions have nevertheless proven potentially problematic for programming: integrating such (potentially infinite) sets isn’t easy in machine calculus. Our approach of spatiotemporal manifestations of events will allow in particular to develop an efficient tool for overcoming such difficulties, reducing considerably the number of objects assumed in ontology, in order to meet (i).

Concerning (ii), the collaboration with the Japanese team will allow relating our approach of spatiotemporal manifestations of events with the analysis of non-logical inferences, such as ‘Maybe John will come (if he’s not dead)’, in light of the study of the Japanese linguistic data.