Descriptions and Explanations of Temporal Variation

This paper presents ongoing research on the semantic representation and model-theoretic evaluation of descriptions referring to abstract objects of temporal variation, i.e. events and states (”eventualities”). While common ontologies for temporal entities are geared to the surface of natural language descriptions (in the spirit of [Vendler, 1957], [Davidson, 1967]), I propose to model the relation between descriptions and temporal entities by a constitutional ontology of distinctions among temporal variations based on the type of explanation which is used for the segmententation, identification and consequent description of the respective temporal entity. With respect to explanations of temporal variation, recent investigations propose a threefold distinction between causal, behavioral and intentional explanation [Dennett, 1989, Dretske, 1988, Hartmann and Janich, 1991] which I adopt to enrich the representational formalism of Discourse Representation Theory (DRT, [Kamp et al., 2007]) with operators that specify how temporal processes are related to descriptions by means of explanatory identification of temporal entities. Technically, this can be achieved by an integration of branching-time logic (CTL*, [Emerson, 1990]) into the framework of DRT, a direction of research which except for [Asher and Singh, 1993] has not received the attention it probably deserves. In addition the proposed treatment of temporal entities allows for an elegant integration of insights from artifical intelligence [Schank and Abelson, 1977], psychology [Zacks et al., 2001] and linguistics [Kamp, 2007]. I illustrate my proposal for an action-theoretic on∗[email protected] tology of temporal entities by an analysis of accomplishments, where I argue that the temporal profile of an accomplishment is identified by an intentional explanation of temporal variation.