Durations: Temporal Intervals with Gaps and Undetermined Edges

Abstract

The study of the meanings of temporal expressions in natural language can proceed in two ways. The first consists of borrowing an ontological theory concerning how time “really” is, and then showing how temporal expressions are interpreted in this model. Let us call this the physicalist approach. The other approach is to start off by studying the temporal presuppositions employed in the language, and defining a model as the structure which satisfies these conditions. This approach we shall call the analytist approach. The two approaches share their goal, which is to provide a temporal model for natural language. However, they are driven by different motivations. The physicalist approach is ontology-driven. It views time as part of reality, and its structure as independent of linguistic considerations. It attempts to interpret the language in this model, possibly by adding assumptions about the language. The analytist approach, on the other hand, is language driven. It assumes that language “mirrors” reality to the extent in which it is describable in the language, and tries to define a temporal model based only on linguisticconsiderations. A theory is “complete”, in a sense, if it combines both approaches and its (analytist) language- driven model can be constructed in a (physicalist) ontology-driven structure.

This paper was supported in part by the National Science Foundation, grant number IST81-18393. It is based on an unpublished manuscript (Manor 1979). The current version is the result of much encouragement from Hans Kamp, Johan van Benthem, and Dov Gabbay who commented on the earlier version, and from Stanley Peters and Arthur Cody, who commented on and made several suggestions to improve the current version.