Processes as Continuants (Abstract) Antony Galton
SECaM, University of Exeter, UK
[email protected]
In philosophy it is common to introduce a high-level ontological distinction between continuants and occurrents. Continuants are entities which endure through time: for this reason, they are also called endurants. A continuant can undergo change and yet preserve its identity through those changes. Its parts are spatial parts, and it exists, as a whole, at each moment of its lifetime—notwithstanding the possibility that at different moments that whole might consist of different parts. Occurrents, on the other hand, are extended in time: they ‘perdure’ (hence, they are called perdurants). Unlike continuants, occurrents have temporal parts, e.g., the first half of the concert and the second half of the concert. Because an occurrent as a whole spans the time period from its beginning to its end, it cannot be said to change (although it may be a change), and it does not exist, as a whole, at any one moment during its temporal extent. Ordinary physical objects are clear examples of continuants, and events are clear examples of occurrents. Compare a human being with a human life, an aircraft with a flight, an orchestra with a performance, an army with a battle, a volcano with an eruption, or a heart with a heartbeat. In each case the former is a continuant, the latter an occurrent. A ‘snapshot’ of the world at one time may contain human beings, aircraft, orchestras, armies, volcanoes, and hearts, all existing as wholes. But we do not see, within this snapshot, human lives, flights, performances, battles, eruptions, or heartbeats: at least, we do not see them as wholes, but only minimal parts of them. Objects are extended in space, whereas events are extended in time. This apparently neat and tidy picture, which for example has received a clear expression in modern ontology as the SNAP/SPAN distinction advocated by Grenon and Smith [2], becomes problematic when we turn attention from events to processes. We can see a hint of this when we notice that in ordinary discourse we often seem to ascribe change to events, e.g., ‘my life is becoming more difficult’, ‘the battle became fiercer’. If events cannot change, how can we say such things? The answer is that we are not referring to events here but rather to the processes of which they are composed. Processes and events are clearly intimately related, but even a cursory scan of the literature reveals a striking lack of consensus as to just how they are related. Allen [1] regards processes as similar to events except that unlike events they satisfy a weak form of dissectivity. Mourelatos [5] classifies processes and events as distinct categories under the more
general heading of ‘occurrences’. Sowa [6] regards an event as a kind of process; conversely, Moens and Steedman [4] regard a process as a kind of event: either way, if events are clearly occurrents, then processes must be too. In this talk I propose a radical view of processes as continuants rather than occurrents. Processes, like objects but unlike events, can be the subject of change: the water flow increases, the heartbeat speeds up, music becomes louder. A corollary of this is that our snapshot of the world at one time must contain processes as well as objects; snapshots are thus not static but have an intrinsic dynamism which may be thought of as providing the ‘power source’ for the generation of events. I use Lyons’ [3] distinction between experiential and historical modes of description to underpin the essential contrast between, on the one hand, the world of dynamic snapshots containing objects and processes, and on the other hand, the fixed history of events as faits accomplis, as it were the fossil record of once-active processes. I shall show that this idea is entirely in accordance with standard mathematical ways of modelling the changing world, provides an appropriate explanatory framework for handling the phenomenon of aspect in natural language, and can even throw light on ancient puzzles such as Zeno’s arrow paradox. Finally, I shall also indicate how the conception of processes as continuants can provide a sound basis for the logical modelling of dynamical systems.
References [1] James Allen. Towards a general theory of action and time. Artificial Intelligence, 23:123–54, 1984. [2] Pierre Grenon and Barry Smith. SNAP and SPAN: Towards dynamic spatial ontology. Spatial Cognition and Computation, 4(1):69–104, 2004. [3] John Lyons. Semantics. Cambridge University Press, Cambridge, 1977. 2 volumes. [4] Marc Moens and Mark Steedman. Temporal ontology and temporal reference. Computational Linguistics, 14:15–28, 1988. [5] A. P. D. Mourelatos. Events, processes, and states. In Philip Tedeschi and Annie Zaenen, editors, Tense and Aspect, pages 191–212. Academic Press, New York, 1981. [6] John F. Sowa. Knowledge Representation: Logical, Philosophical, and Computational Foundations. Brooks/Cole, Pacific Grove, CA, 2000.
Proceedings of the Thirteenth International Symposium on Temporal Representation and Reasoning (TIME’06) 1530-1311/06 $20.00 © 2006
IEEE
