Friday, October 29, 2004

On Event-Causation

Brandon at Siris writes:

Some of you philosophically-minded types who hold that event causation (understood as a relation between two distinct events) is the primary sort of causation can perhaps help me here. I'm currently working on a paper that touches on issues of event causation, and want to make sure I don't do anyone any injustice. I can understand someone thinking that causation is an event (I think so myself); but why would one think of causation as a relation obtaining between two distinct events? Such a view rules out any possibility of causation being something that is done, because events don't do anything - they just are, and come before and after and during each other. So causation can't be an act or action, even a relative one.

BV: Consider an example. A ball hits a window, and the window breaks. Is it the ball that is the cause of the window’s breaking, or is it the ball’s hitting the window that is the cause? If the former, the cause is a substance; if the latter, the cause is an event. Substances and events presumably belong to distinct ontological categories. Saying that the ball is the cause of the window’s breaking seems to leave something out since the ball at rest, or the ball moving in some region of space away from the window, cannot be the cause of the window’s breaking. Thus it is very plausible to say that it is the ball’s hitting the window that is the cause – in which case the relata of the causal relation would appear to be events.

Likewise, it is not the sun that causes the stone to become warm, but the sun’s shining on the stone. It is not the moon that causes the tides but the moon’s gravitational attraction. It is not sodium that is part of the etiology of hypertension, but the ingestion of sodium in certain amounts by sodium-sensitive people. The ingesting of sodium is an event.

If this is right, then causation is not an event but a dyadic relation between events. But now extend the example. Jack Jackovsky threw the ball that broke the window. The throwing -- an event -- caused the ball’s moving, but what caused the throwing? A whole series of events that lead back to the brain. And beyond that? Jack’s intention to throw, which is a mental event. One problem is how a mental event can hook onto a series of physical events. Put that on the back burner. Consider the mental intention to throw. What caused it? Jack the free agent caused it. For some this makes no sense. They reject agent-causation entirely.

Still, one naturally thinks that causing is a doing, and that doing requires a doer, an agent to substitute Latin for Anglo-Saxon. Brandon is right: events don’t do anything. But how can we make sense of the notion that substances like a ball or the sun do anything? Hume’s point was that causing is not empirically discernible. How meet Hume’s challenge?

Second, if causation were a relation obtaining between distinct events, what about the relation prevents it from obtaining between things other than events (e.g., substances, or numbers on a number line, and so forth)?

BV: Event-causation by definition relates events. It is not clear that there is a problem here. Consider a parallel question: what prevents the subset relation from obtaining between entities other than sets? Well, because it just makes no sense to say that a fetus, say, is a subset of its mother. Neither fetus nor mother are sets. The subset relation is defined over sets; the event-causation relation is defined over events.

Third, what reason do we have for thinking events can be distinct objective entities rather than just arbitrary intentional markings-out of regions of spacetime?

BV: One might say that an event is an individual’s having a property at a time, or through an interval of time. More generally, an event is an n-tuple of individuals’ exemplification of an n-adic relation at a time or over an interval of time. (Definition inspired by Jaegwon Kim.) We could then say that events are individuated by their constituent individuals (substances). Thus Socrates’ falling asleep is individuated by Socrates. Events would then be no more arbitrary than their constituent individuals.

I hope these rough remarks are of some help.