Sunday, May 8, 2011

Actions, reasons, and causes

I've been hunting through Davidson to find the arguments for the claim that reasons are causes. He says:
a person can have a reason for an action, and perform the action, and yet this reason not be the reason why he did it. Central to the relation between a reason and an action it explains is the idea that the agent performed the action because he had the reason.

That sounds fine. So, we might imagine two subjects with the same reasons for acting who act for different reasons and demand some causal story that explains why the reasons for which they acted differ. Why would such a story identify the reasons as the causes of the action? All we need is some causal difference between the stories, so I don't see why the causes themselves have to be the reasons. Here's a view. Reasons are facts. There are also what Audi calls "reason states", states that represent reasons. So long as there are different causal relations among the reason states, we give Davidson everything he wants without accepting the view that the agent's reasons for acting consisted of beliefs, desires, or combinations thereof.

1 comment:

ADHR said...

Well, Davidson does say that reasons-explanations are a species of causal explanation, so that gets close to the claim that reasons are causes. I also note that he doesn't actually say that beliefs and desires combine to form a (primary) reason -- he says beliefs and pro-attitudes, the latter being defined as including desires but also moral beliefs. (Going from memory here, so he may not say "beliefs" as such.)

All that said, it's not clear that the different reasons problem needs a causal solution. Gordon has a paper where he defends a counterfactual solution to the problem, contra Davidson. (One could also question whether it's even a problem, but that's a whole 'nother issue.)