Friday, March 16, 2007

Difficulties with Davidson's account of action (II): What is the force of the "because" that links action and desire? (by guest blogger Joshua Rust)

It is worth drawing attention to the flexibility of the subordinating conjunction, "because". I can think of at least three importantly different senses of the word. It can flag a causal explanation: "my car stopped because it was out of gas." It might indicate an inferential relation: "My car must be out of gas because it stopped." Finally, it can denote a clarificatory relation: "He is a bachelor because he is an unmarried male" or "He was precious because of his skills and experience." In both of these cases, we are specifying what is meant by, or what counts as, "bachelor" or "precious." We are not inferring his preciousness, nor do his skills or experience cause him to be precious.

Davidson is concerned to explicate sentences like, "He flipped the switch because he wanted to turn on the light." In what sense is the word "because" being used here, where turning on the light is an action? Indeed, he frames his own project in these terms: "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. Of course, we can include this idea too in justification; but then the notion of justification becomes as dark as the notion of reason until we can account for the force of that 'because'" (Davidson 1963, 691).

The orthodox reading of Davidson sees the subordinating conjunction which links reason and action as exclusively causal. Reasons, Davidson says, "rationalize" or "causally explain" action.

But, if as I argued in the previous post, actions are defined in terms of their causes (primary reasons), it doesn't seem as though these same causes can then be cited as a causal explanation of that action.

What, then, is the force of that 'because'? It is obviously not inferential. We don't come to know that he flipped the switch because of his beliefs and desires. But I think a case can be made for the subordinating conjunction which links reason and action as being clarificatory.

Consider the case of motion sickness, which is also defined in terms of its causes. If I asked "why am I experiencing motion sickness?," you cannot answer "because of the motion." If you did answer in this way, your answer might rightly be construed as a rejection of the question—"what else," you might be wondering, "did you want to know, that you didn't already know?!" Now consider the following response to the same why-question: "because of the movement of the boat." This seems like a perfectly good response to the same question. But notice that boat-movement is just a kind of movement. Why would citing a kind of movement be a good answer to the question, when citing movement in general is not? In fact, I don't think it is an answer to the why-question at all. It's rather an informative rejection of the question. Why-questions are answered when we cite some event in the causal stream leading up to explanandum, which is not already specified by the explanandum (see Woodward's non-triviality requirement in the previous post). Since the sickness is already known to be caused by motion, saying that motion is, further, boat-motion doesn't satisfy this requirement. What the response is doing is not causally explaining the motion sickness, but rather clarifying for the interlocutor what kind of motion sickness he/she has—boat or seasickness. The subordinating conjunction in the claim, "I am experiencing motion sickness because of the movement of the boat," is clarificatory rather than causally explanatory. It says what I mean by motion sickness in this instance—seasickness.

Back to Davidson: If I ask, "why did you flip the switch?," when I already know the body movement to be an action, you may not explain this action by saying that you had primary reasons. This much is already entailed by its being an action, if actions are defined in terms of their causes—primary reasons, consisting of beliefs and desires. But you may nevertheless respond, quite appropriately, "Because I wanted to turn on the light". This is not a causal explanation; if I already knew the body movement stemmed from a desire (=action), in telling me what kind of desire you have you are not citing a cause that falls outside the scope of the explanandum, as required by the non-triviality requirement. But have nevertheless learned something: the flipping of the switch is an instance of turning on the light, and not alerting a prowler. This is analogous to your telling me that my motion sickness is, moreover, seasickness. But notice that the "because" which links my desire to turn on the light and my flipping the switch is not causally explanatory at all, but rather clarificatory.

I don't think that Davidson would agree to my characterization of his view. He clearly thinks that desires rationalize or causally explain action. But as my first post indicates, I'm not sure this view is tenable. But Davidson also says of his view that "The defense no doubt requires some redeployment, but not more or less complete abandonment of the position, as urged by [Anscombe, Hampshire, etc.]". These authors, I think, would be sympathetic to the idea that the "force of the 'because'" is more clarificatory than causally explanatory.

1 comment:

Roman Altshuler said...

Maybe I am completely missing the point, but I am still not clear on why we must reject the view that (1) what makes x an action in general is that it is caused by a belief/desire pairing, and (2) the causal explanation of x is given by a further specified belief/desire pairing.

My confusion is about why rejecting this account for the reasons given does not lead us to reject causal accounts as explanations altogether. Let us say that my refrigerator exploded and I want to know why. It turns out that my meatloaf was there for so long that it turned explosive and was set off by a spark from the refrigerator light. This ostensibly provides a causal explanation, if anything does.

But what makes this an explanation at all? In looking for a causal explanation for the explosion of my refrigerator, I am looking for something that is (1) a cause, and (2) a physical event. So I must already be assuming that the explosion can be explained by some cause that is a physical event.

But why, then, does it add anything to the explanation to mention a meatloaf and spark? After all, we already assumed that the explosion can be explained by a cause that is an event. Why does specifying the kind of event and, for that matter, specifying the kind of cause, add anything by way of causal explanation? If it does, then why doesn't specifying the particular reason (belief/desire pair) add a causal explanation to our account of an action?

Of course we might know that an action in general, by virtue of being an action, is caused by some reason. But when we want to causally explain specific actions, naming the specific reason that rationalizes it does seem to provide a causal explanation--at least, it does tell us something we didn't already know, just as the meatloaf/spark story tells us something we didn't already know.