Thursday, January 25, 2007

David Velleman, The Possibility of Practical Reason, "Introduction" (Part II)

[This is a continuation of the summary of our discussion of the "Introduction" to David Velleman's The Possibility of Practical Reason, which begins in the previous post.]

Velleman's Account of What Makes Behavior Into an Action

Velleman dismisses the "standard model" and the "hierarchical model" of what makes behavior into an action. What is Velleman's model?

Velleman's striking view is the following (this, as usual, is a rough summary): what makes something an action is that it is done with the "higher-order aim of knowing what [one] is doing". The Freudian slip case isn't an action because the speaker doesn't utter "I hereby declare this meeting closed" or "I live in a building with a hated pool" in order to know what he is doing. The climber dropping his partner and the speaker's crying don't count as actions for the same reason--the person doesn't drop his partner or cry in order to know what he is doing.

The workshop consensus about this view was that it was very strange. It seems straightforwardly false that I do the various things I do in order to know myself. Jason has often used the example of saving a drowning child to illustrate the strangeness of Velleman's view. Say I see my child drowning and I jump in to save his life. A philosopher asks me, "Why did you do that?" It seems that Velleman would think it reasonable to say "So as to better know myself". But that would be a weird thing to say.

There has been some feeling among some of the members of the workshop that we're not really grasping something important about Velleman's view, because the (rough) way of presenting it that was just given looks very implausible. I think one of David's reasons for assigning this "introduction" this week was that it looked like Velleman had an account of how the desire for self knowledge was not an "agential" reason, but something "sub-agential", and so not the kind of thing that you'd cite in an action-explanation. So, perhaps with self-knowledge as a sub-agential reason, you wouldn't get strange explanations of why you jumped in the river to save your child like the one given in the previous paragraph. But after a closer look, we couldn't find anything more substantial than Velleman's claim that the desire for self-knowledge is "sub-agential", and there was not an explanation of what that means. Does being "sub-agential" mean the desire for self-knowledge doesn't, or can't, figure in ordinary action-explanations? It's not clear.

There was a fair amount of discussion of points of detail, but I will conclude with one foundational question that was pressed (like many of the others) by Jason.

The Constitutive Aim of Action
Why think that there is some one thing that is the constitutive aim of action? It looks like a specifically philosophical urge to find a single, overarching principle that holds together everything that falls within the grab bag of behavior we call actions. Maybe some of our actions aim at world peace, and others at achieving some personal satisfaction, and some aim at amassing wealth, and so on? Need there be some aim that all of these share? Jason suggested that there need not be. Giving up the search for such a single aim present in all actions might radically change the shape of a theory of action, possibly for the better.

The workshop will meet again in two week's time to discuss some portion of John Campbell's Reference and Consciousness. See you there.

Wednesday, January 24, 2007

David Velleman, The Possibility of Practical Reason, "Introduction" (Part I)

At the end of last meeting, David suggested that we read the introduction to Velleman's The Possibility of Practical Reason to wrap up our discussion of Velleman, since he thought it would provide the best view of the position as a whole and maybe answer some of the questions that have been raised over the past quarter about Velleman's view of action.

So we read it, and tonight we met to discuss it.

As one might expect, Jason gave a very clear and incisive introduction to Velleman's "introduction", and raised some worries about it.

Velleman is interested in answering the following question: given an event, and an agent, what makes it the case that the event is an action of the agent? More specifically, he's interested in a constitutive, non-circular answer to that question. Jason noted that thinking that question can be answered is already to make a substantial assumption.

But let's say we assume that such a constitutive account of what makes an event an action is possible. What's the best version we can give?

The Standard Model

What Velleman calls the "standard model" attempts to answer this question in the following way: an event is an action (of an agent A) provided it is caused by a belief and a desire (belonging to A). More specifically, "we want something to happen, and we believe that some behavior of ours would constitute or produce or at least promote its happening" (5). The belief and the desire are supposed to be both the cause and the reason for A's action.

Velleman says that the Standard Model (SM) "runs afoul of obvious counterexamples". In the counterexamples, "behavior is caused by a desire and a belief but fails to constitute an action performed for reasons".

Counterexample #1: Velleman gives his version of Davidson's mountain climber case. A speaker "desire[s] to win the sympathy of his audience, and his belief that nothing short of tears would suffice [to win the sympathy of his audience]" "frustrate[s] him to the point of tears" (7). So the belief and the desire cause the speaker to cry, but not in the way that is required for the belief and the desire to count as a reason for crying.

The SM could be modified to exclude a case like this, possibly in the way proposed by Davidson, where the belief and the desire have to cause the action "in the right way" (which is not meant as a non-circular way of specifying what's required to count as a reason), or in the not obviously circular way suggested by Velleman: the belief and desire have to "exercise their characteristic powers in causing the behavior".

There was some argument between Jason and Will about whether the Davidsonian way of handling this kind of counterexample was the same as the one mentioned by Velleman (the consensus reached was that they're not, because Jason said that "characteristic powers" is a phrase that has its home in reductive naturalistic accounts of action).

Counterexample #2: Even if the standard model is modified in the way proposed by Velleman (by adding the requirement that a belief and a desire have to cause behavior according to their "characteristic powers" for the behavior to count as an action), there is still another problem that shows it can't be an adequate account of what makes a bit of behavior an action.

Velleman thinks that "activity" like Freudian slips and "bungled actions" (which don't turn out to be actions at all, on Velleman's account) pose a problem for the standard model. When Dr. Katz says, in a conversation about his ex-wife, that he lives in a building with a hated pool, his utterance was caused by a desire to express his hatred of his ex-wife, and a belief that by saying "hated pool" instead of "heated pool", he would express that belief, and his motive (his belief and desire) would be exercising their characteristic powers in this case. And yet Velleman says that in the case of a Freudian slip like this, Dr. Katz's utterance "doesn't qualify as an action" (8).

So the standard model classifies things that aren't actions as actions. So we need a better account of what makes behavior action, rather than mere "activity".

The Hierarchical Model
Velleman then considers a Frankfurt-like "hierarchical model" of what makes a bit of behavior an action. Roughly, according to the hierarchical model, when a belief and a desire plus a higher-order desire for the desire cause a bit of behavior, that behavior is an action (12).

Velleman thinks that "the hierarchical an improvement on the standard model, because it requires the subject to be reflectively aware of his motives in order to act autonomously. A Freudian slip takes its agent by surprise, thereby casting him in the passive role of observer...Such a lack of self-awareness would not have disqualified the resulting behavior from being an autonomous action according to the standard model, but it is indeed disqualifying according to the hierarchical model. For an agent cannot want or be content to be motivated by a desire he is unaware of having" (12).

Several people (Jason most of all) took issue with this claim. Why can't someone have unconscious second-order desires? Jason gave (a version of) the following example: In therapy, I try to understand why instead of making fun of some hapless coworker, I inadvertently protected him from ridicule by doing something unexpected (comically crashing into a cubicle, say). I reason about my behavior as follows: "I must have wanted to help my co-worker avoid ridicule, and I must have thought that that was a worthy thing to do (desired to desire it)". So it's at least not obvious that there can't be unconscious second order desires.

But this isn't a serious objection to Velleman's overall project, because Velleman doesn't think the hierarchical model suffices as an account of action. But we had a worry about why Velleman thinks it is insufficient. Velleman's view that the hierarchical model is insufficient rests on a version of Freud's case of the president who opens the meeting by saying "I hereby declare this meeting closed". If the president was depressed about the opening of the session, he might thereby have desired to close the session, and so had a second-order desire to desire to declare the session closed. But in such a case the president, according to Velleman, would not have been "autonomous"--his utterance would not have counted as an action: "If anything, [the cause of the president's behavior] would have expressed a lack of will on his part, under the weight of a psychic force that is usually regarded as pathological or alien" (13).

The worry (again raised by Jason) about Velleman's account of the inadequacy of the hierarchical model was that it seems like Velleman is committed to the odd view that behavior that is caused by depression isn't action. This reflects a recurring tendency of Velleman's to say certain things people do (Freudian slips, bungled "actions", behavior at least partly caused by depression or ennui) are "alien", when it seems intuitive that they are still kinds of actions.

to be continued...

Second Meeting of the Winter Quarter

The Mind Workshop met tonight to discuss the introduction to David Velleman's The Possibility of Practical Reason. This concluded our quarter-and-a-half-long discussion of Velleman's theory of action. I will post a summary of the central concerns that have been raised during our discussions soon.

In two weeks, we will meet to read and discuss some portion of John Campbell's Reference and Consciousness.

Sunday, January 07, 2007

First Meeting of Winter Quarter

The Philosophy of Mind Workshop will meet on Wednesday, January 10th to discuss David Velleman's "From Self Psychology to Moral Philosophy". The workshop meets from 6-8pm in Cobb 101.

See you there.