Agency, akrasia, and Velleman

I've been reading a bit about practical rationality and reasons for action, and find David Velleman's theory quite interesting (see here, esp. chapter 3). But here's something curious: apparently both Velleman and I take it to be an advantage of his theory that it provides a plausible explanation for akrasia (weakness of will)--and yet we seem to have diametrically opposed beliefs about what exactly this explanation is.

Now, this probably just means that I've misunderstood his view, but, A, I find my interpretation independently interesting, and B, this is my blog, so there.

So. Briefly, Velleman's view of agency is that it consists in being guided by a motivation towards self-understanding, a motive that is constitutive of agency and hence inescapable. He likens it, persuasively, to a particular sort of improv: improvising the role of oneself, where a bad performance, an inauthentic performance, is one that fails to make sense. (Making sense is here cashed out in either a folk-psychologic or narrative sense.)

Now, he's not claiming that this motive dominates all others, merely that it exists and is an essential aspect of agency itself, and hence guides the formation and regulation of our other motivations. It plays the functional role of selfhood, basically.

How does this explain akrasia? Well, it seems like--at least in an earlier work--he sees the virtue of his explanation to lie in demonstrating the genuine agency of the akratic actor. He seems to be saying that akratic actions are those in which either the motive of self-understanding correctly overrides mistaken beliefs about what reasons there are, or the proper weighting of them, or correctly refrains from action, despite being aware of the right thing to do, because the awareness doesn't here constitute a full grasping of how the action fits into the agent's self-narrative.

My objection to this story, put simply, is that it makes akrasia sound okay. But it's not okay; it's awful--and if there's one thing I'm intimately familiar with, it's the phenomenology of akrasia. Under this reading, it seems like what's gone wrong isn't the "will" at all, which is doing fine; it's rather that the belief subsystem hasn't quite caught up to it. And this seems to utterly miss what's so disturbing about the phenomenon.

When I first read Ch. 3 of his manuscript, though, I felt it invited a dramatically different interpretation, one much more attuned to its tragic aspect. On this reading, is not that what the motive of self-understanding grasps has come apart from what the agent believes to be justified, or even to make the most sense. On the contrary--the problem is that the motive of self-understanding, while active, is simply too weak to enforce a coherent narrative. The akratic action may be undertaken while under the influence of that motive while remaining out of step with its demands. As a result the agent feels torn between his self-understanding as someone whose actions are under conscious control and his self-understanding as someone who doesn't do whatever it is he just, akratically, did.

On this interpretation, akrasia is worse that mere non-agency; it's an active subversion of agency, because it engages with the agent's self-understanding while undermining the narrative whose construction is its entire aim.

(I think a similar analysis helps illuminate one aspect of what's so awful about extreme depression.)