Skip to main content.
October 20th, 2005

Defending Functionalism

Over at the new Fake Barn Country (now with real fake barns!) Jonathan Ichikawa argues against functionalism about belief. I’m always up for defening functionalism, so I want to resist.

Here’s Jonathan’s argument.

  1. In general, I can’t just decide to believe p.
  2. In general, where A is the kind of thing that could be an action of mine, I can decide to do A.
  3. So belief can’t be explicated in terms of what actions I do.
  1. So functionalism is false.

The problem is all in the step from 3 to 4. (Although I’d quibble with 1 as well.) Functionalists, at least clever ones, say that there are three characteristic attributes of beliefs.

  1. Input states
  2. Internal roles
  1. Output states

Jonathan only focuses on the third. But beliefs aren’t constituted by what people do, but by these inputs and these internal connections as well. And since I can’t just choose what my inputs will be, or how my mental states will generate other mental states, I can’t just choose my beliefs.

Jonathan sounds like he denies the second of these, but I don’t quite understand that position. If I could just choose to, for instance, take some random p to be grounds for believing q, then I would be (tacitly) choosing to believe If p then q or p is a reason to believe q or something similar. And by hypothesis that’s meant to be ruled out.

But there’s a deeper reason the functionalist should resist the step from 3 to 4. We don’t analyse belief in terms of action (since victims of paralysis can have beliefs) but in terms of dispositions to act. And as someone who has all kinds of bad habits he’d like to kick, let me tell you it isn’t just that easy to adjust your dispositions to act.

For instance, I have the habit of running away when a car is bearing down on me. (This isn’t one of the bad habits – I don’t reveal that on a blog.) I couldn’t easily lose that disposition, even if I decided playing automotive chicken was one of the highest of human experiences so I seriously wanted to lose it. As far as I can tell, it’s almost exactly as easy to change these kinds of dispositions as it is to change my beliefs. In this case I really mean exact; the only feasible way of changing my disposition would be to form the belief (I mean the real belief, not a probabilistic version) that the cars would stop.

So these kinds of plasticity considerations that Jonathan adduces, rather than telling against functionalism, seem to me to be weak evidence in favour of functionalism.

Posted by Brian Weatherson in Uncategorized

1 Comment »

This entry was posted on Thursday, October 20th, 2005 at 12:34 am and is filed under Uncategorized. You can follow any responses to this entry through the comments RSS 2.0 feed. Both comments and pings are currently closed.

One Response to “Defending Functionalism”

  1. Jonathan Ichikawa says:

    Thanks Brian, this is helpful. (Some of these points have also been made at FBC.) The point about input states is well-taken and very relevant; you’re right that I neglected them. But I didn’t mean to be only considering action, as you suggested. I took my argument to apply to internal roles that beliefs play as well. About this suggestion, you say:

    If I could just choose to, for instance, take some random p to be grounds for believing q, then I would be (tacitly) choosing to believe If p then q or p is a reason to believe q or something similar. And by hypothesis that’s meant to be ruled out.

    I’m not sure this is right. Isn’t it at least question-begging against the anti-functionalist? The relevant assumption seems to be something like any time I commit myself to infering q from p, must I have the belief that if p then q. But I’d like to deny this. I could just adopt that rule for myself. This would probably be very irresponsible, and I’d have to work against some old habits, but I don’t see why I couldn’t do it. And I don’t see why if I did do it, it’d have to be a belief.

    As for dispositions, well, yeah, habits are hard to kick — I just admitted as much in the last paragraph. But it seems to me as though habits are hard to kick in a very different way from the way that beliefs are hard to kick, or to adopt. It’s not just that it’s hard to believe that the earth is cubical — if I really put my mind to it, I won’t be any closer to succeeding than if I just tried a little bit. The only way to do it is to set of a complex chain of self-deception.

    Compare this with quitting smoking, or refraining from fidgeting with pieces of chalk while lecturing (a particular habit of mine). These can be hard, that’s just because they’re hard, not because they’re the sort of thing we can’t just do. Sure, maybe it’d help to get some sort of complex feedback system, but if I really put my mind to it, I could stop fidgeting. Or if I couldn’t, that’s a failure on my part — I may wish I had more self-control. The case with adopting beliefs we don’t have is different.

    Maybe my concession to — or at least my non-rebuttal of — the point about inputs is all the functionalist needs. That’s cool with me. My real project, the reason I’m worrying abou this, isn’t really to rebut functionalism; it’s to establish that there could be something that plays most of the functional roles of belief that isn’t belief, and that we can have a lot of voluntary control over it.