Tuesday, May 04, 2004

Choosing Freedom - Choosing Determinism

Continuing on from my previous post about Free Will, I want to more closely examine the different conceptions of freedom which are on offer:

  • (F1) X's action was not deterministically caused.
    i.e. Holding everything constant (including X's mental states), X could nevertheless have done differently. Even if his beliefs, desires, and reasoning were exactly the same, a different decision may have resulted.

  • (F2) X's action was caused by X's own desires, values, etc.
    i.e. X could have done differently if he had wanted to.

  • Let's examine these alternatives by way of an example:
    Suppose I'm driving along the motorway, when the thought crosses my mind that I could swerve across the road and crash into the oncoming traffic. I have no reason to do such a thing (for I believe that doing so would cause injury or death, and I have a very strong desire to avoid injury or death). But suppose that, after going through the reasoning which surely should result in my decision not to swerve... suppose instead that I (inexplicably) decide to swerve... I die moments later.

    Does that sound like a free action to you? I really don't think it is. I would actually feel much more free if I knew for sure that my actions were reliably caused by my beliefs and desires. I would feel much more free if, in that given scenario (i.e. given my current beliefs & desires) , it were truly impossible for me to kill myself (and possibly others) like that. The possibility that I could go through all the reasoning necessary to reach a good decision, and yet have the opposite decision spontaneously result... it sounds more like some kind of mental malfunction, than 'freedom'.

    I think that sort of example demonstrates quite nicely the advantage of F2 over F1. I am free (according to F2) because I could have swerved if I had wanted to. But I didn't want to! Given the fact that I didn't want to, surely I can only be said to be free if my actions conform to my desires in a rational manner. Surely I am only free to the degree that my desires deterministically influence my behaviour. That is to say, surely I am truly free only if (given my beliefs & desires) it is ensured that I would not have swerved at that moment.

    Freedom requires that our actions be caused by our mental states (beliefs, desires, etc). Yet F1 seems to require that our actions be uncaused (or, rather, caused in some unpredictable, or "random" manner). The indeterminate causation required by F1 thus strikes me as being not freedom at all, but rather, an obstacle to freedom's realisation.

    I think what the libertarian needs to do (if he wants to hold onto a coherent notion of freedom) is modify F1 to allow actions to result determinately from our mental states, whilst simply insisting that our mental states not be externally determined.

    To clarify: the compatibalist (F2) merely requires that our desires (etc) be the proximate cause of our actions - but it's okay that our desires were themselves caused by preceding events that were outside of our control. In contrast, the libertarian (F1) would require that our mind be the ultimate cause of our actions. The mind itself - according to this view - must be somehow uncaused ("transcendent", perhaps), and "free" from external influence.

    Put like this, the libertarian view at least makes a bit more sense. But it asks the impossible. Our personalities don't come from nowhere... we are hugely influenced by both nature and nurture - our genes and upbringing - both of which are external to us, i.e. outside our control. If that makes us unfree, then freedom is an impossible ideal, which has never been attained.

    Alternatively, we could choose F2 - a concept of freedom which is fully consistent with determinism, yet nevertheless provides a useful and plausible account of free action (i.e. action as a result of our mental states and reasoning).

    Update: see also Jason's excellent post, Evil Robots, which shows how a machine could have free will.

    1 comment:

    1. [Copied from old comments thread]

      F2 seems so much better to me also. However, at least in Nielsen's article, Nielsen seemed to think that it wasn't enough to leave F2 as "if he had wanted to". He was concerned with cases of X wanting to do the action because of compulsion, and kleptomania. So he added them into his account of freedom:

      1.He could have done otherwise if he had chosen to.
      2.His actions are voluntary in the sense that the kleptomaniacs are not.
      3. Nobody compelled him to choose as he did.

      I think 3 especially makes thinks more problematic for F2. 'Compelled'? Does this mean a kind of conditioning? As in X willfully chose to do the action - although it was only because someone had come along and convinced him otherwise through hypnosis or somesuch? If Nielsens meaning is the one i just gave, then where do you draw the line between hypnosis and simple social conditioning... nurture.
      Patrick Kerr | Email | 9th May 04 - 8:41 am | #


      That is a tricky one. I'll try to flesh out that idea for a "compulsion scale" we were talking about a while back. Here's the core idea: we are free (uncoerced) to the degree that a change in our desires would cause a change in our behaviour.

      Mental disorders such as kleptomania presumably compel their sufferers regardless of their desires. ie, no matter how much a kleptomaniac wishes he could stop stealing, he just can't help himself. A change in desires won't change his behaviour, so he is not free.

      In a situation where you are being compelled (e.g. a guy points a gun at your head and tells you to steal something), the compeller is threatening your deepest desires (eg the desire to live), which cannot easily be changed. A change in your superficial desires (e.g. whether or not you actually want to steal things) is not going to affect your behaviour (the guy is gonna kill you - who cares what else you want!?). Thus you are unfree to a significant degree.

      The hypnosis example clearly fits my criterion for coercion - but how, you ask, is it different from socialisation?

      Well, social conditioning actually forms our desires, whereas hypnotism only temporarily overrides them. If hypnotism was permanent, and shaped our long-term desires in the same way that socialisation does, then I would say that a hypnotised person's later actions are no more coerced than yours or mine. If their desires were different, then they would behave differently. So they would be "free". The fact that their desires in actual fact couldn't be different, is irrelevant.

      So the crucial feature you're looking for is whether the conditioning is long term, i.e. whether it shapes the actual desires of the person (in which case we call them free), or if it temporarily overrides those desires, thus coercing them.

      This distinction may seem unfair. I'll discuss the pragmatic justification for it in my upcoming post on freedom & morality.
      Richard Chappell | Email | Homepage | 9th May 04 - 1:16 pm | #


      It does prima facie seem unfair, but at the same time I'm not sure that I disagree with you!
      Patrick Kerr | Email | 13th May 04 - 8:26 pm | #


    Visitors: check my comments policy first.
    Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)