Thursday, March 17, 2005

General Freedom

I previously advanced an expansive understanding of freedom, such that any constraint whatsoever counts as an impingement upon it. The laws of nature are such that I am not free to jump over the moon. But one might question the importance of this conception. At least, it needs to be supplemented by some account of our 'on balance' freedom, or freedom simpliciter. Although I lack the particular freedom to jump over the moon, it would be most implausible to conclude that I am thereby unfree in any general sense. Some particular freedoms are more important than others. In this post, I want to explore the question of how we might weigh particular constraints as being more or less important, which may then serve as an account of general freedom.

The whole point of freedom, by my earlier conception, is that it enables us to achieve our goals or desires. So this strikes me as the logical place to look for a scale against which to weigh constraints. The simplest view which springs to mind might be called the 'actual desire' account:

(AD) Constraints are to be weighed according to the strength of the desires that they thwart. I thus have 'general freedom' to the extent that I can get what I want.

One objection appeals to the happy slave scenario, where we feel the happy slave is unfree despite the fact that he can still achieve his strongest desires. If one is unfree, you cannot change this simply by ceasing to care about it. The facts are unchanged - you remain unfree whether you care or not.

To some extent I agree with this - the happy slave has particular unfreedoms insofar as he must work for his master and lacks other options. But 'general freedom' is a rather different concept. It is not just about in what ways you are constrained; it is about which constraints actually matter. Insofar as the happy slave is able to achieve everything he desires through his servitude, he suffers no significant constraints.

Alternatively, we may feel that (AD) is too narrow. After all, if the slave had changed his mind, he would not be able to act on his new desires. The fact that he lacks this ability might be deemed significant, whether or not his desires are actually thwarted. If you feel swayed by such considerations, you might prefer a broader, 'possible desires', account of general freedom:

(PD) Constraints are to be weighed according to the strength of a possible desire they might thwart, and the likelihood of your having such a desire. I thus have 'general freedom' to the extent that I can get what I want in this and all close possible worlds.

This view has some nice results. It suggests that the happy slave is unfree, because he could very easily have had different desires and thereby felt the force of the constraints upon him. But (PD) also suggests that I am not made unfree simply because I cannot jump over the moon. There is no nearby possible world where I would have a genuine desire (as opposed to mere whims) to jump over the moon. So this constraint is not a significant one.

One might object that my desire-based approach effectively conflates 'general freedom' with wellbeing - either actual wellbeing, as in (AD), or our wellbeing in close possible worlds, as in (PD). I'm not sure whether this is really a problem, though. My fundamental 'freedom' concept is that of particular freedom, which remains appropriately distinct. It is only when we ask which freedoms really matter that our answer collapses into wellbeing. But since wellbeing is what matters, it should come as no surprise that the forms of freedom that matter are those that enable our wellbeing! It's almost tautological. But perhaps I've been discussing the wrong sort of significance. Perhaps general freedom is not about those freedoms that are significant qua good life, but rather, those freedoms that are significant qua freedom.

This notion may be illuminated if we consider the flipside of freedom: coercion. On G.A. Cohen's analysis, one is 'forced' to do something when there are no reasonable or acceptable alternatives (e.g. "your money or your life!"). This suggests a 'reasonable alternative' account of general freedom:

(RA) S is free in doing A iff S has a reasonable alternative to A that she could have chosen instead.

This is fairly vague, and it isn't immediately clear how to cash out this notion of a 'reasonable alternative'. One option would be to appeal to a probabilistic or statistical notion of what the average person would consider reasonable in that situation. But I think a better option is to judge this based on the specific individual in question. We could do this through a possible worlds analysis, i.e. something like:

(PW) B is an acceptable alternative to A iff it is chosen by the agent in some close possible worlds (and the relevant details of A and B remain fixed across these worlds).

(This bears some rough similarities to an account I've suggested before, though in a somewhat different context.)

Cohen suggests an expected utility analysis of reasonable alternatives:
(EU) B is an acceptable alternative to A iff: either B is not worse than A, or B (though worse than A) is not thoroughly bad.

This then leaves the notion of 'thoroughly bad' unspecified, but it could be spelled out as some absolute threshold of wellbeing.

Now, these reasonable alternative views of freedom end up being quite radically divorced from our notion of wellbeing. If I can push a button to get everything I desire, then I'm obviously never going to accept any alternative. The option is much better than any other, so much so that I would choose it in all close possible worlds. If we further suppose that the alternatives (which I never seriously consider anyway) are "thoroughly bad", then, according to (RA), I am thus 'unfree'! Conversely, if all my options are equally miserable, then I am apparently 'free', because whichever I end up choosing, there would have been an alternative just as good that I could have chosen instead.

Given such consequences, I'm not really sure that (RA) is a very interesting or important concept. I actually prefer the concepts of general freedom that are tied more closely to wellbeing. They make it more obvious just why freedom is worthwhile. If freedom is not tied to wellbeing in any way, why should we care about it?


  1. Obviously, we disagree pretty fundamentally about what freedom actually is. I continue to be baffled by people who don't agree with what I've said on this issue, but hey, that's life...

    I like the metric of pr(desire occuring) * strength of desire * pr(desire thwarted) as a measure of unfreedom. I can see how that brings clarity to a lot of situations.

    However, I still maintain there is a real difference between a physical constraint and, as you have termed it, an unfreedom.

    As such, I would say it is not proper to term the inability to leap over the moon an unfreedom, but merely a constraint.

    As for the example of a man trapped in a cave, he may not be free - but only in the sense that he is trapped. He remains at liberty to do as he pleased in context. I think that the context of liberty is the proper context for considering freedom, and that other senses of the word are best left to informal discussion.

    As for the formula, it might not be perfect. As we learn and experience what is an is not possible, we also learn to mitigate our desires. For example, we learn our morality - what not to do - which modifies our desires. Our belief about what is possible alters the probability of us arriving at a particular desire.

    There is a sense in which the equation fails to capture the dynamism of a real person with a history, however I think it is an excellent moment-in-time metric.

  2. being able to jump over the moon is a freedom rather like being able to think complex thoughts is a freedom for you but one an insect (as far as it thinks at all) might see as implausible.

  3. how much you want either of those things probably has quite a bit to do with to what degree evolution thought you needed them.


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.)

Note: only a member of this blog may post a comment.