Saturday, January 10, 2009

Deliberation and Freedom

Hi everyone,

I'm not sure if everyone is on this any more. If someone who is and knows that others, new grad students for example, are not, maybe we could invite them.

I've been meaning to post for a while on a question I have about deliberation and free will. I don't have a well-developed view at all, but I have a strong intuition that seems to be at odds with at least a few claims made by people I don't immediately discount as crazy.

My question is: Why should we think that deliberation requires free will? That is, why should we think that, if determinism is true, and incompatibilism is also true, we could not truly deliberate? As I've said, my view is not well-developed. As of now, I just respond to those who think determinism + incompatibilism precludes deliberation with a confused look and a challenge to explain why my unfree actions cannot stem from deliberation. So, any thoughts on why?

From what I gather, the view that being unfree or determined precludes deliberation tends to stem from a conceptual argument. People like Peter van Inwagen argue that an unfree agent who recognizes her lack of freedom would be paralyzed and unable to deliberate because of her knowledge that only one road was truly open. Presumably this type of conceptual argument is backed up by intuitive cases. For example, if I know that there is no milk in the fridge, I do not deliberate about whether to get up and get myself a glass of milk. I recognize my desire for milk, but recognize also that deliberating about getting milk is precluded by facts about the contents of my apartment. Presumably lots of other everyday cases like these can be found.

However, it seems like the crucial detail in the milk case and others we might think about is our knowledge of the state of the world (i.e. whether milk is in the fridge). If we know there is no milk, we do not deliberate. If we did know there was milk, then we would dliberate about whether it was worth getting up for. Most importantly, if we did not know whether there was milk or not, we might still deliberate about whether to get up and get some.

I think that the last situation most closely resembles our hypothetical situation as determined or unfree agents. While we know that only one future state is possible, we do not know what that future state is and do not know which of our actions are determined and which precluded by determinism. Of course, we MAY have souch knowledge in some cases and in those I have no problem sayin that we would not (and probably could not) deliberate. Oedipus, for example, was fated and knew about his fate, even if he didn't believe it to be his fate, but was still able to deliberate about how to go about avoiding it. To say that he was not really deliberating about how to avoid his fate only on the grounds that he was fated and knew this fact seems silly.

Does someone knows this literature, of which I hear there is lots, including some work by Dana, and want to explain how I'm missing the point here? Lots of smart people seem to hold this view. Epicurus and his followers for example fall in the "no deliberation without free will" camp. Peter van Inwagen holds this position also.


Matt Brown said...

So do Kant and McDowell, so far as I can tell.

The problem (for Kantians) comes down to what does it mean to reason or be rational, or what does it mean to follow a rule? To be subject to a law of nature is to be absolutely compelled (on the concept of law that you're working with), to not have any options. To act on a reason or follow a rule is to CHOOSE to conform your thinking to some norm / practice. By definition, it requires that one have options.

So, since deliberation has a normative side (you can deliberate well or poorly, i.e., rationally or irrationally), then it isn't compatible with hard determinism. A computer program, for example, is only rational or rule-following in a derivative sense, on the basis of some free being able to check / vouch for its correctness. It doesn't have options for how to process information.

Of course, someone who denies this (like Shaun Nichols, for example), will just say that being rational is just following the rules of logic, which is just to be like the computer program. This seems to deny that there is anything at all normative about logic, and also to run afoul of Wittgensteinian considerations about rule-following and game-playing.

soup said...

matt, could you say what those Wittgensteinian considerations are and how nichols runs afoul of them?

Matt Brown said...

Insofar as I understand Wittgenstein's point, it seems like it is that a rule is not an abstract entity that one can uncontroversially and unproblematically apply to action. Whatever the application of a rule is, it's not a transparent, explicit algorithm.

A similar point can be made without Wittgenstein, I think, but instead with Lewis Carroll's article on Achilles and the Tortoise.

soup said...

Per, there's a lot of stuff going on in here I think. There are at least two different views you could be disagreeing with/confused about. The first says that we can't deliberate in the absence of free will. The second says we can't deliberate if we know we aren't free.

The first would at first glance seem like a metaphysical thesis and/or a conceptual claim about what it is (what is required) to deliberate.

The second seems, at least largely, a psychological claim. When Van Inwagen says that "an unfree agent who recognizes her lack of freedom would be paralyzed and unable to deliberate because of her knowledge that only one road was truly open," it suggests at least two things: 1) An unfree agent who didn't know this might well be able to deliberate and 2) a free agent who was convinced that she wasn't free would not be able to deliberate because she thought she knew that only one road was open.

This idea of being paralyzed by such knowledge can only be maintained by insisting that hard dterminists, when they do manage to deliberate (ignoring the conceptual/metaphysical issue for now, and only focusing on the issue of whether one need be paralyzed by the knowledge/belief that one is not free), are 'acting under the idea of freedom' as Kant would put it. That is, they forget or put out of mind somehow their lack of freedom. Or perhaps he could say they never really believe it in the first place, but that's a far less plausible move.

Finally, this supposed paralysis invoked by Van Inwagen doesn't rely on incompatibilism being false, but it is a big part of why he regards it as false. If it is in fact the case that we would be paralyzed in our decision-making/deliberations by believing/knowing that (say) physical determinism was true, then one could leave it open whether compatibilism is true or not. Knowledge/firm belief in determinism precludes (at least synchronic) deliberation, as a psychological fact. Perhaps 'true free will' (I can't believe philosophers people still take such an idea seriously) is compatible with determinism, but it will still be the case that we can't deliberate with (at least synchronic) awareness of determinism.

Now, what of the actually unfree agent who doesn't know/believe that she's unfree? Can she deliberate? I don't remember what Inwagen says about this, but I would guess not. But the reason can't be that she's paralyzed for the above reasons. There would now presumably have to be some arguments of a different sort. Or rather, perhaps Inwagen would say that she can deliberate, but not freely; again, now there would have to be other kinds of arguments for incompatibilism.

I think you're totally right to insist on the importance of the idea that we don't yet know which way things will be. Knowing that we are determined to choose something, but not what that is, is completely different than knowing that we are determined to choose a particular thing. In the latter case, perhaps we cannot deliberate. In the former, I see no clear reason why we can't.

As I said before, van Inwagen would have to deny that all the determinists (they don't have to be hard, recall) out there believe their determinism when they're actually deliberating.

First, there are surely very many cases where even this isn't a problem. As matt notes, nichols says that being rational is like following rules of logic. In lots of cases, my deliberations will take the rough form of simply evaluating likelihoods of certain outcomes, such as if I'm deliberating about whether to make a certain wager. I don't see any problem with making such calculations under the explicit idea that all the calculations are predetermined to come out a certain way, and so therefore are my actions. Maybe this doesn't count as deliberation for inwagen. No matter.

At any rate, not all deliberations are like this. Most people in the free will business are concerned with decisions with ethical import, and which cannot be reduced to the application of rules of logic.

So let me say where I think some of the intuitive purchase inwagen is after comes from. Suppose you're deliberating about whether or not to eat meat. You are trying to weigh the considerations for and against, but the thought that the result of all these thoughts, and your ultimate course of action is already written in the stars keeps intruding. 'Only one road is truly open to me' you tell yourself. You don't know yet which one it will be, but this idea that only one of them is 'open' and the other is 'closed' conflicts with the sense you have during deliberation that you could choose either one, and which one you do choose will be at least in part the result of deliberative processes that evaluate the merits of each.

This is van Inwagen's core idea, as well as Kant's (I think). You can't deliberate about how to choose between putative options and at the same time believe that at least one of them is not really an option.

In my opinion, the source of the conflict is that in this case someone is trying to adopt two practically (as opposed to theoretically) incompatible standpoints. From the standpoint of a deliberator, it's true that both must seem to be options in some sense. From the standpoint of someone regarding the process of deliberation, if determinism is true, these options will not seem to be 'genuine' in the way they are for the deliberator. If you try to adopt both standpoints at the same time, the standpoint required for deliberation is undermined by the other one.

But so what? Here's another way to become paralyzed. Suppose we are 'radically free' such that all 'voluntary action' is free action and in principle open to deliberation as to its rectitude. Try to keep that standpoint ('There are indefinitely many roads open to me' as opposed to 'there is only one road open to me') at the same time you try to make mundane or even significant decisions in life. Going to a movie? But there are so many other things that you could be doing that would be better for you and/or more morally praiseworthy. From the standpoint of a deliberator, you might decide that you just want to watch a movie or get a beer with friends rather than work tonight. Or (not) eat meat rather than the reverse. From the standpoint of one regarding one's decisions as a radically free agent, these options seem absurdly limited. There are a million things you could do! There are thousands of possibilities between eating meat of whatever kind whenever you want and never eating any. There are thousands of things to other than work or go to a movie. What about writing poetry or a novel or in your journal or a short story or learning spanish or french or japanese or chinese or italian or working out or becoming a war correspondent or joining PETA or going on a campaign against PETA or becoming a doctor or lawyer or scuba diving instructor or any of hundreds of other things you could do? Surely whatever you're doing, there's something better you could be doing. What's the very best possoible thing you could be doing right now?

I submit that this standpoint is every bit as paralyzing as the other, if one adopts it while actually going about the business of one's life, or even during deliberation. Does this show that we're not really free? I don't think so.

I think that both standpoints are in tension, or practically incompatible, with making ordinary decisions in life. What kind of conclusions about free will can be drawn from this? As a psychological point, it seems that in order to effectively deliberate, we need some way of reducing the perceived options to a number below hundreds or thousands and above 1.
Perhaps we can't operate under the (synchronic) idea that there is really only one option, but I submit we can't operate under the (synchronic) idea that there are thousands of available options either. Does this show that there aren't 'really' thousands of options, or that there are 'really' at least two? Of course not, and I doubt van inwagen would think it does (but who knows). I think it shows 2 things. One is that effective deliberation occurs when the perceived options are bounded, with 2 at the bottom. I also think that there would be a problem with deliberating between even two options if one were synchronically thinking 'I am entirely free to choose either of these'. This is perhaps a special case of the tension between effectively doing something and regarding oneself as doing that same thing at the same time. And without the requirement that there be the synchronic (or rapidly switching) thought 'Only one road is open to me,' I don't think van Inwagen's claim is plausible, since there seem to be plenty of hard determinists who deliberate just fine (bracketing metaphysical/conceptual objections), at least so long as they are not attempting to occupy the two practically incompatible standpoints at the same time.

And not occupying the theoretical standpoint should not be confused with not believing its content. To say that I'm not thinking at a given moment about paris being the capital of france doesn't imply that I don't believe it.

So, it seems that we can both believe in determinism and deliberate at the same time (as a psychological matter at least) so long as belief is understood dispositionally, interpretationally, or any other way that doesn't require belief to be consciously and explicitly held at a time to exist at that time (which would be a stupid notion of belief).

The alternatives are 1) to deny that the people who profess determinist beliefs really have them at all, or 2) that they really deliberate. 1) is just desperate. 2) leaves open two further possibilities. The first is that non-determinists do deliberate, but not the determinists. The second is that none of us really deliberate (for metaphysical/conceptual reasons). Again, the first is pure desperation and unworthy of response. The second could be true, depending on your conception of deliberation. But the more plausible (whether psychological, physical, etc.) determinism becomes, the more reason to change your conception of deliberation such as to no longer require non-determinism. One could say that non-determinism is an essential requirement, and that giving it up requires giving up deliberation. But the problem with that move is that it is obviously false, and again, like a lot of defenses of free will (compatibilist and incompatibilist alike) smacks of desperation.

Now I am desperately uncomfortable. Damn that was long, but I'm too uncomfortable to do a lot of editing.

Peace out.

Per said...

I definitely agree that the question of whether deliberation is precluded by determinism has both a conceptual and a psychological side.

On the conceptual side it seems, as Eric said, to be a question of the metaphysical requirements on possibility of deliberation. It seems like that is what Matt is saying the Kantian problem would be for determined deliberation. To be subject to the laws which the definition of causal determinism invokes is by definition incompatible with the type of event that deliberation is. I'm tempted to respond to this type of position by saying that I think deliberation is being defined too narrowly. Or by saying that I can, in fact, act on a reason even if the resulting action was the only one available to me. That seems to be a perfectly coherent case of deliberating without metaphysically open alternatives.

As for the psychological side of the claim that determinism precludes deliberation, I like Eric's point that paralysis seems to follow from even the false belief that determinism is true. I take this to be a reductio of the paralysis argument, at least insofar as we think that there have ever been determinists that held (truly or falsely) believed themselves determined. There are plenty of self-proclaimed examples of this: Calvinists, certain Islamic sects, Diodorus Cronus, etc. Shawn Nichols actually catalogues some of these in his "After Incompatibilism". Of course, it's possible that they just don't hold the belief that they're determined at the moment they deliberate or decide. I'd want to hear more, though, about why a belief in determinism MUST paralyze us. Or, as Eric said, I want an account of how my self-proclaimed uninterrupted belief in determinism is not actually uninterrupted, or not a belief.

This post has mostly been agreement with Eric and restatement of my dissatisfaction with the claim that determinism precludes deliberation, but are there other arguments than the ones pointed to by you guys (Matt and Eric) that might hold more water?

Is there something besides a psychological, and empirically verifiable (and disconfirmed I would think), claim about paralysis in the face of determinism or belief in determinism?

Is there better motivated conceptual argument that argue that the concept of deliberation requires the denial of determinism?