Monday, October 08, 2007

You're all just a bunch of robots, man.

I've been thinking a lot lately about the "determinism makes us robots" assumption that seems more or less pervasive in the popular literature on free will and moral responsibility. It popus up especially in popular articles about the impact of neuroscientific work on free will. The idea seems to be that the truth of determinism would preclude reasoned judgment, which appears to be at least a significant requirement for free and responsible action. But the robot metaphor carries with it the image of self-alienation, as if determinism would have us carrying on mechanistically in spite of some deper free self.

There was an interesting exchange about this on The Garden of Forking Paths recently over an article of Dana's. And since then contributors have been linking to (mostly popular) articles that make this assumption or seem to take it as a serious position that needs to be dispatched.

All this strikes me as interesting both sociologically and philosophically. There's ample evidence to show that lots of people think this way about determinism. What makes, not just incompatibilism, but this particular interpretation of its consequences, so attractive to undergraduates and science journalists? Is compatibilism really that unintuitive? And so what if it is? Eddy Nahmias and Joshua Knobe have pushed this for some purpose that is a bit opaque to me.

Anyhow, thoughts on the determinism to robots move, either anecdotal or otherwise?

12 comments:

Matt Brown said...

Is compatibilism really that unintuitive?

Yes, it is. At least, it always has been to me.

And so what if it is?

Well, that depends. If you think that the debate over free will and moral responsibility turns on folks' intuitions, then compatibilism is screwed. If you don't, then you're going to have to tell me how to otherwise adjudicate the debate.

Per said...

First of all, in response to Mike, I have always been interested in the question of determinism and whether it makes us robots. I've wanted to ask for a while whether people still believe that it does, in fact, make us robots. I think I can give good arguments for the truth of the determinist thesis and I don't immediately sense any desire to obey Asimov’s three laws. But, obviously it's not accepting the thesis that makes us robots, it's the truth of the thesis. Nonetheless, I am surprised to hear that there is a strong contingent of philosophers out there that deny the truth of determinism (or think it entails this heavily negatively connotated robot claim). But apparently there are and I should just look at the things Mike pointed to in his post.

However, since I'm here now, I'll just say what I have to say without the research to back it up. It seems like the claim that the truth of determinism entails some claim about ourselves as agents requires substantial metaphysical grounds that aren’t forthcoming. Supposedly it precludes reasoned judgment which, in turn, limits or negates the possibility of free and responsible actions. But I’ve never heard a good argument for how it would limit free and responsible action or even how it precludes reasoned judgment. Usually, it is just assumed that reasoned judgment is a kind of judgment that is not determined. Maybe this statement betrays shallow understanding of the literature, but I’d like to hear a detailed account of why (and how) determinism makes what we experience as reasoned judgment, and what we define scientifically as reasoned judgment, incoherent or otherwise impossible.

On to Nahmias and Knobe. Their claim, I guess, is that if people are natural incompatibilists, then we should rethink our philosophical intuitions about what we should rightly expect from the folk in terms of actions and motivations for actions. I agree that there is an interesting sociological point here. There are certain facts, both scientific and sociological (sociology is a science of sorts I know, I don’t mean offense), that we might like to know before judging other people’s actions and setting up social institutions (like courts) whose purpose is to judge people’s actions. But I think they take too narrow a view of folk intuition. Affecting those intuitions is the system of judgment (and affiliated institutions of judgment) already in place as well as the more subtle social and cultural structures that provide the intuitive support for more formal structures (again, I’m thinking of courts). There is a vast network that’s responsible for the intuitions Nichols, Knobe, Nahmias, etc are “discovering” in various samples of the folk. This network has been around for a while (choose whatever pseudo-anthropological account you want—e.g. Nietzsche, Hobbes, Freud) and has changed as we’ve learned more about what it is to be a free and responsible agent. Indeed, it seems like history is a continuing lesson in humility concerning what (kinds of) actions we can justifiably be held responsible for. All of the scientific evidence that’s coming on the scene and disrupting New Yorker subscribers’ breakfasts is just the latest installment in that lesson. Do Knobe etc want to know what this new evidence and their poll results mean for how we conceive of (and should conceive of) ourselves as free and responsible? Will we eventually be given enough evidence that we realize that we are robots? I doubt it. Will the class of events for which we hold one another responsible slowly shrink to nothing? Probably not. Obviously the should formulation of those two questions is the more important one. Of course, new evidence should lead us to revise the content of our laws and question the intuitions that lead us to make certain immediate judgments. But the question of whether we should allow our framework of intuitions about free action and responsibility to evolve into one that resembles our framework of intuitions about toasters and teakettles is more difficult to answer.

Now that I’ve written this, I see that I’ve hardly addressed Mike’s question at all. There are probably some arguments for why a determinist world just is a world of robots and I haven’t even tried to imagine an answer to such an argument. But I think the perspective on what it would be like to slowly become more and more concretely aware that something like determinism is the case is an interesting one given the sometimes loud responses in the popular press to evidence that we’re determined in some way or other.

Dan said...

Excuse my ignorance of the literature on compatibilism, but can someone explain to me how determinism does NOT imply that we are just extraordinarily complex robots? It sure seems to me that anything that "can't do otherwise" is a robot, however else one might want to tweak the picture. I don't mean for robot to have any negative connotation here. The key feature of robots, as I see it, is just that they are mechanical, which seems to be precisely what determinism affirms of us (however complicated you want to make the mechanism).

Matt Brown said...

Per said: "It seems like the claim that the truth of determinism entails some claim about ourselves as agents requires substantial metaphysical grounds that aren’t forthcoming. Supposedly it precludes reasoned judgment which, in turn, limits or negates the possibility of free and responsible actions. But I’ve never heard a good argument for how it would limit free and responsible action or even how it precludes reasoned judgment. Usually, it is just assumed that reasoned judgment is a kind of judgment that is not determined. Maybe this statement betrays shallow understanding of the literature, but I’d like to hear a detailed account of why (and how) determinism makes what we experience as reasoned judgment, and what we define scientifically as reasoned judgment, incoherent or otherwise impossible.

The argument for this, basically, is Kant's. I won't pretend to be able to accurately report it after being so far away from it, but the idea would be that, since rational judgment requires cognition that is rule-governed, and being determined means that one is ultimately governed by mechanical rules rather than rules of reason, then determinism is incompatible with rational judgment. Probably the best secondary source on this sort of stuff can be found in Henry Allison's "Idealism and Freedom."

Per said...

Matt said: "the idea would be that, since rational judgment requires cognition that is rule-governed, and being determined means that one is ultimately governed by mechanical rules rather than rules of reason, then determinism is incompatible with rational judgment."

So there's some sort of need-for-autonomy argument underlying the claim that determinism undermines the capacity for reasoned judgment. One needs to provide laws unto oneself, or some such claim. I guess it makes sense why this is an accepted view still, but I still think more needs to be said about why actions that are reason-governed cannot be determined. This is what I was saying about incompatibilists saying that reason is itself undermined by determinism. How is our current notion of reason potentially undermined by determinism? We have a strong philosophical and scientific understanding of what it means to reason and, even on a number of Kantian-style readings (Moran 2000 for one), it doesn't seem like determinism must necessarily render those notions suspect.

But I'm going to think about this some more. There's a Kantian line I like concerning taking a deliberative vs. a theoretical stance toward one's own decision and I should think harder about whether it would be undermined by determinism. Would determinism preclude taking the deliberative stance?

Adam Streed said...

Per asked: "Would determinism preclude taking the deliberative stance?"

Korsgaard says no, following Kant. If I remember correctly, she considers several ways one might formulate the determinist thesis and then claims that, even if true, none of them will help you decide what to do (this is in The Sources of Normativity, maybe chapter 3). The deliberative stance requires the idea of freedom.

She also has a line that I find alternately glib and brilliant---something like "You can't change the past, and that sounds like a joke but I mean it. Freedom is the ability to do otherwise, not the ability to have done otherwise."

I can track these references down and give real quotations if anyone cares.

Adam Streed said...

Also, to weigh in on the survey, I have also found compatibilism obviously false (until very recently), and belief in determinism very, very depressing.

Mike T. said...

See, this is what I'm saying. Determinism should only be depressing if its truth would have some kind of effect on our phenomenal experience, or the value of our relationships and lives. You're toeing the robot line, Streed.

And on the Korsgaard you cite, I'm not sure how those two insights are supposed to fit together. It sounds like she both wants to claim that determinsim doesn't preculde freedom, but that freedom is the ability to do otherwise in some sense. There are people believe in this kind of compatibilism, they provide conditional analyses of "could have done otherwise" that somehow bear on the actual causal sequence. But this doesn't sound like what Korsgaard is after either (given your paraphrase) since it seems to reject exactly this kind of conditional analysis-- it's the abilty to do otherwise, not could have done otherwise...

Other approaches to compatibilism give re-interpretations of the notion of "freedom" required so that it doesn't appeal to "doing otherwise" at all. But again, this doesn't jive with Korsgaard as you present her.

Adam Streed said...

Mike said: Determinism should only be depressing if its truth would have some kind of effect on our phenomenal experience, or the value of our relationships and lives.

On the consequent's first disjunct: I can't see why this is true. There are many things that might be true but have no effect on phenomenal experience, and yet still I ought to care: maybe I've got a disease, maybe someone I care about has recently been harmed, maybe US policymakers are right now making terrible mistakes. You might object that, unlike these possibilities, determinism could never impact my phenomenal experience. But I don't see why that should make a difference. The goodness or badness of some state of affairs doesn't turn on whether it appears to be so (except maybe in special cases, like some artworks). So the fact that I would never appear to myself to be determined doesn't make the truth of determinism irrelevant. To put it another way, the necessity of feeling free is no consolation if I'm actually determined. It just means I'm deluded, which is not itself all that reassuring, either.

On the second disjunct: the thing you're denying is exactly what determinism seems to entail. I know this is mere table-pounding, but no choices or efforts seem valuable if they're determined. Nothing seems valuable if determinism is true. You could say I'm begging the question, but I think our positions here are symmetric.

Here's the passage from Korsgaard, p. 96 of Sources:

"[The theoretical claim that we could not have done otherwise, if true], might show that we aren't responsible.[fn] But it is a different question whether determinism is a threat to responsibility. Freedom is the capacity to do otherwise, not to have done otherwise. No one has that capacity, because you cannot change the past. That sounds like a joke but I mean it. The freedom discovered in reflection is not a theoretical property which can also be seen by scientists considering the agent's deliberations third-personally and from outside. It is from within the deliberative perspective that we see our desires as providing suggestions which we may take or leave. You will say that this means that our freedom is not 'real' only if you have defined the 'real' as what can be identified by scientists third-personally and from outside."

If I understand her, Korsgaard thinks that freedom is the ability we have to take our desires as providing (or failing to provide) reasons for action. So no semantics of 'could have done otherwise' here. But I don't see how this point that freedom is a practical, not theoretical, matter is supposed to help me out. Knowing that it's a practical necessity to act under the idea of freedom is, to my ear, knowing that I'm deceiving myself.

Matt Brown said...

Adam, I think the key to Korsgaard's point is here:

"You will say that this means that our freedom is not 'real' only if you have defined the 'real' as what can be identified by scientists third-personally and from outside."

A certain kind of overzealous scientific realism would indeed define "real" in this way, but Kant, Korsgaard, and Dewey (though the later on very different grounds) would not do so. And why should we? Given all the different ways we have of being involved with the world, why is the one way that scientists are involved be the only way to get at truth?

Mike T. said...

Adam: You're right that I'm not convinced by your response on the depression issue. On the feelings thing: I guess I imagine depression as a kind of feeling, so what I thought you were saying was that, if you found out determinism were true it would make you feel bad. So while you're right that there are things that might be true but wouldn't affect your feelings. I thought you were claiming that determinism wasn't one of those things. Now it sounds like you've changed your mind.

On the value claim: I'm not sure I understand how you're using "symmetrical" here, but if I get it, you think that we're just asserting different (opposed) positions on whether determinism precludes value -- that we're both just begging the question against one another as a result. But it's totally possible that people's claims can be symmetrical and one of them can be wrong. Is there a God? Yes. No. Someone's wrong.

I don't like thinking about determinism as the inability to do otherwise, but taking this common reading on for a moment, why would my inability to avoid something valuable make it non-valuable? Maybe it would undercut the praise or blame you get for having that valueable thing, but certainly not it's value proper.

So I'm unconvinced. If you meant that determinism would make you feel depressed, I say, weird... doesn't do that to me. But I agree that it would make me depressed if it made living a life that I value impossible. But it doesn't seem to do that. I'm not pounding the table on the compatibility of value and determinism because I have reasons for believing they're compatible.

Lenoxus said...

Four months later, and the debate is still unresolved? Pathetic. ;)

Here's my two cents: To me, believing in strict incompatiblist determinism is rather like believing in solipsism, or nihilism. Both philosophies can be proved with impeccable precision, and while they may be unintuitive to many of us, they manage to grapple and remove the sort of counterintuitiveness that creeps into any part of life when we think about it too long — such as Zeno's paradoxes.

Hey, like, even if we grant the existence of consciousness, how the heck it is possible for there to be "more than one" consciousness, many of them contradicting others in their nature? Wouldn't that be like the universe having more than one set of physical laws? No! Life has to fit in a single box! It just has to, regardless of whatever cost that has for our (that is, my) deriving "meaning" from life. "Meaning?" How? What external force, what homonoculus, gives meaning to life? What then gives meaning to the meaning-givers? Are you saying it's turtles all the way down? That's impossible! Everything must either be physical or supernatural, no convenient metaphysical ground allowed!

Ah, I love writing dialogue for my straw people.

I admit that I know no coherent proof of free will. I also know no coherent way of getting by without it, so there you are. (By the way, things like free will and the existence of other minds are among the few things I take on "faith" alone, simply because I feel that the consequences of doing so are a heck of a lot better than not doing so. I don't make this sort of exception for, say, God, because I know of plenty of "substitutes" for what God is supposed to be in our lives, but none for those other things.)

I suppose, to put it simply, I don't and can't believe in hard determinism because it's so boring, like intelligent design. To my ear, "I'm hardwired, you're hardwired, we're all hardwired" leaves just as little room for humankind's incredible exploratory imagination as "God did it, God did it, God did it." At the end of the day, I don't really give a damn if it's "the Truth," any more than if "the Truth" were that the world is going to end, no matter what, in the next three minutes. I'd do my best to get on with life anyway.