November 1, 2004

Does determinism matter?

Hi all,

Had a thought today about determinism in P&E that I want to explore. In light of Heidegger's da-sein, can we put the determinism we get out of physical reductionism into the same place as external skepticism? That is, maybe we're brains in vats or being toyed with by an evil demon, but so what, it doesn't feel that way so it doesn't really affect our lives at all. Similarly, if we somehow found out tomorrow that our lives and actions and decisions ARE reducible to physical principles and everything is determined, it wouldn't change the way we live our lives since it never at any moment FEELS as if that is what is going on. I for one would still care about all the same things and still make conscious decisions and still have the same goals. I would still "take care" in the world, as Heidegger would say. For that is what it is to be human. If we lost this condition of "taking care" we could no longer be classified as human beings (da-seins).

Now, this is all being said under the assumption that we somehow discovered determinism was true without being able to use it to know in advance everything that was going to happen. Naturally, if we did we WOULD make some changes in our cares and decisions and goals. For example, if we could foresee that Harry was going to commit some heinous crime next week, we might find it in our power to prevent this from happening, but problems would arise immediately here.

One, we would also foresee that we try to prevent Harry from committing the crime, but he must somehow commit it anyway if we foresaw it. But then why go through the trouble of trying to prevent it? Or, if we DID successfully prevent it then what we foresaw was wrong. But to be fair to determinism and escape the immanent paradox, we never would have foreseen him committing the crime, we would have foreseen ourselves foreseeing the crime and preventing it, but not actually foreseen the crime happening.

Does anyone want to pick up on the next step here? Does this make sense or sound interesting? I know it is a lot like "Minority Report" but I've never seen it. No time to continue this at the moment, but I want to expose that there is still a paradox here, and also that moral issues are raised (as they always are when determinism is involved).

Labels: , ,

6 Comments:

Blogger Winston said...

This comment has been removed by a blog administrator.

11/02/2004 01:53:00 AM  
Blogger Dub! said...

I agree with Winston on the impossibility of a Laplacean demon... it follows from Turing that Laplacean demons can only exist if they are outside the universe they are observing and cannot have any causal relationship with that universe.

That being said, Heidegger would probably not think these arguments are relevant to his study of Dasein one way or the other. While his Dasein is 'in the world', it's not in the world in the same way that objects of sceintific study are. It's this 'kinda-alongside-the-world-but-not-really' feature of Dasein that I personally don't get. It feels to me like a kind of crypto-dualism. I have a hunch that what it comes down to is the fact that Dasein is essentially a normative force, and on Heidegger's view, we can't naturaize normativity. I disagree with him on that point; but it seems like that's what's at stake.

On the subject of The Minority Report... this seems like a good place for me to gripe about how upset I was about that movie. The Phillip K. Dick short story had a really interesting and consistent take on deterministic prediction and how the very act of prediction could influence the future. Read it! The movie, however, screwed up all the interesting philosophical implications of the story. Here are three features of the movie's take on 'fate' that I really didn't like: (spoilers)

1) The precogs could be wrong, and in pretty serious ways. They weren't really deterministic predictors at all, which allowed for all the mushy talk about choice and free will at the end. If we just know that the precogs are good predictors without knowing how they work, then they doesn't have any special philosophical import. In the story, the precogs were *always* correct except in very specific interesting and principled cases.

2) The precogs tell the future through images, as if they have little video cameras flying around through the temporal universe. I wouldn't complain about this if the movie didn't have a plot point hinge on the fact that *only* visual information was available to the precogs (when the villain disguises himself in order to frame someone else). That seems ridiculous. It also might contradict the fact that the machine spits out little balls with the criminal's name on them. Where is it getting the name from? An analysis of the foreseen visual image? When the villain disguises himself, he just puts on different clothes but the Precrime unit presumably got a little name-ball with the framed person's name on it. You can't tell me that Precrime would identify future criminals merely based on the clothes they wear!

3) The precogs make predictions about futures that could only exist if that future were foretold. For example, the kidnapped precog sees that Tom Cruise will get away if she tells him to give the bum some money, so she does so. There's nothing wrong with this in itself, but it really highlights the differences between the movie and the book.

Also, the title "The Minority Report" made great sense for the short story. It was a useless title for the movie.

11/03/2004 09:31:00 PM  
Blogger Winston said...

I want to explain my view a little bit more, because I might not have been clear: I think that the Predictor/Lapacean Demon can exist and interact with its own predictions, but what I was tring to point out was that there is a limit to that ability. I don't see a problem with a predictor that does a "what if" prediction like this: you're standing under a very large wrecking ball hanging from a crane, and the predictor has the option of either raising or lowering the ball. You ask it to do whichever will lead to your survival. The predictor could test what would happen in each case and then pick the right option: to raise the ball. Here the predictor is interacting with the world it's making a prediction of without a problem.

The problem is that it can be forced into situations where it cannot give a correct prediction, such as the one where it turns on light A if it predicts it'll turn on B and vice versa, or where you ask it to predict if you'll press button X or Y, and then you deliberately press the other. If, by definition, a Lapacean Demon can always predict the future, then I guess there can be no Laplacean Demons that interact with the world. But there can still be a predictor, with limitations.

11/03/2004 10:23:00 PM  
Blogger Dub! said...

I wrote Turing in my first comment when I meant to write Godel. Oops.

Winston,

Does the predictor know for certain that choosing to raise the ball will save you? No. It's possible that in the process of computing what the future holds, an errant logic switch in its circuit board will throw off an electron that will cause some some wild Rube Goldberg physical chain of events, making the world such that choosing to raise the ball will cause the chain to snap and kill you but choosing to lower the ball will do nothing. It's *highly* unlikely, but unless the predictor has access to its own physical states, it has to use probabilistic induction to rule out that case. And then we're not talking about a deterministic predictor/Laplacean Demon at all. It's just a very good predictor.

I don't think your second example works. It'll give *you* a false answer because you're both forcing it to communicate and limiting the ways in which it can communicate, but it'll still know what the future holds either way. So it *is* predicting, in a sense. It's just lying to you. It's keeping its real prediction a secret. (If you limit an omniscient computer to only be able to output facts about baseball, that doesn't mean that it isn't omniscient).

None of this stuff really touches Daniel's humanistic concerns.

11/04/2004 05:31:00 AM  
Blogger Winston said...

Richard -

I disagree on the wrecking ball case. It seems that your saying that there could be an errant electron or something like that in the predictor, but it can't know that except through probabilistic reasoning. I think this is false. Suppose you have two perfectly formed spheres formed of some element, where the atoms are in a perfectly arranged lattice. Then if you want to predict how gravity is going draw them together, you can effectively compress this information by saying that there's a center of mass here and another one there, and thereby predict their motions, perfectly, with no randomness at all. You're just throwing out information that cancels out (this hints at Jessica Wilson's talk last week).

Similarly, there's no reason that a predictor couldn't have access to a summary of its own state from which it could deduce that it's working properly and will continue to do so. This is just a straightforward reduction. Since we're assuming a deterministic world, I'm not going to worry about random quantum effects, which throw a monkey wrench in all of the thought experiments so far.

You might object that the predictor might not be able to tell if there was an errant electron that distorted its deduction of whether it's working properly. I agree that it's possible. But if you press this line, I'll respond that your Laplacean Demon is vulnerable to the same problem -- it could make an error and not know about it. So if you're going to talk about the possibility of mistakes, then you first have to solve it for the Demon before you come after the wrecking-ball predictor. It seems to me that if your Demon can legitimately believe it's making no mistakes, then so can my predictor.

My predictor may work in a different way from your Demon; that is, by doing a reduction (when possible) instead of calculating the path of every particle, and if that makes it not a Demon, then I have no quarrel with that. But it's still a predictor.


As for the second example: I'll agree that you could have a predictor that lies to you, but I take it as essential to Daniel's paradox that it always tell the truth about its predictions. If you assume it can lie to you, then there's no infinite-regress paradox of the sort Daniel was proposing, just a problem of whether to believe the thing or not.

Yeah, I agree that this has gotten way off the topic that Daniel posted... I wish I could comment intelligently about the human problem of interpreting determinism (or perhaps determinism plus randomness) versus free will, but I don't have much to add. It seems like this is also a big problem for philosophy of mind. Anyone feel like making a new post about this question in particular? I'm interested to see what everyone has to say about it.

Footnote: Here's a somewhat more down-to-earth example of a system that checks if all of its future caculations will be error-free. Suppose you have a computer program stored in memory, and that each bit has three possible states: the normal states 1 and 0, and the error state (akin to the errant electron) is 0.5. If they're all ones and zeros, then the program can be certain that all its future states will be error free. It could sweep through its entire memory, bit by bit, and upon finding all ones and zeros, it could conclude that all its future states will be error-free. This is the sort of reduction I mentioned before.

As I noted earlier, this doesn't save the program from skepticism about the correctness of its own error-checking, but then, the Demon is just as vulnerable to this self-skepticism. You might object that the program can't tell if the hardware it's running on is error-free, but that's irrelevant. The abstract system is itself meant to be taken as the deterministic "universe," not the physical realization of it on an actual computer. I think that if we're going to assume a deterministic universe, then this abstract formal system is a perfectly acceptible model of it. Picture Conway's Game of Life -- that can serve as a model of an abstract system which is a deterministic universe.

11/04/2004 04:57:00 PM  
Blogger Dub! said...

Winston,

Because we've gotten lost sight of Daniel's topic, I'll post a response to your comment in the comments section of the post you made.

11/04/2004 06:06:00 PM  

Post a Comment

Subscribe to Post Comments [Atom]

Create a Link

<< Home