Marchal wrote:
>
> Hi all,
>
> Great number of interesting and deep posts. I will try to comment some
> now (actually I will answer only Chris), and others later.
> For exemple SLP and Hans' discussion on
> consciousness is very interesting but bears on difficult questions that
> I want not to dwelve into now if only because I want to think
> a little more ...
> For a little "quasi-objective" theory of consciousness, look at
> my (godelian) chapter 5.
>
> >Christopher Maloney (about my post to "Romanos Avaqian"):
> >Bruno, when I read your posts, I feel small! Thanks for this,
> >you have swelled my reading list yet more.
>
> I don't intend to make you feeling small and
> I am sorry if I have donne so :-(
>
> I feel that Romanos Avaqian's question is a very fundamental one.
> Fair enough, Romanos' question "can the subject become an object for
> himself?" gives me an opportunity to give, not *my* answer but the answer
> any one will get from an interview of a self-referentially correct
> machine,
> modulo the "theetetic definition of knowledge". (An *avant gout* of my
> chapter 5, which is self-implying independant).
>
> The anwer is no. No subject can recognize itself as an object.
I'm not sure what you mean by this exactly, but I suspect that I
disagree with it. Let me reproduce here what I recently wrote in
a reply to GSLevy:
For any intelligent, information processing system, "thinking about
something" involves making internal models of some (perhaps external)
apparition. The models are necessarily elided, or simplified.
While it might be possible for an AI to know, in principle, all
of its own internal workings down to the lowest level,
it would nevertheless be impossible for it to be able to "simulate
itself". Using a simple von Neumann description, a perfect fidelity
self-simulation would involve emulating its own program, _and
duplicating all of its internal data_. The latter is obviously
impossible.
So when we're thinking about ourselves, we're just manipulating and
mangling a model, the same as any other model of any external object
like a cat or a poem.
>
> But I should mention you can get the same answer "in a self-implying
> way", by using a thought experiment in which you imply yourself both in
> first
> person (subject) view and in third person view like your own duplication
> thought experiment (see your URL) which shows, among other things, that
> duplicated people cannot recognize themselves in their respective doubles
> (doppelgangers).
> A machine can only BET that "itself" is an object from another machine's
> (first person) point of view. Like a patient relatively to his surgeon,
> or a
> computationalist (practionners) in front of his classical
> teletransportation
> machine.
I lost you here. What do you mean that "a machine can only _bet_ that
"itself" is an object from another machine's perspective"?
> This reflection is also linked to Lucas-Penrose misuse of Godel's theorem,
This actually sheds some light on your thinking to me. I agree that
Penrose has gotten all screwed up in applying Goedel's Incompleteness
theorem to consciousness. This is another example (I think you all
remember a prior one for me on this list) where I encounter an otherwise
intelligent person holding onto a concept that is utterly ridiculous.
So let me see if I'm understanding this correctly. Penrose says that
consciousness cannot be the result of computational processes because
there will always be true statements which a computational process
would find undecidable (Goedel statements for that process). But
humans,
on the other hand, would be able to "see" the truth of Goedel-esque
propositions because of our amazing, non-computational intuition. But
the counter to this argument is that:
1. if a particular human's consciousness is the result of a
computable process,
2. there will be Goedel statements for that particular human, but
3. they will be of such a complexity that the human would not
possibly be able to decide the truth or falsity without the aid
of a computer, which,
4. changes the computable process into a human+computer system,
5. rendering the original statement decidable in the new system.
When you say that this is related to the topic above, I think you
must be talking about the inability of any machine to fully model
itself. Is this right?
> and Benacerraf wonderfull and super-important paper (although wrong from
> the
> very beginning to the very end as Benacerraf himself recognizes): 'God,
> the devil and Godel', the monist 51, 9-32, 1967.
>
> Concerning my swelling of your reading list. I'm afraid it's only
> a beginning. The mind-body problem and the everything stuff belongs
> to a new (not yet named) field which is at the intersection of all fields
> !
>
> >Christopher Maloney (to Steve Price) :
> >Ugh, I don't even want to *talk* about zombies. I find the whole
> >concept patently absurd, and in a way, ethically repugnant. By
> >that I mean that it seems to me awfully grandiose and anthro-
> >centric to even presume that some other creature that exhibits
> >all the traits of consciousness and thoughtfulness, could some-
> >how not be conscious or thoughtful.
>
> I have the same ethical aversion of the zombie notion, Chris, but ...
> But sometimes for logical purpose it is useful to go beyond your
> feelings. As a matter of fact the notion of zombie IS very useful
> in philosophy of mind, ... if only to address the "consciousness"
> problem. I agree with Steve Price's anwer to your post here.
Well, I replied to him offline. We have to choose our battles,
nicht wahr? I was just stating an opinion, and I was not trying
to be scientific one way or the other. It's my perogative to opt
out of a debate whenever I want.
> For exemple, some philosopher of mind take the zombie as an absurd
> concept (like you and me, indeed), SO if someone argue that a theory X
> leads to the existence of zombie, she gives a kind of reductio ad
> absurdum against the theory X.
>
> As an illustration I think zombies are unavoidable for those who
> defends the De Broglie-Bohm interpretation of QM. (Hint: AI and quantum
> computing, you can help yourself with Deutsch's argument that MWI is
> experimentally testable, you can find it at the Michael Clive Price FAQ
> http://www.hedweb.com/manworld.htm#detect).
>
> In fact those "De Broglie-Bohm" zombies are even incorporeal !
> They are not made with particules !
Okay, let me again try to see if I understand where you're coming
from. This is related to your crackpot proof that "not comp" or
"not phys-sup", correct? You accept comp, which means that you
reject phys-sup, and De Broglie-Bohm QM, with "hidden variables",
implies phys-sup? And that would be because a hidden variables
version of QM would be deterministic, with just one "branch", and
would not produce consciousness, because it does not implement
counterfactuals?
If I'm on the right track here, then I'm afraid I'm diverging from
you on the phys-sup issue. I've been thinking quite a bit about
Hal's excellent summary of the Maudlin paper. I haven't gotten my
butt down to the library yet, but I will. I think there's a
hole in this line of reasoning, but I can't put my finger on it
precisely yet.
> Nevertheless most of them acts like us, including long talk on
> consciousness and bodies, they makes love and children, and all this,
> without consciousness and without
> bodies. And all these uncorporeal bodies and unconscious talks interfere
> with our real quasi-newtonian-like unique universe, full of particules !
> Surely you were joking mister Bohm !
> It is indeed my dislike of the idea of zombie which makes me sceptical
> about any no-wave-packet-reduction-but-single-world interpretation
> of either QM or COMP. Bohm's theory is a kind of cosmic solipsism and
> "zombie" are useful for making that statement clear.
But I don't see how phys-sup implies zombies. People who adhere to
phys-sup, I would suspect, don't think that it produces zombies.
That's just your conclusion. And how do you justify it, from first
principles? Even if I accept your crackpot proof (just kidding about
that, of course) why shouldn't I accept phys-sup instead of comp?
Finally, thanks for all these book recommendations. I will
definitely pursue this line of research. Hell, I'm a programmer,
one would think I ought to know something about the science of
computability. I am anxious to look over the Raymond Smullyan
books. I know him from two short essays of his that appeared
in "The Mind's I", both of which were excellent. His writing
is lucid and imaginitive. Very reminiscent of Jorge Luis Borges.
(Any Borges fans here?)
>
> There are a lot of excellent introductory book. You force me to make
> difficult choices.
> I guess that the best book on logic and computability, including
> the modal use in godelian provability, is
>
> Boolos G. S. and Jeffrey R. C., COMPUTABILITY AND LOGIC, 1989 (3d
> edition),
> Cambridge University Press.
>
> A classical very nice and little introduction to theoretical computer
> science is:
>
> N. J. CUTLAND, Computability, an introduction to recursive
> function theory, Cambridge University Press, 1980 (repr. 1983).
>
> A very good introductory text to mathematical Logic + Computability
> is also:
>
> MENDELSON Elliott, Introduction to Mathematical Logic, 3d ed. 1987,
> Wadsworth & Brooks Inc..
>
> But relatively to this list (and my thesis!), I think Boolos & Jeffrey is
> the best you can find.
>
> I urge everyone interested in my chapter 5 to read Smullyan penguin
> book "Forever Undecided". It is one of the best simple introduction
> to "self-referentially correct" reasoners.
>
> I guess people in this list will also appreciate a lot his "the tao is
> silent" and "5000 B. C". In this last book he uses
> the "dream" concept to illustrate a relativistic approach to many-reality
> and immortality. It is better than Platon !
> I discover this recently. More on Smullyan later. (Without Doubt).
>
> BTW, people, do you know the book by Jozef Gruska on Quantum Computing.
> I find it very good ! Even on the philosophical level !
> (Quantum Computing, McGraw-Hill, 1999).
>
> In my next post I will answer/comment to Jacques M Mallah, Devin Harris,
> Hal Finney, and Wei Dai ...
>
> Bruno.
--
Chris Maloney
http://www.chrismaloney.com
"Knowledge is good"
-- Emil Faber
Received on Wed Jul 14 1999 - 21:02:09 PDT