Re: who's on 1st

From: Gilles HENRI <Gilles.Henri.domain.name.hidden>
Date: Sun, 25 Apr 1999 12:26:07 +0200

>> If you meet somebody who insures you
>> that he is you, would you believe him and if not, why?
>
> It depends on the definition you use. If he is implementing the
>same conscious computation as me, then he's the same as me in that sense.
>(And this is the one that's involved in the duplication thought experiment.)
>But he's not me in the sense that there are still two of us. If he's just
>implementing ones that are very closely related to mine, such as my future
>self would do, or my counterparts in other MWI 'worlds' would do, then
>he's close enough to me for many practical purposes. For legal purposes
>(e.g. property rights), he's not me.

I agree basically that the fact that two implementations are considered as
"identical" depends on the definition of identity that you assume (as I
said previously, you could "assume" that you are identical to your twin, or
any human being, or even an animal following what you mean by "close enough
for many practical purposes"). But "COMP" is (if I understood it correctly)
a stronger hypothesis: it is that at some finite level, you could reproduce
or duplicate EXACTLY your conscious state, or at least you could simulate
it "to an arbitrary degree of accuracy" (which is already somewhat
different!) (James):

>that's necessary. I built an arificial neurone out of integrated circuits in
>1984; I have no doubt that if I had enough of them I could simulate myself
>to an arbitrary degree of accuracy. To deny this is to bring in sprit by
>another name.
>James

I apologize to insist, but what I ask is: what do you mean exactly by that?
Jacques, how do you define " If he is implementing the
>same conscious computation as me?", James, which features of yourself
>would you try to simulate "to an arbitrary degree of accuracy", how long,
>and how would you check it?

I wouldn't like to appear to be too fussy. However, it seems that we
discussed a lot of paradoxes or original theories based on the possibility
of "duplicating" a conscious being, and that it deserves a careful
definition of what we mean by that. As a matter of fact, theories like
relativity and QM have emerged only when somebody began to ask : how would
you PRACTICALLY measure that? (simultaneity of events, or position and
momentum of the same particle) and realized that it was not THAT obvious.

Again the reason why it doesn't seem THAT obvious for me is that
consciousness in the usual sense does exist only because we have a
representation of ourselves as "subjects", and that this representation
requires a particular place in space and time (whatever these things are).
If you think that you have built a neural network almost identical to
yourself, you know that YOU have built A MACHINE. But what does YOUR
MACHINE know?

Again I'm not saying at all that we are made with a "spirit" of any sort. I
mean that the computation that makes us conscious is so peculiar that it
can't be anything else than unique in a given "classical" world, because
the analog input it uses (our physical environment) is unique at the place
and the time we occupy. And if you succeed in building an intelligent
machine (which IS possible in my opinion), this machine SHOULD know that it
is a machine, and hence that it is not you, exactly for the same reason why
you know that you are you, and not anybody else.

Gilles
Received on Sun Apr 25 1999 - 03:32:39 PDT