Bruno Marchal wrote:
>Le 19-juil.-06, à 17:30, Jesse Mazer a écrit :
> > Stathis Papaioannou:
> >> Bruno Marchal writes:
> >>>> I think I have more basic difficulties also, like the Maudlin
> >>>> argument re the handling of counterfactuals for consciousness to
> >>>> occur:
> >>> It is a bit harder, no doubt. And, according to some personal basic
> >>> everything philosophy, the Maudlin argument is important of not ....
> >>>> is this requirement just to avoid saying that everything implements
> >>>> every computation?
> >>> Jacques Mallah makes that point some years ago (in this list), and I
> >>> think Hal Finney has developed that point. I think their argument are
> >>> valid. But then I don't think the Putnam-Mallah-Chalmers is really a
> >>> problem once you get the idea that the physical world emerge from the
> >>> mathematical world of computations. Personally I have never seen a
> >>> convincing argument that everything implements every computations,
> >>> just
> >>> perhaps some tiny part of some computations.
> >>> I will postpone saying more on the movie-graph/Olympia type of
> >>> argument
> >>> (if only to avoid to much simultaneous threads and to modulate the
> >>> difficulties).
> >> It seems to me trivially obvious that any sufficiently complex
> >> physical
> >> system implements any finite >computation, just as any sufficiently
> >> large
> >> block of marble contains every marble statue of a given >size. The
> >> difference between random noise (or a block of marble) on the one
> >> hand and
> >> a well->behaved computer (or the product of a sculptor's work) on the
> >> other
> >> is that the information is in the >latter case presented in a way
> >> that can
> >> interact with the world containing the substrate of its
> >> >implementation.
> >> But I think that this idea leads to almost the same conclusion that
> >> you
> >> reach: it really >seems that if any computation can be mapped to any
> >> physical substrate, then that substrate is >superfluous except in
> >> that tiny
> >> subset of cases involving well-behaved computers that can handle
> >>> counterfactuals and thus interact with their environment, and we may
> >>> as
> >> well say that every >computation exists by virtue of its status as a
> >> platonic object. I say "almost" because I can't quite see >how to
> >> prove it,
> >> even though I suspect that it is so.
> > But just because you can map any physical activity to any computation
> > with
> > the right mapping function, that doesn't necessarily mean that some
> > physical
> > processes don't contribute more to the measure of certain
> > observer-moments
> > than others--Chalmers would say that there are "psychophysical laws"
> > governing the relationship between physical processes and conscious
> > experiences, and they might specify that a physical process has to meet
> > certain criteria which a rock doesn't in order to qualify as an
> > instantiation of a given mind.
>Those specifications have to make physical processes NOT turing
>emulable, for Chalmers' idea being coherent. The price here would be an
>explicit NON-COMP assumption, and then we are lead outside my working
>hypothesis. In this way his dualism is typically non computationalist.
Why would Chalmers' version of dualism be non-computationalist? As I
understand him, he does argue that there is a one-to-one relationship
between computations and conscious experiences, and he certainly believes
that a sufficiently detailed simulation of a brain would *behave* just like
Anyway, without tying my argument to closely to Chalmers' beliefs, what I
meant when I talked about "psychophysical laws" was just a rule for deciding
when a copy of a particular computation has been instantiated physically,
with each instantiation contributing to the total measure of that
computation. You don't even have to postulate a special "physical universe",
you could just ask how frequently copies of a smaller computation are being
instantiated in a larger computation (like a computation representing the
evolution of the universal wavefunction, or a computation representing the
universal uovetailer). For example, let's say I identify a given
observer-moment with a particular computation O which represents all the
computations going on in the observer's brain during that moment (with a
'moment' of subjective experience presumably corresponding to computations
spread out over tens or hundreds of milliseconds in the physical brain)
which are relevant to what the observer subjectively experiences in that
moment (and there might be plenty of things going on in the physical brain
that *aren't* relevant, like random thermal vibrations of atoms in neurons).
Suppose I also have a larger computation E which is a detailed simulation of
a physical environment that happens to include a brain that seems to be
doing what appears to be a similar set of computations; in this case the
"psychophysical laws" would be some rules that would tell us whether the
larger computation E does in fact contain an "instantiation" of O within it.
And if we postulate some ultimate base-level computation representing
"objective physical reality", like the simulation of the universal
wavefunction or the universal dovetailer, this might contain instantiations
of every possible program, but it might instantiate some more frequently
than others, thus giving some observer-moments higher measure than others.
Finally, even if we reject the idea of a single base-level computation, the
idea of larger computations containing instantiations of smaller ones within
them might still be relevant, since if you identify observer-moments with
particular computations, some observer-moments might contain copies of
smaller ones within them, and this would have to be taken into account in
the sort of "measure battle" that you talk about in other posts.
So, do you think that the idea of "psychophysical laws" whose sole purpose
is to decide if larger computations contain instantiations of smaller ones,
and therefore contribute to their overall measure, violates
"computationalism" as you understand it? If so, why?
>I met David Chalmers in Brussels in 2000 (at the Brussels ASSC
>meeting). He *is* indeed quite coherent, in the sense that he considers
>that in the self-duplication Washington/Moscow experiment the first
>person must feel to be at the two places simultaneously.
>This is coherent also with his dualist interpretation of Everett. Now,
>I personally agree with Hans Primas, and David Deutsch, that Everett's
>move is motivated by a search for a monistic view of (quantum) reality.
What do you mean by "feel to be at the two places simultaneously"? As in, I
would have a single subjective experiences of looking out of two pairs of
eyes in two bodies, even though each of the bodies acts as if they have no
knowedge of what the other body is experiencing? This seems at odds with his
general ideas about subjective experiences tracking with actual third-person
reports of one's experiences, which play a critical role in his "dancing
qualia" argument, for example (see http://consc.net/papers/qualia.html ). If
he didn't explain in detail what he meant, is it possible you misunderstood
him? Perhaps he just meant that both copies would have equal claim to be
later versions of the original, not that they would have any subjective
knowledge of each other's experiences.
> > Although there is some difficulty figuring
> > out exactly what these criteria would be (matching counterfactuals, for
> > example?), it doesn't seem obviously hopeless,
>Only by jeopardizing the comp hyp., and introducing an explicit dualism
>(as he does).
>I have no problem with that. I respect all hypothesis, but I
>concentrate myself on the comp hyp.
>I only rarely argued in favor of comp, or of any hypotheses. I prefer
>to study their consequences. Now, I could say that I find Chalmers
>approach as an highly speculative approach build just to save the
>Aristotelian conception of physics/nature/reality, which I already
>known to be incoherent with comp.
>The point is that Chalmers approach is coherent with mine: he just
>proposes a different theory.
Well again, leaving aside the details of Chalmers' ideas, do you think there
is anything non-computationalist about the type of "psychophysical laws" I
suggest above, which are solely concerned with the question of how larger
computations contribute to the measure of smaller ones by containing
instantiations of them? Such laws might very well say that the movie-graph
or the ticking of a clock does not qualify as a valid instantiation of some
complex computation corresponding to a particular observer-moment.
> > which is why I'm not ready to
> > accept Bruno's movie-graph argument or Maudlin's Olympia argument.
>OK, but my feeling is that you need to abandon comp to be able to cut
>down such form of reasoning. Or you should perhaps point on some
>precise step you judge not convincing.
Can you repost that link to the steps of your argument in order? Thanks.
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to email@example.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at