Re: against computationalism

From: Gilles HENRI <Gilles.Henri.domain.name.hidden>
Date: Mon, 2 Aug 1999 17:06:32 +0200

CM:
>
>> comp2 contradicts comp1 because the essence of comp1 is the independance of
>> the results with respect to the material implementation, whereas comp2
>> requires the precize definition of this implementation (of course the
>> simulation made in comp2 could be runned on any TM, but the object of the
>> simulation must be one precise physical system).
>>
>
>I don't see that as necessarily true. Couldn't comp2 be an ensemble of
>computations, each of which represents one "universe", or one set of
>physical laws? We are interested in that ensemble which contains SASs
>that can be mapped to ourselves. So we can be existing in multiple
>universes simultaneously.

GH:

In different universes, the exact definition of "we" is somewhat subjective
(see previous mails) but it doesn't matter.
Of course if you accept comp2, each computation can be "mapped" to a
Universe, (however you have a big uncertainty in how to do this mapping,
because all simulations of a physical Universe assumes a preexisting
physical law, that was my point 1). But now comp1 is false, i.e. the brain
is NOT like a computer, storing ALL the necessary information at a much
higher level than its physical structure. It can NOT be duplicated by
another physical structure (or if comp2 is true, by a computation coding
another physical structure).

CM:
>
>Even if not, I don't follow your reasoning that comp1 is contradicted.
>When you say "of course the simulation made in comp2 could be running
>on any TM", I understand that to be thing in comp1 that you are saying
>is contradicted. So where's the problem, exactly?

GH:
see above. I repeat: the necessary information is stored in the entire
physical structure (analogic, or digitally modelized) which is the contrary
of comp1 (as I have expressed it).

What I stress here is that comp2 goes much farther than the idea that you
can make a digital computation on any platform. When you let a normal
computer run, you have basically two levels of information : the
information I1 stored in its memory bits (a few tens of Mo, say) and the
physical information I2 in its physical structure, which is enormously
larger, but irrelevant for the most part for the computation.

The information stored in I1 is enough to determine exactly the
computation. The physical information I2 is more tricky. If you calculate
it by the physical entropy, you will get around the number of atoms 10^23,
but this physical information is limited by the quantum uncertainty, i.e.
you forget any detail beyond the Heisenberg inequality. This is perfectly
compatible with the impossibility of measuring the quantum state. However
if you want to calculate deterministically the quantum evolution, you must
describe the wave function with much more accuracy, although this accuracy
is not measurable experimentally! For example if you want to describe the
wavefunction at a Planck scale length, you need a factor (l(de
Broglie)/l(Planck))^3 = 10^75 (say) more information, although this
information is completely hidden in reality.

There is so a fundamental difference between digital computers, for which
I1 exists and is self-consistently determining EXACTLY its evolution, and
so can be duplicated independantly of I2, and analogic ones where I1 can
exist but is insufficient to predict its evolution (in this case, I1 is in
general observer dependant, it is only a partial information ,e.g. the
speed of the wind and the temperature of the atmosphere at selected points,
etc...) The required information I2>>I1 necessary to predict the real
evolution is
a) depending on physical laws
b) hence still unknown (we don't know the physical laws of the Universe)
c) unlless I1, probably far beyond all possible measurements due to the
quantum nature of the world.

* comp1 states only that the brain (more exactly here the consciousness)
implements EXACTLY a digital computation as defined above, which can be
obtained by any machine with an equivalent information I1, whatever its
physical information I2 is (like computer programs).

*comp2 recognizes that comp1 is false. Note that there IS actually a low
information content I1 in the brain ("low" means here of course much lower
than the quantum detailed information. It is still higher than all what our
computers can store!), which determines our conscious state, our memory,
etc... that does NOT appear to be very sensitive to the microscopic state
since our neurons are constantly changing, dying, etc...without strong
changes in our thought. However, it is not "digital" enough to be exactly
predicted at this level.
 comp2 assumes that
a) there is a level I2>>I1 for which the evolution of I1 can be exactly
reproduced.
b) ALL physical properties (or at least, consciousness) are conserved
through a simulation of I2 with a digital machine (whose physical
information I3, of course, is much larger than I2 !!)

Do you agree that it is a very strong assumption,which is not obvious at all?

For example if I do a simulation of the sun in my computer, I can reproduce
rather exactly the temperature, luminosity, pressure etc... of its
interior, but my computer will not trigger a neutrino detector !! Why
should a digital simulation of the physical state of my brain, or best of
THE WAY I AM REPRESENTING the physical state of my brain, have the same
"intrinsic" properties as the brain itself?

CM:
>
>I disagree completely with your analysis in the last paragraph. The
>Chinese room thought experiment, as well as Maudlin's argument, are
>entirely bogus. In short, let me try to explain why. Let me draw an
>analogy between the common-sense view of physical reality and your
>"library" example. In the former, there are physical laws which govern
>the motion and interaction of innumerable sub-atomic particles in my
>brain, which map in some way into a computation.

GH:
I disagree also. The motion and interaction of sub-atomic particles are not
THE physical reality. It is one way by which we can represent
(approximately) the reality. BTW in Quantum Field Theory it is extremely
hard to imagine the quantum state as a collection of "real" particle, and
we have definitely no theory to represent all particles + interactions
including gravitation. The reality itself is unknown. Taking the physical
laws for the reality is just as taking the characters of a movie as real
persons- physical laws are just a little bit more accurate.

CM:
 They map as surely
>as a macroscopic digital computer maps -- but in a much more complicated
>and diffuse way.

GH:

No, for the last reason. A REPRESENTATION of reality can be mapped (in
general approximately) by a computation, not the unknown reality itself. A
digital computer is built to implement exactly computations (without
approximations, except residual bugs in Pentium...), and so cannot be
exactly equivalent to reality.

Note that Bruno's conclusions is that the reality does not exist at all !!!
I think it exists but it is beyond any representation.

CM:
>
>In the latter (your library example), you have to consider the entire
>system - library, books, you, the food you'd eat - everything. Then you'd
>find that that entire system maps onto a computation by virtue of the
>actions you'd take in performing your job. So I'd say that the system as
>a whole is conscious. It might be a ridiculously precarious consciousness -
>as you say - you might not be very reliable in doing your job. But that
>would, I'd suggest, be analogous to (in the common-sense physics view)
>having a fatal disease that could kill me at any second.

GH:
You did not answer me on what I have actually to do for this job for you to
live!!

CM:

>
>
>> A third and last difficulty is that comp2 does not solve the problem of
>> consciousness. For if everything is assimilable to computations, what makes
>> some computations or parts of computations conscious or not (see Wei)? So
>> how to found Bruno's "computational psychology" ? What is the dream of a
>> string? Complexity is not enough, because for example the chemical
>> evolution of a thinking brain is not more complex to that of a dead brain
>> leading to putrefaction. At the analogic level of comp2, you have lost the
>> information level of comp1 ! I agree that the problem is the same with
>> materialism - I just point out that it is not easier with comp2.
>
>Asking a rhetorical question, "what makes some computations or parts of
>computations conscious or not," is not the same as a refutation. I don't
>see how you have presented any problem about comp2 in this preceding
>paragraph.

GH: Here I just stress that asking "what makes some computations or parts of
computations conscious or not," is a priori not simpler than asking "what
makes some physical systems or parts of physical systems conscious or
not,", so comp2 does not simplify this problem. I see some unconsistency in
thinking that consciousness can be produced by computations of physical
laws that do not exist!

CM:
>
>> So I think that pure computationalism, either comp1 or comp2, is very hard
>> to maintain. Another comp3 proposition?
>>
>> One remark is that all "thinking" devices based on digital simulations of
>> the analogic state of the brain handle in fact much more (and too much)
>> information than the brain itself, which is totally unaware of its own
>> material structure. A very important fact is that they ALL require an
>> external structure able to store the relevant information and program them
>> adequately, which is NOT the case of actual brains (and I guess of possible
>> future thinking machines).
>
>What!!??? Of course the program being run by a brain requires an external
>structure - the brain! What do you mean by this last paragraph?

GH:
I meant here that the physical laws governing the activity of the brain are
the same that those governing its very formation, and the brain is
"building itself" without any OTHER pre-storage of its information content.
I suspect it is a requirement for a conscious system to actually
work...(not proven of course!)

Thank you for answering my provocative mails!


Gilles

******************
Gilles HENRI

Laboratoire d'Astrophysique
Observatoire de Grenoble
414, rue de la Piscine-BP 53
F 38041 GRENOBLE Cedex 9
FRANCE
Maître de Conférences à l'Université Grenoble I (Joseph Fourier)
Membre junior de l'Institut Universitaire de France

Tel (33)-0-476-51-42-11
Fax (33) 0-476-44-88-21
mailto:Gilles.Henri.domain.name.hidden
******************
Received on Mon Aug 02 1999 - 08:11:42 PDT