Gilles Henri wrote
>I suspect that the comp hypothesis would in fact favour the solution where
>there is actually no external world at all, but only your (for me, my!)
>mind, because it is much shorter to describe ONLY a brain state than the
>whole Universe surrounding it, although perfectly equivalent regarding our
>sensations. So no world, no friends and no physical laws. We would be back
>to some kind of solipsism, which has been known for a long time to be
>unprovable and undisprovable- just desperately useless.
The comp hypothesis favour indeed a solution where there is no external
world at all, nor even any material (whatever it means) brains.
But that is exactly why I say:
"We can find ourselves ONLY in those structures which are relatively
numerous, not to survive but to REMAIN WITH OTHERS and a (relatively)
In short the computationnalist (objective) idealism protect us from the
solipsism (subjective idealisme).
In fact it is not true that the description of ONLY one brain is much
shorter than than the description of the whole Universe surrounding it,
because DeWitt-Wheeler equation, or any UD (universal dovetailer
algorithm) are compressed description of all possible universes and all
possible self-aware substructures in there.
Now, it is a quasi-theorem that any such computationnal substructure, if
and when they look into their (inexisting but apparent) environment,
BELOW their own level of description, will see Many Universe or Many
Histories, indeterminism, entangledness, etc.
In this sense, computationnalism "explain" some, if not most, quantum
>However it is not obvious at all why such a mind would require the
>appearance of physical laws. Most things are too complicated to be
>described by precise laws, although they are handled rather properly by the
>mind: feelings, language, and almost all human behaviours do not obey any
It is a question of level. I believe it is quite plausible that human
beeings obey schroedinger equation at a low level, but obey to "vague
folk-psychology" at a higher level.
Another exemple is arithmetical truth. The only theory which describes
correctly and completely arithmetical truth is ... arithmetical truth
itself. There doesn't exist axiomatisable theory of arithmetic.
This transform your argument in favour of comp.
>It is perfectly CONCEIVABLE that physical
>experiments could show that matter behaves like human beings, without any
>good mathematical description.
Yes. It is even a consequence of comp. If comp is correct then comp is
undecidable, then non-comp is consistent with comp. Some machine will
consistently argue against mechanism!
Comp makes CONCEIVABLE that comp is false. To accept artificial brain or
Quantum Suicide are necessarily a matter of personal opinion.
>Even if we are made of this matter, precise
>mathematical laws are not required to make an organized system work:
>civilizations are made of individuals, and they are well structured despite
>the absence of a good mathematical description of each individual.
OK. What is the point ?
>I think in fact that there are indeed much more "brain -like" computations
>without this constraint than with it, and the appearance of physical laws
>would thus be difficult to understand.
Sure. That why I insist on the big work which remains to be done by
computationnalist for solving the mind-body problem and the problem of
the origin of the physical laws.
>A(nother) stone in the garden of computationalism?
Yes. A usefull stone to build a way toward a theory of mind, in the
jungle of computationalism :-).