On 17 Apr 2015, at 14:35, Bruce Kellett wrote:
Stathis Papaioannou wrote:
On Friday, April 17, 2015, Bruce Kellett <bhkell...@optusnet.com.au
<mailto:bhkell...@optusnet.com.au>> wrote:
Stathis Papaioannou wrote:
On Friday, April 17, 2015, Bruce Kellett
<bhkell...@optusnet.com.au <mailto:bhkell...@optusnet.com.au>>
wrote:
Stathis Papaioannou wrote:
Physicalism reduces to computationalism if the physics
in the
brain is Turing emulable, and then if you follow
Bruno's
reasoning in the UDA computationalism leads to
elimination of a
primary physical world.
But physics itself is not Turing emulable. The no-cloning
theorem of
quantum physics precludes it.
Do you mean because you can't exactly copy a given physical
state? That doesn't necessarily mean the physical world as a
whole cannot be emulated. And if it turns out that physics is
continuous rather than than discrete you could still come
arbitrarily close with digital models of the brain; if that
was
not good enough you would be saying that the brain is a
machine
with components of zero engineering tolerance.
An exact copy of an unknown quantum state is not possible. It most
certainly does mean that the physical world as a whole cannot be
emulated. Quantum mechanics is based on the incommensurability of
pairs of conjugate variables. Because you cannot measure both the
position and momentum of a quantum state to arbitrary precision
simultaneously, we find that there are two complementary
descriptions of the physical system -- the description in position
space and the description in momentum space. These are related by
Fourier transforms. Any complete description of the physical world
must take this into account.
You can't copy an arbitrary quantum state, but you could copy it by
emulating a series of quantum states.
?
If a quantum state could be duplicated, then you could measure
position exactly on one copy and momentum on the other. Exact
values
for these two variables simultaneously contradicts the basis of
quantum mechanics. And there are very good arguments for the view
that the world is at base quantum: the classical picture only
emerges from the quantum at some coarse-grained level of
description. You cannot describe everything that happens in the
physical world from this classical, coarse-grained perspective.
If you had an actual Turing machine and unlimited time, you could
by brute force emulate everything. However, that is not the point.
If you car needs a part replaced, you don't need to get a
replacement exactly the same down to the quantum level. This is the
case every machine, and there is no reason to believe biological
machines are different: infinite precision parts would mean zero
robustness.
I think you miss the point. If you want to emulate a car or a
biological machine, then some classical level of exactness would
suffice. But the issue is the wider program that wants to see the
physical world in all its detail emerge from the digital
computations of the dovetailer.
By the FPI on all computations. This will be a priori not computable.
That the universe looks some much predictable is the mystery with
comp. We must fight the white rabbits away.
If that is your goal,
The result is that we have to do that if we assume computationalism in
the cognitive science.
then you need to emulate the finest details of quantum mechanics.
Comp explains why this is impossible. The finest details of physics
are given by sum on many computations, the finer the details, the more
there are. To get the numbers right up to infinite decimals, you need
to run the entire dovetailer in a finite time. We can't do that.
This latter is not possible on a Turing machine because of the
theorem forbidding the cloning of a quantum state.
Same with comp.
Quantum mechanics is, after all, part of the physical world we
observe.
It might be part of the reality we live, but it might be explained by
the arithmetical FPI on the computations seen from inside. IF QM is
correct, and if comp is correct, QM has to be a theorem in comp, that
is, the logic of []p & <>t have to give a quantization on the sigma_1
arithmetical sentences. And that is the case.
([]p is Gödel's beweisbar(x), meaning provable(x), and <>t is the dual
~beweisbar('~(1=1)').
Don't confuse Digital physics (the universe is a machine) and comp (my
body/brain is a machine), as they are incompatible (and as Digital
physics entails comp, but comp entails ~Digital-physics, so digital
physics entails ~digital physics, so digital physics is self-
contradictory. With, or without comp, we are confronted to something
non Turing emulable. No need to go outside arithmetic, as we know
since Gödel, Church, Turing, Post, ... that arithmetical reality is
much larger than the computable part of it.
Bruno
Bruce
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
