On 17 Apr 2015, at 08:26, Bruce Kellett wrote:
meekerdb wrote:
On 4/15/2015 11:16 PM, Bruce Kellett wrote:
LizR wrote:
On 16 April 2015 at 15:37, Bruce Kellett
<bhkell...@optusnet.com.au <mailto:bhkell...@optusnet.com.au>>
wrote:
Bruno has said to me that one cannot refute a scientific
finding by
philosophy. One cannot, of course, refute a scientific
observation
by philosophy, but one can certainly enter a philosophical
discussion of the meaning and interpretation of an
observation. In
an argument like Bruno's, one can certainly question the
metaphysical and other presumptions that go into his discourse.
Yes, of course. I don't think anyone is denying that - quite the
reverse, people who argue /against /Bruno often do so on the
basis of unexamined metaphysical assumptions (like primary
materialism)
And the contrary, that primary materialism is false, is just as
much an unevidenced metaphysical assumption.
But it's not an assumption in Bruno's argument. As I understand it,
his theory in outline is:
1. All our thoughts are certain kinds of computations.
2. The physical world is an inference from our thoughts.
3. Computations are abstract relations among mathematical objects.
4. The physical world is instantiated by those computations that
correspond to intersubjective agreement in our thoughts.
5. Our perceptions/thoughts/beliefs about the world are modeled by
computed relations between the computed physics and our computed
thoughts.
6. The UD realizes (in Platonia) all possible computations and so
realizes the model 1-5 in all possible ways and this produces the
multiple-worlds of QM, plus perhaps infinitely many other worlds
which he hopes to show have "low measure".
I leave it to Bruno to comment on whether this is a fair summary of
his theory.
It is not a theory. It is an argument. it is dangerous to sum it by
"thought = computation" . The only axiom is that consciousness is
locally invariant for a digital substitution made at some level. It is
a very weak version of Descartes Mechanism. It implies all form of
mechanism and computationalism studied in the literature. It is my
theory if you want, but my theory is believed by basically all
rationalists by default. Only precise and rare people, usually
philosophers, but also some scientists, like Penrose, defends
different theory.
What makes it stronger than the STRONG AI thesis, is that it is
supposed to apply to us.
What makes it weaker than most computationalist thesis, is that there
is no bound delimited for the substitution level.
Then, I argue that this leads to the fact that all first order
specification of any universal machine/program/number gives a TOE. In
particular the laws of physics have to de derived in any of those
TOEs. It gives actually much more and the whole stuff I like to call
it theology, because it is arguably isomorphic to Proclus theology,
and Plotinus, Plato. But all this are in the results. The theory is
only that I am Turing emulable. Even if the brain is a quantum
computer (which I doubt), I remain Turing emulable, (see the paper of
Deutsch).
I have difficulties with several of the points you list. But that
aside, one has to ask exactly what has bee achieved, even if all of
this goes through?
Everyone knows that Aristotle physics has been refuted. Already by
Galilee.
The achievement here is a refutation of Aristotle's theology, in
computationalist frame (the one believed usually by materialist,
atheists, but also many religious people).
I do not think it explains consciousness.
That was not the goal. But yet, I can argue that 99% of the conceptual
problem is solved, and that the remaining 1% is simply unsolvable. But
for the origin of matter appearances, the explanation is conceptually
100% solved. In that frame, and assuming it true, as the result is
also that this can be tested.
It seems to stem from the idea that consciousness is a certain type
of computation (that can be emulated in a universal Turing machine,
or general purpose computer.)
Not really. Consciousness is 1p, and it the math explains why
consciousness, like truth, are not definable in arithmetic, unlike
computations. In fact consciousness is not definable in any third
person way.
It certainly does not ring right, that consciousness would be a
computation, and already the FPI suggests that consciousness is
related to infinities of computations, and in the meaning or semantic
of those computation, which the machine are unable to define entirely
by themselves.
This is then developed as a form of idealism (2 above) to argue that
the physical world and our perceptions, thoughts, and beliefs about
that world, are also certain types of computations.
Not at all. It is just that if your brain is Turing emulable, it is
Turing emulated infinitely often in arithmetic (in a tiny part of the
standard model of Peano Arithmetic, say).
That is a theorem.
You are the one coming up with a sort of God, the primary physical
universe, and tell me, without proof, that it is able to select one
computation among infinities of others.
But this is no nearer to an explanation of consciousness
UDA explains the problem.
You need to study AUDA (Arithemtical UDA, the interview of the
universal machine, in sane04) to have an idea how consciousness and
matter are explained.
than the alternative model of assuming a primitive physical universe
and arguing that consciousness supervenes on the physical structure
of brains, and that mathematics is an inference from our physical
experiences.
OK. But the point is that this view is refuted in the theory assuming
the brain is Turing emulable. That *is* the point. You have to find a
flaw in the reasoning, if you want continue to believe in both
mechanism and materialism.
Consciousness supervenes on computations? What sort of computation?
All of them. That is made mathematically precise by using Church's
thesis. They are all realized in the standard model of Peano
Arithmetic, although from inside, things get much bigger.
Why on this sort and not any other sort?
My original thesis contained 300 pages to explain this. But I ahve
been asked to suppress it, as it is obviously known to everybody. It
is indeed "just" elementary theoretical computer science. With
Church's thesis, the universal machine is "really" universal for the
notion of computations. You can program the universal dovetailer in
LISP, but when he run, he will dovetail on all universal numbers, and
thus make all computation is all the possible way to compute anything;
It will emulate all quantum topological computers, all fortran
programs, all patterns of the game of life, and it will even
approximate all "Garden of Eden" cellular automata configurations.
There is one thing it will never simulates exactly though. It is the
physical universe. If you grasp step seven, you know that a priori,
you cannot emulates exctly anything physical, as it is "only" the
result of a statistics on an infinity of computations. Comp predicts
that below your substitution level, the FPI must become indirectly
observable, and indeed the MWI, or just QM without collapse, confirms
this startling consequence.
Similar questions arise in the physicalist account of course, but
proposing a new theory
It is not a new theory. You can find it in chinese taosit text, in the
"question to king milinda", in Plato. It has been forgotten, and
renwed by Descartes, Hobbes, LaMettrie, and it is base of the whole of
cognitive sciences, physiological sciences, etc. It is usaully
attacked by religious or pseudo-religious people.
I just show that mechanism is incompatible with (weak) materialism.
that does not answer any of the questions posed by the original
theory does not seem like an advance to me.
Indeed. In theology, we need to backtrack 1500 years, but we have a
new companion at the discussion table, the Löbian numbers.
But indeed, to grasp what I ahev done is graspin that the mind-body
problem is two times more complex with computationalism, because, not
only we have a the problem of the mind, in the mind-body problem, but
we have now a problem of matter also. WE need to explain matter in
purely arithmetical terms, with internal probabilistic interpretation.
the crazy remarkable thing, is that incompleteness offers those
internal views at the second the universal machine become Löbian (that
is: realized that she is a universal number).
Then to pursue, you need to solve the open problem.
At least physicalism has evolutionary arguments open to it as an
explanation of consciousness
It can't work. Evolution needs a typical bet on computationalism. But
by UDA, with computationalism you cannot explain the matter used in
evolution as the ultimate base of evolution.
Nicely, computationalism extends Darwin to the birth and evolution of
the physical laws, although it is more a sort of percolation of number
dreams.
The physicalist model has the advantage that it gives the physical
world directly -- physics does not have to be constructed from some
abstract computations in Platonia (even if such a concept can be
given any meaning.) If you take the degree of agreement with
observation as the measure of success of a theory, then physicalism
wins hands down. Bruno's theory does not currently produce any real
physics at all.
That is debatable. The logic of the observable is not nothing,
especially when you see how close Quantum logic is from an explanation
of quantum mechanics. The only difficult things is gravity/energy-
space-time, but even for that, there are hints of where they come from.
But anyway, comp is not physics: it is an axiom in theology (the bet
that you remain alive for that digital substitution), and it is
testable, because in that theology physics is derivable from addition
and multiplication.
The discussion of the detailed steps in the argument Bruno gives is
merely a search for clarification. As I have said, many things seem
open to philosophical discussion,
Nothing is subject to that. You miss the point completely. the point
is that we accept comp, such question admit purely mathematical
formulation, and indeed we can already interview machine's like PA,
and they already have an explanation which fits Proclus and Plotinus
theology, and its is indirectly testable by the physics which is
entailed by computationalism.
You must not compare this with physics, but with theology. The
constructive result is that comp refutes physicalism, and it is
constructive as it gives the means to recover the appearances.
and some of Bruno's definitions seem self-serving. When I seek
clarification, the ground seems to move beneath me. The detailed
argument is hard to pin down for these reasons.
It is all normal if you have not yet grasped what is a computation, in
the sense of the mathematicians.
It is all normal if you have not got the vertigo of the complexity of
the consciousness problem.
It is all normal if you have some difficulties to doubt the
aristotelian dogma (the existence of a primitive physical universe).
Nowhere I argue for or against mechanism or materialism: I show them
incompatible, and then I interview the universal machine to get her
opinion. The main work here is Gödel, Löb, and the key result of
Solovay (reference in my URL or papers, or ask).
For the definitions, I use the most common one used in the discipline
I cross.
OK. My pause-café is finished. Back to work. Bruce, to study a paper
asks very often to study the main references. You need some good
passive understanding of computer science and elementary mathematical
logic.
I give you an exercise, which should be easy if you grasp step seven:
shows that the physical matter cannot be duplicated exactly, once we
assume comp. Like step 3 entails indeterminacy, step 7 entails non-
cloning. The answer is even above, I think.
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.
