On 23 Jun 2016, at 03:01, Bruce Kellett wrote:
On 23/06/2016 3:04 am, Bruno Marchal wrote:
On 21 Jun 2016, at 04:08, Bruce Kellett wrote:
On 21/06/2016 3:14 am, Bruno Marchal wrote:
On 20 Jun 2016, at 04:00, Bruce Kellett wrote:
On 20/06/2016 4:09 am, Brent Meeker wrote:
The alternative, which Bruno actually suggested once but
disowns, is for explanations to form a "virtuous circle" in
which everything is explained in terms of other things
ultimately forming loops: NUMBERS -> "MACHINE DREAMS" ->
PHYSICAL -> HUMANS -> PHYSICS -> NUMBERS I call this
"virtuously circular" if it is comprehensive so that everything
is somewhere in the circle.
The thing about such a loop is that you can start at any point
-- for instance, PHYSICAL, HUMANS, PHYSICS, or anywhere else.
The question then is whether this actually achieves you anything?
Just stop on the simplest theory.
"Simple" is an undefined term. You might think the integers are
simple, I might think that physics is simple. No-one is right in
any absolute sense.
Why? We can define a theory to be simpler than another if it has
less assumptions.
The primary criterion for the adequacy of a theory is that the
theory be consistent with the data. Simplicity is at best a
secondary consideration, and then largely in the eye of the
beholder, or the way in which the theory is formulated. It is like
the choice of a coordinate system for astronomy -- the calculation
of the trajectory for an earth-moon mission is extremely difficult
in a coordinate system centred on one of the moons of mars but quite
simple in the coordinate system centred on the earth. But the
physics is the same whatever the coordinate system. "Simplicity" is
a loaded word, a word that should be avoided.
Simplicity is just occam. We prefer the theory according to which the
earth is not fixed with all star and planet going around because it
make us avoiding many needed supplementary hypothesis. For two
theories with equal theorems we prefer the one which assumes the less.
Ig you can explain QM and consciousness from elementary
arithmetic, you make a gain compared to starting from physicalist
QM, which assumes Matter, QM (and thus arithmetic, as QM already
assumes arithmetic).
Well, you should formalize your theory so that we can at least
compare.
My theory is that the external objective physical world exists,
independently of you or me, or even of consciousness.
Consciousness is a property of certain forms of matter, and matter
achieves those forms, and hence consciousness, by the process of
evolution.
OK. So your theory asks for an ontological commitment in some
Nature or Matter, being independent of us.
And your theory requires an ontological commitment to the existence
of the integers.
Actually it does not. It needs only an agreement with the usual
elementary arithmetical proposition. And, any Turing-complete theory
would do. We don't need to make existence metaphysical, and indeed
that was the genious progress of the mathematical logicians.
And all what I prove is that you cannot be simulatenously rational,
agrees on elementary arithmetic (= give sense to the word "digital"),
be Turing emulable, and use a notion of physical primary matter to
rely consciousness (or first person) with the third person describable
reality. And so it you are computationalist (like 99,99% of scientist)
you have to derive the physical laws from the theology of numbers.
There is just no choice in the matter.
You don't actually have fewer ontological assumptions.
I do science. I make zero ontological commitment. I make only
hypothesis, and computationalism is ontologically neutral a priori.
You need to understand, though, theorem in arithmetic, like Euclid's
proof that there is no bigger prime number.
In the epistemology, given by the Löbian number, I use, with her,
freely the smallest number principle. if some numbers have the
property P, I assume there is a least one. In classical logic, this
can be shown equivalent with Löbianity, or with the induction
principles. Of course, P has to be describable by an arithmetical
formula, as I avoid second-order logic (there).
Who said that the integers are simpler than nature? A number
theorist might not agree!
Nor a logician, as we know that natural numbers, or integers, when
they get the additive & multiplicative structure, the true
propositions esacape all machines or effective theories. In that
sense, numbers are unboundably complex. but that complexity is in the
semantic: the theory is simple in the sense above: it has few
assumption. Kxy = x with Sxyz = xz(yz) are already enough.
Note that the ontology is the integers, not some set of axioms.
It is not necessarily an ontology. You just need to play fair and say
if you agree that zero is not a successor, that zero, when added to a
number, gives that number, etc. I do not assume something that any
theoretical physician does not assume.
As I said, the progress are
The numbers
Dualist unintelligible theory of mind and matter
SWE
The numbers
Computationalism
SWE
The numbers.
(indeed, once you have derived physics from arithmetic, you can still
throw away computationalism, logically, and still introduce any gods
you want).
Note also that the derivation provides works with transfinitely
possible weakening of computationalism. Most arithmetical gods get the
same physics. Oracles does not add much to the theology. (Turing
oracles, not the delphic one!).
Axioms are statements that are true of the pre-existing integers,
axioms are not ontological statements.
Good! Just my point. That is why I do much less ontological commitment
than we need to assert physicalism.
Keep in mind that I am ever doing philosophy, or if I do philosophy, I
do it as a scientist, interested in a precise theory, which provide
much light on my question of interest (the mind-body problem) and
which is experimentally testable.
By the UDA, the theory obtaoned is obviously testable, as it gives the
tools to extract mathematically physics from arithmetic (through the
"head" of the Löbian machine/number), and so we need just to compare
that physics and the physics inferrred from the observation.
Then you know my point: you can't say yes to the doctor or you need
to refute Church Thesis.
I deny this. Saying yes to the doctor merely requires you to accept
that the functionality of the computer is the same as the
functionality of your brain. This is as possible in the physicalist
account as in any other. Why should not the physical world be Turing
emulable?
Which physical world? If you accept that you survive when the digital
functionality is preserved, then you survive already in infinitely
many occurence of your brain re^presentation, at the correct
substitution level in arithmetic. Saying that only this or that
reality or number counts is just invoking magic. below our
substitution level, there are infinitely many universal numbers
competing to provide your consistent continuations.
Of course, like Bohm, you can assume that there are particles, and
conspiratorial potential, but that looks like Ptolemeaus epicycles,
and worst, they prevent the computationalist theory of consciousness
to apply. Scientists don't do that. Only blind believers do.
This might not be true for a computationalist account, but I reject
that account: if physics can be seen as possible a simulation run
by some alien civilization, then physics is certainly Turing emulable.
Which is not the case. The alien can fail us only for a finite time. I
explain that to Brett Hall: but computationalism makes it possible to
see if we are in a normal emulation, or at the physical bottom, the
things which is the first person sum on all emulations (by the FPI).
The phenomenal physics is not entirely Turing emulable, but that might
be no more than the presence of a random oracle on some near-
equivalent computations.
You must get out of this habit of claiming that certain things are
impossible for the physicalist, while these things are merely
impossible *if you assume computationalism* The physicalist rejects
computationalism, so your "refutations" are baseless.
Well, all my work is a modest proof that computationalism is
incompatible with physicalism (and some weak amount of occam). Then I
show it testable, by giving the logic of the observable proposition,
and thanks to QM and Quantum logics, it works till now.
From a physicist pov, I just extend Everett on all computations,
instead of cheating and taking the quantum one copied from Nature.
We cannot copy Nature, because this deprives us to test the theory
with Nature later, if you want.
Mind (and consciousness) is a property of brains and other
configurations of matter that have similar functionality.
If the functionality is Turing emulable, then your theory cannot
work, without involving some non-Turing emulable element paying
some role in mind and matter to assure the identity link.
Rubbish. It is no different from your "yes, doctor" scenario.
Requiring "something" to assure the mind and matter identity link is
pure dualism. Nothing extra is required. There is no need for
anything to be not Turing emulable. Don't you understand? Once you
have explained the functionality, there is nothing left to explain
-- there is no magical mind-body link to explain -- that is dualism.
Exactly. Physicalism is dualism, has it take the numbers and the mind
(when they are not elmiminativist of course), but still introduce a
primary matter, or assume physics cannot be explained by something
simpler, despite those are metaphysical assumptions never needed in
actual physics. The law of numbers are enough. They are so simple that
they belong to the primary school curriculum, and with
computationalism, we get the begining of an explanation of where the
physical laws come from (number dreams coherence condition inherited
from incompleteness).
Is that not cute and elegant?
I am open minded that it can be false, and indeed, I worked 30 years
to refute it, and well, it is not refuted, but it gives apparently
only the same weirdness that the experimental physicists is already
confronted from.
We learn about these things, and about the basis of consciousness,
empirically, by applying the scientific method in our study. There
is no "hard problem" of consciousness, because once we have
understood the functionality, we have understood all that there is
to it.
The contrary happens. Once we understand the functionality, it
looks like if consciousness is no more needed, but then why would
evolution endows us with it?
Because consciousness is part of the functionality.
Part? It is certainly related to it, but "part" seems introduce a
category error. Consciousness is 1p, functionality is 3p. You cannot
identify them. That would mean here that consciousness is part of []p,
when consciousness is in the association with truth: []p & p.
(Truth here means satisfied by the standard model of arithmetic).
There is no mystery here, apart from that arising from your
inherently dualist understanding of consciousness.
?
There is no mystery once you abandon physicalism, yes.
If not, explain to me how a Turing universal machine can distinguish a
physical computation from an arithmetical computation. Give me a
theory of matter, and use it to explain to me how that matter select
the computations and make them real. It looks like a god-of-the-gap to
me, but if I am wrong, just show me.
That is the hard problem: solving the describable easy part of it
makes the harder part only more harder, ... until you get the
Theatetus point, and remind that a brain is supposed to have some
relation with truth.
What is truth?
With computationalism we can limit ourself to arithmetical truth, and
even (but this is in G* minus G) just the sigma_1 arithmetical truth,
which even PA can already define.
There has been a lot of progress in epistemology and the notion of
"truth" since Plato's time. Maybe you should get up to date.
Ad hominem.
There is no "hard problem"-- to think so is a simple category mistake.
May be you don't understand the problem.
Your "Yes, doctor" thought experiment is actually saying much the
same thing: the functionality is all that matters -- once you have
understood the function, you have understood consciousness. The
doctor doesn't have to transfer "your consciousness" once he has
replaced your brain with a functionally equivalent computer. There
is no duality.
Indeed, but there is still a sense in saying that the consciousness
has been preserved.
Your inherent dualism is showing again........ You separate
functionality and consciousness -- that is dualism.
I don't separate them, indeed computationalism associate them, but
that is not a reason to identify them. Indeed that would be the
category mistake. A pain in my thumb *is* not a bunch of neuronal
firing, like a winning chess play is not a bunch NAND electricity.
Now, relatively to a universal environment, a pain in a thumb might be
a semantic of a digital neuronal firing, but then that's already the
case for all Turing equivalent (intensionally too) level. Church
thesis entails the intensional Church thesis, so each computer can
emulate the behavior at all substitution level possible.
And this will work if we accept that the brain does not produce
consciousness, it makes only possible for a consciousness (a
person) to manifest itself relatively to some computations shared
with other persons. But this eventually is what will make physics a
dream-sharing theory, if we do get the right relative measure when
solving the measure problem.
Get rid of the white rabbits, you mean? You can't do this, you know.
On the contrary. The fact that the logic of "proability" one on the
computations/sigma_1 sentences gives a quantum logic and a
quantization is an evidence that we could.
If you have a proof that we can't, then I would be pleased you give it
to us.
I have already give three equivalent version of "my" theory,
which is probably the same of yours minus assuming mind and
matter and a mysterious link between, as far as I understand.
As I understand it, your theory assumes the existence of numbers,
or at least of the integers. Your base ontology can be used to
support axioms, giving RA, PA and so on. Once you have some axioms
and rules of inference, you can prove theorems. You then identify
"existence" with the existential quantifier of mathematics -- if
we can prove Ex(x = y), then you say that y is also part of the
ontology.
I would say only that if the machine M can prove Ex(x = y), the
machine will believe in the existence of y. Then it happens that
all humans believe in RA and PA, at least all those willing to give
sense to "digital mechanism", so if M proves Ex(x=y), we can take
it as true.
Believing is not knowledge,
Indeed, I define, with Theatetus, knowledge by true belief. (and
knowledge-for-sure as true and consistent belief).
and truth does not imply an ontology --
So glad to ear that, but this is exactly why (immaterial/arithmetical)
mechanism is so simpler than physicalism. many people assumes both.
Even Hao Wang, a good philosopher and logician made the error of
literally confusing mechanism and materialism. I guess we have just
not completed the Enlightenment Period: we still keep half the Gods of
Aristotle. But Church's thesis rehabilitates the Neopythagorean.
"triangles have three corners" is a true statement, whether or not
triangles exist in any objective sense.
Yes, and that is why I insist "It exists prime numbers" is true,
independently of how you interpret arithmetic. It is true in all
models of arithmetic, and we don't claim that there is one. Indeed,
with computationalism that would be enough to become inconsistent.
You might read the book "Inexhaustibility" by Torkel Franzen which
analyses very well this point, which I exploited since long, as Church-
thesis makes this possible.
Computationalism illustrates that we can do, and should do, theology,
with the same attitude of not doing any ontological commitment. The
terms used to agree on axioms admits interpretations, themselves
amenable to mathematical reasoning.
The same with applied physics. To go on the Moon or on Mars, you don't
need to believe in the objects Moon or Mars. You need only to hope the
possible hallucination is stable enough and lawful. Same in the first
step of the UDA. You need to assume doctor and brains, but not in
primary doctor or primary brain.
When doing science, we just do not commit ourself ontologically.
We stay in the hypothetico-deductive + testable realm each time we
open our mouth.
But without assuming the numbers to start with, you can never get
to theorems and the existential quantifier.
That is why I assume x + 0 = x, etc.
The existential quantifier merely points to an already existing
entity. We can say Ex(x^2 = 4) only because the integer 2 already
exists. It is a pointer, not an ontological operator!
OK. But such existence is the mundane existence, not the one of some
metaphysical commitment.
The existential quantifier is not then a definition of what
"existence" consists in.
It is the simplest one, as I can explain in very simple conceptual
term the theological and the physical existence (notably by the
simple arithmetical modal [i]Ex[i]P(x), and [i]<i>Ex[i]<i>x.
If physicall real atoms are supposed as fiundamental, I might need
to define the numbers by brain configuration of apes, and get the
definition of zero with a many volumes book reducing the ape's
brain to superstring theory, and this in a version of superstring
theory never assuming natural or integers.
If you use superstring theory then you are most likely wrong! No,
all you need is a physics in which the basic endurance of objects is
ensured by symmetries and Noether's theorem.
Your ontology is assumed before you get that far, before you have
arithmetic even.
?
What I said above: you have to assume the existence of at least the
integers before any arithmetic can get off the ground.
If you stop at 2+2=4, then I can't do anything for you. But physicists
use 2+2=4 as much, in their theories.
Just keep in mind the progress in the TOE research:
The number laws.
Dualist unintelligible theory of mind and matter
SWE
The number laws.
Computationalism
SWE
The number laws.
Computationalism.
especially that this last one justifies both the quanta *and* the
qualia (and much more, with the eight modal internal views).
So you are actually no better off than the physicalist -- you
still have to assume a primitive ontology:
Yes. Without some theory, we don't have a theory. We need to agree
on somethings. But my theory is just elementary arithmetic. It is
believed by everybody (except sunday philosophers), and it is
assumed by physicists too.
No, your ontology is the integers, and these are not assumed by
physicists -- integers are derived by physicists from experience
of the physical world.
Not in their theories, or show me a physical theory which does not
assume the numbers and where they are derived from purely physical
things. That does not exist. That's why Wigner, and Einstein, Galilee,
where astonished by the comprehensibility of nature and by the role of
numbers and math. Of course, that becomes obvious with computationalism.
this might form a model for arithmetic,
The (standard) model is the well known structure (N, 0, +, *). It
was taught in high school explcitly sometimes ago.
but then so does the physical universe.
To assume an ontologically physical universe at the start is no
better, for me, than to say God made it.
God made the integers, all else is the work of men! "Die ganzen
Zahlen hat der liebe Gott gemacht, alles andere ist Menschenwerk".
(Kronecker)
God made the integers, all else belongs to the integer's dream (your
servitor).
Anyway, I thought naively for a long time that all scientists knew
that as long as the mind-body problem is not solved, we should be
open to change our mind of what is more fundamental (number,
matter, mind, etc.)
The physical has the property of containing distinct objects --
hence already has an ontology of integers, but in addition, things
exist in the physical universe without the superstructure of
axioms and theorems -- we just look and see! Physicalism is
actually simpler!
Grandmother theory of gravitation (everything falls down if not
supported) is simpler than Einstein's theory. Then.
The goal is the search of the truth.
What is truth? The physicist only seeks theories that are not
contradicted by the data -- "truth" is an optional extra!
Yes, but when they found that the mass of the photon is 1/12 minus
(1+2+3+4+5+6+7+ ...) they are happy when the number theoreticians tell
them that it makes sense to say 1+2+3+4+5+6+7+ ... = -1/12 (something
PA already know, when said precisely).
Here by truth, with computationalism, eventually we can limit
ourselves to sigma_1 truth, which is already definable in PA.
Arithmetical truth itself cannot be defined in PA, or by any Löbian
machine, unless using second-order arithmetic (which does not help as
it relies on set-theoretical truth, which is even more difficult to
define). This reflects our own Turing universality, we quickly
understand that we cannot define the numbers without being circular.
But here, with comp, *that* is a virtual circle, and machine can use
it, although there is some perils (but that's life in numberland).
And my point is just that mechanism is incompatible with
materialism, and that we can test them experimentally.
I don't defend any truth anywhere.
So don't talk about it so much,
Sorry, but it is a key concept in logic and computer science.
We need it to understand that incompleteness makes knowledge obeying a
different logic than belief, etc. We need it to have semantics. But
the notion of truth here is mundane, and ask not more than what we use
in elementary set theory or analysis.
and don't criticize alternative theories just because they are
incompatible with your theory.
I don't criticize any theory. I just show the incompatibility of
existing theories, and shows how the physical evidence are on the side
of mechanism. Indeed, the quantum appears to be explained by the logic
of self-reference at the place the UDA shows it has to appear.
And then we get the qualia, at a time where distinguish physicalist
eliminate them (Churchland, Dennett, etc.) or bring back some form of
dualism, (Searles, Chalmers, etc.).
I provide a reasoning only. Then I explain that QM (without
collapse) provides evidence for mechanism, and thus again (weak)
materialism.
That is why I like computationalism: it makes a large part of
philosophy, metaphysics and theology experimentally testable.
Until you have some solid predictions that are different from those
of a materialistic theory, and your predictions are borne out by
experiment, you don't have anything of any particular value.
See my long text "conscience and mechanism" for a program generating
the currently testable propositions.
You clearly criticize something that you have not yet studied.
You can't oppose philosophical taste with science, if that was needed
to be reminded.
I am open to some error, but usually, when people try, it quickly
looks like bad faith, with insult, ad hominem remark, or eventually
recognition that they don't find the error, and may be that there is
none.
It is just new for those who took for granted Aristotelian
Metaphysics. Are you able to doubt it? If yes, all you have to do is
study a bit of computer science and mathematical logic.
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at https://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 [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
