On 10 Sep 2012, at 16:39, benjayk wrote:
Bruno Marchal wrote:
On 08 Sep 2012, at 15:47, benjayk wrote:
Bruno Marchal wrote:
even though the paper actually
doesn't even begin to adress the question.
Which question? The paper mainly just formulate a question, shows
how
comp makes it possible to translate the question in math, and show
that the general shape of the possible solution is more close to
Plato
than to Aristotle.
The problem is that the paper is taking the most fundamental issue
for
granted,
Absolutely not. I am open that UDA could lead to a refutation of
comp,
either purely logical, or by the possible testing it implies.
My opinion on the truth or falsity of comp is private, and to be
honest, varying.
You want me to be more than what I am. A logician. Not a philosopher.
It is simply not my job.
OK, but if you are solely a logician, you should concern yourself with
logical proofs. You don't even define the assumption of your paper
in a
(theoretically speaking) logical way and your proof contains many
philosophical reasonings.
You confuse formalism and rigor. When you *apply* logic, the
hypothesis can refer to non logical concept. You have to tell me
exactly where you have a problem.
Especially step 8, which is criticial in your reasoning.
If you get the step 3, it is already enough. And with the next steps
up to seven, you have already the indeterminacy, non locality and non
cloning. I give this as exercise, but I can say more if you are
interested.
It uses occams
razor (which is philosophical, and not necessarily valid in any
mathematical
or logical context),
It is used in the last step when we apply it to reality, in its
weakest form, as it is always used when we bet on a reality, even
looking for a beer in the fridge. If you don't like it, you can take
the weaker consequence that primitive matter is epinoumenal (like
invisible horse).
you use appeals to absurdity (with regards to aribtrary
inner experience being associated to null physical activity),
Thats what logicians do all the time.
you use
additional philosophical assumptions (you assume materialist mechanism
cannot mean that physical computations are not *exactly* like abstract
digital computations, just enough to make a practically digital
substitution
possible),...
?
So take my criticism to mean that your proof is simply not what you
present
it as, somehow being beyond philsophy (which is always on some shaky
ground).
Not with comp.
This is what I perceive as slightly dishonest, because it allows you
retract
from the actual point by demanding to be given a precise refutation
or a
specific error (as required in logic or math). But your paper is
philosophical, and here this logic does not apply.
Easy. I illustrate that *in* the comp theory, we can get conclusions.
If you were correct, you would help to add the clarity which is missing.
If you'd admit that I am perfectly happy with your paper. It does show
something, just not rigorously and not necessarily and not for
everyone
(some may rightfully disagree with your reasoning due to philosophical
reasons which can't be proven or be precisely stated).
I use comp to avoid such arbitrariness.
It looks like if you want that philo remains on shaky grounds. Only
comp might be shaky, as applying to reality, but once assumed, I don't
see the flaw.
I have test the reasoning on a lot of scientist, and I have never had
any problem, unlike with media and some philosophers.
Improve it, if you think that possible (as it is probable), or find a
flaw, but don't tax the subject as being necessary fuzzy, as with
comp, the assumption are clear, and the consequences too.
If someone believes that physics behaves perfectly like abstract
computations would and if he doesn't want to invoke some very
mysterious
form of matter (which does not rely on how it behaves and also not
on how it
feels or is perceived to be) to sidestep the problem, yes, than your
paper
may indeed show that this does not make much sense.
Unfortunately most materialist do actually believe (perhaps
unconsciously)
in some very mysterious and strange (and IMO meaningless) kind of
matter, so
they won't be convinced by your paper.
And that was the point. It applies to rational enough agent, not to
people ready to put magic to escape the conclusion. I am sure I can
make this clearer in the step 8, as it is hard to refute magic through
logic.
Bruno Marchal wrote:
("kinda digital", "digital at some level" are not enough for a
strict
reasoning).
You also say that a 1p view can be recovered by incompleteness, but
actually
you always present *abstractions* of points of view, not the point
of view.
What could that mean? How could any theory present a point of view?
I think you are confusing level. You could as well mock the quantum
analysis of the hydrogen atom as ridiculous because the theory cannot
react with real oxygen.
That's the point. A theory cannot conceivably present and acutal
point of
view. But then your theory just derives something which you call
"point of
view", which in fact may have little to do at all with the actual
point of
view.
In UDA, the points of view are the memory content.
In AUDA, I use the classical theory of knowledge (S4), that is the
most common one, and show that the Theaetetus' definition can works
thanks to incompleteness and lead to it (actually a more precise
version of it: S4Grz).
QM does not claim to show how a hydrogen atom leads to a real
reaction of
oxygen, it just describes it.
To make it coherent, you would have to weaken your statement to "we
can
derive some description of points of view, or we can show how some
description of points of view emerge from arithmetics", which I will
happily
agree with.
But with comp, if the machine agree with your description of points of
view, and have the relevant discourse, you can treat them as zombie.
Unless reifying a magical notion of matter, a bit like Peter Jones.
You omit that among the description of the first person point of view,
the machine admit already that the first person view has no
description at all, but at the metalevel, the machine can use comp,
and describe it by the modal logic S4, like us. So, without confusing
the level, the machine agrees with you. But this invalidate nothing,
on the contrary.
However, this would destroy your main point that arithmetics and
its point of view is enough as the ontology / epistemology (we need
the
*actual* point of view).
So, this does not follow.
Bruno Marchal wrote:
Bruno Marchal wrote:
How am I supposed to argue with
that?
There is no point of studying Gödel if we have a false assumption
about what
the proof even is about. It is never, at no point, about numbers
as
axiomatic systems. It is just about what we can express with them
on a
meta-level.
On the contrary. The whole Gödel's thing relies on the fact that
the
meta-level can be embedded at the level.
Feferman fundamental papers extending Gödel is "arithmetization of
metamathematics". It is the main point: the meta can be done at the
lower level. Machines can refer to themselves in the 3p way, and by
using the Theatetus' definition we get a notion of 1p which
provides
some light on the 1//3 issue.
But Gödel does not show this. The meta-level can only be embedded at
that
level on the *meta-level*.
This is just false.
Sorry, I meant on *a* meta-level (not the meta-level that can be
embedded,
obviously).
?
They are the same.
If you think it if false, then show me a proof of Gödel that does
not rely
on a meta-level (beyond the embedded meta-level in arithmetic). I
have never
seen such a proof.
Read cautiously Gödel 1931.
Technically Gödel still use a notion of omega-consistency, and the
complete embedding is due to Rosser, and is clear in Hilbert-Bernays,
and then Löb, and then Solovay.
Bruno Marchal wrote:
Apart from this level, we can't even formulate
representation or embedding (using the axioms of N - except on
another
meta-level).
False. I can only suggest you to study the original paper, or to
follow some good course in logic.
You just miss the most original and admittedly astonishing part of
Gödel's proof.
Sorry, but you miss that the problem with the paper has nothing to
do with
the paper itself or logic.
It's problem is only in that we think it shows something it does not
(and
can not due to the basic limitations of the axioms of arithmetic).
The conclusion that Gödel draws does not eliminate the meta-level.
It just
eliminates the meta-level (the language that is embedded in
arithmetic) on a
meta-meta-level (the language that we use to make sense of this
embedding,
eg english).
Yes, it is correct on a meta-meta-level, but in this case it is a
statement
about the meta-meta-level and arithmetic, not arithmetic itself.
The metalevel needed in Gödel's proof is entirely embedded, so there
is a collapse of all meta-meta-... in the level of elementary
arithmetic. The machine does proof "if I am consistent then I cannot
prove it".
And yes, we can eliminate the meta-meta-level by embedding that in
arithmetic as well, but then we use a meta-meta-meta-level to make
sense of
that.
Not at all. We use only the axiom of arithmetic, and the inference
rule. We don't interpret it as we never do hen working *in* the
theory, and in this case also *on* the theory.
We just create more meta by embedding the meta in arithmetic.
I disagree.
What I wrote may be confusing, but read it a few times, maybe then
you will
get what I was getting at all along.
I think I get it, and disagree.
Bruno Marchal wrote:
You act like Gödel "eliminates" the meta-level, but he does not do
this and
indeed the notion of doing that doesn't make sense (because
otherwise the
whole reasoning ceases to make sense).
Gödel does not eliminate the metalevel. On the contrary it shows that
machines or formal theory can access to it.
That is precisly what I meant by "eliminating the meta-level". It
can only
access it on a meta-level, it itself can not access it (the notion is
meaningless, because it has no axioms to access it, except on a meta-
level).
You are just wrong here.
If you see Gödels paper as a philosphical paper it is actually very
valid
IMO - he express what we intuitively know about math (there are some
things
about every axiomatic system that the axiomatic system does not
capture)
using arithmetic. But he doesn't really prove it within arithmetic
(except
on a meta-level), just *with* arithmetic.
Not at all. It is a statement of any sigma_1 complete part of
arithmetic on itself. You critics here is equivalent with the
statement that a machine cannot think without an observer looking at
the machine. It is again just a sophisticate way to defend non-comp.
Bruno Marchal wrote:
Bruno Marchal wrote:
You just use fancy words to obfuscate that.
It i#s like saying "study the bible for scientific education (you
just don't
understand how it adresses scientific questiosn yet)".
No reason to be angry. It is the second time you make an ad hominem
remark. I try to ignore that.
I am not angry, just a little frustrated that you don't see how you
ignore
the main issue (both in our discussions and you paer), while acting
like you
are only showing rational consequences of some belief.
I am not acting like. This is what I do.
Well, maybe you think you do it, but your paper and how you defend
it here
shows something different - that you using a specific philosophical
approach
(which you can't show to be accurate rationally, or even put into a
clear
assumption)
Saying yes to a doctor needs philosophy or theology. That is obvious.
but comp makes possible to reason (and only reason) from there.
and are emotionally invested in the paper (why bother to defend
with so much effort else).
One cent ad hominem psychology and out of the topic.
Again, I am just being frank here and not intending to attack you
personally.
It is impossible to answer remark like that. Just find a flaw, or a
missing assumption. Here you have just show that you miss the proof by
Gödel.
This concerns only AUDA. It is not used in UDA, and we might
concentrate on it. But your ad hominem remark and your negative mood
makes me think that you are not really interested in the subject
matter. Indeed you already assert that comp is wrong, even that CT is
wrong.
You have admitted not studying the papers nor Gödel, so we are going
nowhere.
Bruno Marchal wrote:
I have said nothing about you, actually you seem to be a genuine,
open and
nice person to me. I am just being honest about what you appear to
be doing
in your paper and on this list. It is probably not even intentional
at all.
So, sorry if I offended you, but I'd rather be frank than to argue
with your
points which don't even adress the issue (which is what perceive as
being
obfuscation).
What you call "obfuscation" is just the originality. I take a problem
usually addressed by philosopher or theologian, and I show that if we
assume comp, we can derive testable conclusion.
I know that some philosophers are sick at this, but that is a
tradition in human history. This is discussed in other threads.
I feel that you getting a bit emotional here (not that there is
something
wrong with that at all, I just wanted to mention it).
Calling it "obfuscation" was a little mean, because it implies that
it is
intentional. Sorry.
OK.
Your paper is just not very clear,
You can ask for clarity for each step. The paper, like all technical
papers contains references for many details. You have to read them also.
It does not make sense to say that something is unclear without being
specific.
and the many debates regarding it cofirm
that.
Which debate?
Note that this is not your fault, IMO there is no clear way of
showing what
you wanted to show. In fact you did a really good job expressing it
as clear
as possible. It just not clear enough, because the subject matter is
not
clear.
We can reason clearly, even on non clear subject matter, thanks to
precise definition. Comp, as I define it is clearer than anything in
the literature, which never make clear the assumption of the level,
usually taken for granted. Then I explain that a machine cannot know
her substitution level, and that whatever she experiences will be
dependent on a measure on its consistent continuations.
Bruno Marchal wrote:
but that's not the issue.
I illustrate the contrary. You seem to be angry against science or
theory in general, and displeased that I use the scientific
methodology in something belonging *traditionally" to religion.
No, I am not angry wth science and theory. I am just not thinking
highly at
all about pretending it is something holy that is telling us what to
think
and what to do and how to view the world, in the same way that I am
not
thinking highly of dogmatic religion. It is harmful IMO and not
conducive to
our well-being.
?
But scienc, well understoodf, never do that. I give a theory, that is
I put my assumptions on the table, and get consequences, and means of
testing them.
I am only slightly displeased that you act a bit like your reasoning
is
beyond doubt by dogmatically rejecting every argument against it
(simply
claiming it does not show a flaw, and then asking where the flaw is).
I will not say that there is a flaw just for the pleasure. It is those
who think there is a flaw who must show it.
I have never said that there is no flaw, so to accuse me of dogmatism
is unfair.
You
say that the argument is wrong and that's it - "read the
paper" (...and you
will see that you are wrong).
I say only: read the paper, and tell us what is wrong *specifically*.
There is no technical flaw in it, because the flaw is that it is not a
precise argument in the first place
What is not precise? The comp assumption is precise. You just repeat
it is not, but you don't say what is not precise.
(vague assumption=vague or wrong
reasoning=vague or wrong conclusion). You never clearly define what
YES
doctor means.
It means saying "yes" to a doctor proposing you an artificial digital
brain, copying your brain at some level.
It is the "essence" of a machine, that we can replace part without
loosing the functionallity.
Bruno Marchal wrote:
The ideally correct machine is shown, by using the most standard
antic
definitions, to have a genuine theology, but you remain free to
consider them as zombie, if that is your theory.
I don't consider *anything* zombie, even the most inert matter.
Machines
just don't have clearly determined points of view, IMO.
But I think most persons don't have one either, which is maybe too
radical
to even seriously propose (without being ridiculed or be met with
hostility).
You are not quite clear,
Bruno
benjayk
--
View this message in context:
http://old.nabble.com/Simple-proof-that-our-intelligence-transcends-that-of-computers-tp34330236p34413396.html
Sent from the Everything List mailing list archive at Nabble.com.
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to [email protected]
.
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
.
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
