On 07 Aug 2014, at 21:58, meekerdb wrote:
On 8/7/2014 12:34 PM, Bruno Marchal wrote:
Comp gives the 3p identity criterium, from the level of
susbstitution you bet on.
What does level of substitution have to do with it. You've defined
the correct level substitution as one below which consciousness is
unchanged. So the the definition already requires that we know
what it means for a consciousness to be the same, i.e. unchanged.
We don't need that definition.
I didn't say we needed a definition. I said we needed to know what
it means for one's consciousness to be the same.
I thought the question was if the person remains the same. With comp,
the teleportation suppose that the level chosen is such that all
memories are preserved, and we define the identity of the person by
its memories. It is "legal" third person definition of identity, which
might be signed by Obama and the governor. We agree that John Clark
from helsinki, is still alive after the duplication, and he lived now
in Washington, and in Moscow. For example. You keep your identity in
all duplication: it is your consciousness which differentiate. The
first person experience, on the contrary, feels to be unique and like
being selected randomly (for the vast majority). But if they have bet
on comp, and understand comp, they know that is has to be like that.
In that case, all the copies are all the same persons. The HWWMMMMMW-
John Clark has the legal right to claim being John Clark, but when
playing chess with his fellow MMWWMMWMWW-John Clark, he knows that he
is not that other guy for now, because, despite they share a initial
life memory, that life has differentiated in different histories since
that (reiterated) self-multiplication.
Then, if you agree that we can consider all the John Clark
reconstitutions, as genuine same person (Our friend John Clark),
despite their multiple variate first person views, you can understand
that we might understand that in fact we are all the same initial
amoeba, and even the same initial universal system, which seems to
need that 'blanche consciousness/conscience state", which would be the
initial differentiating consciousness.
We don't need it no more than we need a definition of "staying
alive" when accepting an heart operation, or accepting taking a
plane, or sleeping a night. We need to bet on a level, but no more
than that.
But we need to know whether or not we've won the bet, otherwise it's
meaningless - one level is as good as another.
There is a fundamental difference between surviving intact, and
surviving tout court.
By definition of comp, the right level is the one at which your
survive intact.
It is just that for discussing the number of person, we need the
thought experiment/experience with amnesy, and lost of things.
All that is not needed in UDA, where I avoid such thought experiments,
but in my notes I have myriad of them.
Then the 1p recognizance criteria is very simple. All the copies
are declared the same person as the one copied, except that now
they differentiate.
And that can help to understand that we are all already the same
person, having multiplied and differentiate.
I guess it would if I were a copy of somebody.
You are plausibly already a sort of half copy of your father + half
copy of your mother, with some variance.
But that's just playing with words. I my father's dead, but I'm not
half-dead.
It is not. It means that sexuality is comp ways to accelerate the
differentiation. Nature does emulate, at the least, universal genetic
algorithm. Very simple one, the bacteria Escherichia Coli's genome, is
already a monster of complexity, and it is provably Turing universal,
as enough variate and long enough DNA is.
Plausibly nature bet on comp too, as with the genetic code, which
is a universal number relatively to the turing universal carbon
chemistry (in some condition of pressure, temperature and humidity).
Then computer science shows that there is indeed a universal
notion of first person, given by the knowability (non arithmetic)
operator, provided by the definition of knowledge by Theaetetus.
Having a notion of person doesn't imply that all persons are the
same person.
Correct. And the number of persons, nor the notion of personal
identity is relevant to understand the reversal.
I think it's relevant because it relies on maintaining consciousness
and personal identity, whether that requires memory or not.
I disagree. To understand the FPI and the reversal, you need only bet
on a level for surviving intact. Then we reason in protocols where the
level is correct, like in the math part, we start bottom up, and
define the correct level at the start, as we build/define the machines.
By definition of comp, there is a level where you survive in the same
sense that you survive when drinking a cup of tea.
But to understand the translation in arithmetic, we need to define
the 3p self. This is done with the usual second theorem of Kleene,
which I explained the basic by the formal diagonal applied to
itself (D"X" = "X"X"", and variants). This leads mathematically to
the logic G and G* (by Gödel, Löb, Solovay).
Then incompleteness makes Theaetetus' definition of
"knowability" (say) working.
I don't see that incompleteness makes Theaetetus definition work. I
would work as well (or poorly) in a complete theory too.
Not if that theory is Turing universal, which is the case where []p &
p = []p becomes non justifiable, non believable by the rational
machine, even if true (proved by G*).
It captures some aspects of "know", but I think it misses the aspect
of causal relation.
You can define "p cause q" by many variant of [](p -> q) in each
hypostases. "causality" is in the mind of the observor.
Comp does not refute physics, but it refutes physicalism. Notion of
cause, must be (re)defined in non physicalist terms.
It models knowledge as only provable theorems (from some axioms).
No. Thats the beliefs. Knowledge conjunct the belief with the truth
notion. That is no more representable in arithmetic, or in the
language of the machine. It obeys a completely different logic.
It implies that knowledge is not particular to a person.
On the contrary, the truth of the self-referencial beliefs, relative
to the machine's inputs, will particularize the person.
Yet it is obvious that some people know things others don't.
yes, the HW-Clark knows something that the HM-Clark does not know, but
can only imagine.
With "'[]p" obeying to G, the new operator "[-]p, defined by []p &
p (in the arithmetical intepretation of "[]", that is Gödel's
beweisbar predicate, and p some arithmetical sentences (for matter,
we restrict them on the sigma_1 sentences).
And we get the miracle needed, which saves both the modern (who
likes []p & p), and the ancients mystic/rationalist/platonist who
like the idea, or respect the introspective data, that the knowing
is non propositional, and the knower being not a machine, and being
not definable.
Gerson thinks that by defining knowledge by true belief, we make
knowledge as a sort of particular belief, and that it makes
knowledge representational. With comp, that is partially true, in
some sense, but only in G*. It makes no sense from its first person
point of view, and correctly so. Gerson confuse the informal "[]p"
and p, with the formal "[]p & p", which for "Tarski-Gödel-
Epimenides" reason, get no more definable in the language of the
universal machine concerned.
Like Judson Webb shows that incompleteness protects Church's
Thesis, incompleteness protects the soul from any possible 3p
definition. With comp, you can luckily bet on a correct sublevel
(too much grained) but in no way you can prove that you get the
precise correct description of yourself capable of supporting your
soul.
What would it mean for a description to support my soul? That sound
like mere poetic metaphor.
?
It is comp. Take the movie of the graph. It does not support a
computation, nor a soul. But each of the movie image can be projected
on a boolean graph, and start the computation, making consciousness
"there" again. Now, with comp, the consciousness is always in
Platonia, and it is only the effective turing universality which makes
the description of the computation supporting the soul.
Buy the description supporting your soul, I meant only the description
of the brain at the correct substitution level, such that its
reimplementation makes its soul re-enacted relatively to you and the
universal systems which support you and the computer enacting your
brain.
If you could, you would be able to prove that []p -> []p & p, and,
although G* can do that, G you can't, and you can't.
Is the S4Grz person a universal person. Assuredly.
Then who are all these people?
The people who does not lie, and are correct. Only God know them. With
comp, we don't know them, but we can evaluate plausibility.
I am pretty sure PA is one of them. A bit less sure for ZF.
All (arithmetically) correct löbian machines define such persons.
Is it *the* universal person?
On your definition of person it would seem that any two universal
persons would have to be the same person.
I just ask the question. I find it reasonable in the WM-duplication-
with-prolongation: where someone is duplicated only once, at the age
of 6, and the copies live their life entirely where they have been
reconstituted. I just say that if you believe that at the age of 90,
they are still the same person (as comp asks for the legal level),
then you might consider that we might be the same person, in the
larger sense involving amnesia (and not needed in UDA).
We work on the mind-body problem, and we have understood (I hope) that
the usual brain-mind identity link is broken, in both comp, but also
in Everett QM.
Bruno
Brent
That is not needed neither for auda, nor of course for uda, but
although I doubt this could make sense, I am not that sure. Thanks
to salvia, I have one more doubt.
Bruno
--
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 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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
