On 25 May 2015, at 13:53, Pierz wrote:
On Monday, May 25, 2015 at 4:10:37 AM UTC+10, Bruno Marchal wrote:
On 23 May 2015, at 12:36, LizR wrote:
I'm not sure why comp would predict that physical laws are
invariant for all observers
I can see that it would lead to a sort of super-anthropic-selection
effect, but surely all possible observers should exist somewhere in
arithmetic, including ones who observe different physics (that is
compatible with their existence) ?
Those with different physics will have measure zero. Why? Because
the laws of physics must be given by the sum on all computations
below the substitution level, whatever any universal machine state
can be in. Only geography will need the anthropic element, the
physics needs only a mathematical statistics on all computation,
going in "actual state" which are any state.
Physics become a theorem of machine theology, itself a theorem of
arithmetic (+ comp).
Of course, today, we don't know how much the "standard model" is
contingent or absolute. String theory diminish a large part of the
contingent parts, but introduces complexity in other direction, with
panorama of different sorts of physics. All this are open problem in
comp.
The goal of comp is to provide an explanation of the relation
between consciousness/mind and appearance of matter and persistence,
and this in some testable way. It is an explanation in the form of
the formulation of a problem, or a reduction of a problem into
another one.
Bruno
OK the 'invariant physics' you refer to is a very low-level one,
i.e., the ultimate unifying laws. However, let me try to put my
point about the substitution level another way to make it clearer.
ISTM that it will be sufficient for there to be another region of
the multiverse where the observable, everyday physics (things like
particle masses etc) are different in order to force us to conclude
that the substitution level must be as large as the observer's
universe. Why? Because let's say there is an observer in another
region of the multiverse whom I wish to duplicate here in *this*
region, where the observable laws are different. I get a copy of
that observer's memories etc and reproduce them here. Now suddenly
that observer has a continuation that experiences the physics of
*this* region. But if that was possible, then the physics of both
regions would have merged,
Not necessarily. It might be that we need to emulate locally, perhaps
with a quantum computer, his own physical laws, and that we interface
it properly, which should be possible, given comp is suppose to apply
to him.
because observers in both regions would already be able to interfere
with one another's measures.
Not necessarily, because the two "region" might have undergo deep long
histories, and the linear multiplication would need white rabbit type
of event for having the interference.
SO, in order to prevent such interference of measures and a merging
of physics into an average measure, it must be that in order to
duplicate an observer in another region, I need to duplicate the
observer to such a deep level that the separation into the two
regions is as it were enforced by the very definition of that
observer. The observer must include the entire computational branch
down to the point at which the two physics diverged.
I am not convinced. Interference needs very similar experience, or
amnesia, but in this case, the other region might have a physics which
can't interfere with us, because they have separated a long "time
ago", and are too much different.
I use the trivia that above the substitution level we are Turing
emulable.
I might miss your point, also. The idea of inviting someone obeying to
different physical laws here is not so easy to imagine. Comp would
allow something like that, but only by simulating well enough the
"different physical laws" locally.
Bruno
On 23 May 2015 at 21:23, Pierz <[email protected]> wrote:
Some time ago on this list I had a fascinating exchange with Bruno
that has stayed with me, fuelling some attacks of 4am philosophical
insomnia - an affliction I imagine I'm not the only person on this
list to suffer from! If you try to nail Bruno down on some aspects
of his theory, he has a tendency to get all Sg Grz* and p[]<>p on
you at a certain point, making it difficult to progress without a
PhD in modal logic - despite the fact that I suspect that the ideas
are fundamentally simple. Nevertheless in the course of the
discussion, Bruno did acknowledge that his theory predicts that the
laws of physics are invariant across space and time, because they
are supposed to arise out of pure arithmetic (being the hypostases
of the FPI bla blas). Indeed, for the dissolution of the material
within the arithmetical to go through (logically), then the
regularities that we call physical law cannot depend on geography,
since ex hypothesi they arise from number relations which are prior
to time and space. Yet physics - or cosmology - seems to be headed
full-steam in a different direction, towards the conclusion that
physical law is indeed dependent on geography, the laws we observe
being dependent upon an observer selection process. That is, we see
physical laws finely honed for life, because life can only exist in
those regions where the laws are conducive to life. I'm less sure
about this, but I think it might still be OK for physical law to
geographically determined in this sense, so long as there are no
other observers in different parts of the multiverse who see
different laws, but to assume such a thing seems foolish. Why
should we believe that of all the possible permutations of the
parameters which determined physical, there is only a single
solution which permits life? There might be many different
So on the face of it, the recent measurements of the mass of the
Higgs boson, which are strongly suggestive of a multiverse might be
seen as empirical evidence against 'comp'. Yet there is a way -
namely an extremely low substitution level. You'll recall that the
substitution level is the level at which a digital substitute can
be made for a brain such that the self (whatever that is) survives
the substitution. This might be quite high - perhaps its sufficient
to mimic neuronal interconnections in software? Or it might be very
low - maybe we need to go down to the molecular level and simulate
chemistry. However, it would be a big surprise I imagine for the
digital survival enthusiasts if the required level was the entire
multiverse! Yet that conclusion seems inescapable if the emerging
multiverse cosmology (and comp) is correct.
Why would a low substitution level save the day for comp? Because,
as stated before, if the physics observed by some conscious being
is dependent solely on number relations (as UDA purports to
prove), and number relations are pure abstractions prior to
matter, space and time, then physics cannot be contingent on
geography, because it is contingent on matter, space and time. So
if comp is correct, and it is also correct that we live in a
multiverse such that observers see different apparent laws in
different parts of that structure, then the only solution (ISTM) is
to make the observer large enough to encompass the geographical
variation.
But such a low substitution level seems counter to most of the
common sense assumptions about consciousness that are the basis for
the logic of UDA seeming plausible at all. It would commit us to
the idea that teleportation of the 'same' consciousness from
Washington to Helsinki is impossible, because we couldn't isolate
the person's consciousness within any reasonable physical limits,
such as their brain or body. We'd need to substitute the entirety
of everything, including Helsinki and Washington themselves! But
what then is the status of a teleported person, if such a thing
could be achieved? If we reassemble the exact same organization of
molecules such that nobody, not even the person, could tell the
difference, then how has the substitution level *not* been achieved?
Perhaps the answer to the conundrum lies in the definition of
physical law? Perhaps things like the cosmological constant, the
masses and charges of particles and so on, which I would normally
regard as aspects of the laws of physics (and which recent results
suggest may not be the same in all parts of the multiverse) are not
the real laws of physics. Rather it is the deeper laws which
underly those geographically contingent apparent laws which are the
true laws of physics, and which derive from number relations.
However, that manoeuvre won't save us, because then in order for an
observer to experience a certain set of apparent physical laws, I
need to specify within which branch of computations (multiverse
region) I am instantiating that observer. That is the same as
saying that the substitution level is very, very low, because in
order to duplicate an observer, I need to duplicate the entire
universe-generating computational branch that they are in, not just
their personal memories and so on.
Then again, maybe I should not be surprised by this substitution
level, because if the wave function is the manifestation of my
computational duplicates and their relative measure, then any
genuine duplicate of me would be part of that quantum wave function
measure, and making a copy of me in the same universe would not
have that effect. Only if I could make a fungible duplicate of the
universe and insert it into the deck would I be able to influence
physics and make a real substitution rather than a poor copy.
This suggests to me that either comp has overlooked something about
the nature of consciousness and is wrong, or that cosmology is
wrong and there is only one physics everywhere, or that, even
though comp is right, the artificial duplication of consciousness
is impossible because consciousness is determined by its
relationship with the entirety of existence, perhaps in much the
same way that the wave function of an electron has to "know about"
all the other electrons in existence in order to obey the exclusion
principle. The whole within the part and all that jazz.
Or I missed something, not that that has ever happened before... ;)
--
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.
--
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.
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.
