On 17 Dec 2011, at 23:52, Russell Standish wrote:

On Sat, Dec 17, 2011 at 11:59:07AM +0100, Bruno Marchal wrote:

... snip ...

In a multiverse, the
counterfactuals are realised, but in different branches.

Not necessarily. If the computation is classical, it is the same in
the normal continuations. The classical counterfactuals are not
realized in a quantum multiverse. You have to put the Klara in some
superposition state to do that. You need a quantum Olympia.


They will automatically be in superposition, being just a classical
device replicated across the branches. But they are not quantum
devices (in the sense of processing qubits).

But it will be in the same state across the multiverse. If different, then it is different computations, and different consciousness, and we are no more in the situation of the argument.







Hence those
"inert" parts are no longer inert in all branches.

I don't see this.


Consider an instruction that monitors the circular polarisation of a
photon. If the photon is left polarised, then the program branches, if
right, the program continues to the next instruction.

OK, we then we are in the case of quantum superposition (or alike).



In Maudlin's initial run, suppose the program didn't branch (we
now eliminate from consideration the MW branch in which the program
did branch prior to the construction of Olympia).

Now Maudlin construct an Olympia with without branching at the
point. Or we do your step of replacing the branch instruction by a no
operation, as the result will be the same. The we run Olympia, with a
Klara attached at that step.

In the Multiverse branches where the photon is right polarised,
Olympia continues on, and the Klara remains inert. In the branch where
a left polarised photon is observed, the Klara springs to life, and
implements the missing branch instruction.

This is what I mean. At all stages, Klara and Olympia are classical
computing devices, embedded in a Multiverse.

Consciousness, with comp relies on interaction in a branch. The other branch can change statistics on the computations/continuation, but on the presence of other branch, unless the brain is a quantum computer, but this only means that the level is lower than usual, and we have to simulate it in a single branch to do the MGA or Olympia reasoning again.






If they were, they
could be removed from the computer altogether, without affecting the
computation.

But that is the case for the computation under consideration.


No. See above.

Either the mutiverse needed for that type of physical supervenience is Turing emulable, or it is not.
If it is, we can do the MG reasoning, if it is not, comp is false.








So you are introducing a different kind of physical multiverse,
which would handle the counterfactuals. But this will not work.
Either this physical multiverse, which plays the role of the
generalized brain, is Turing emulable, in which case I can emulate
it in a single Turing machine, for which the MGA will apply again.
Or it is not Turing emulable, but then the need of it will
contradict the comp assumption.


This step, as I understand it, is a form of dovetailing. Nobody really thinks of the dovetailer algorithm as instantiating consciousness, so
the move is ultimately invalid, I would think.

The problem is there. With comp + the physical supervenience thesis,
the dovetailing algorithm does instantiate consciousness.


The dovetailer instantiates consciousness in exactly the same way that
a random block of marbles instantiates the statue of David.

The dovetailer generates precise programs, and run it in the precise computer science sense. Nobody has ever show me one instance of a program executed by a rock, except for little trivial program. Actually, nobody has succeeded to tell me what a rock is, or in what sense that exists. The universal dovetailer generates immensely big programs, which does not make sense in most possible conception of rocks that I know. That the statue of David is in a random marble of block is irrelevant for the issue of consciousness relate to program execution.




I think that for most people call a shapeless block of marble 'David'
is a bit perverse.

Sure. But that the UD run the many lives of Russell is a correct consequences of the comp Hyp.



It is not a question of the parallel realities playing a role in the
computation, but in the supervenience. Maudlin's argument says If
COMP, then supervenience on single universe is contradictory. But it
doesn't say anything about supervenience on multiple parallel
realities.

Those are relevant for the relative measure on continuations.

Quite possibly. But that is an independent question to Maudlin's
argument. I'm trying to stay focussed here.

My point is just that the use of the multiverse does not change the consequence of the MGA reasoning.





If you then fold the multiverse back into a
single universe by dovetailing, one can then reapply the Maudlin
move.

Indeed. That is the key point.



But then, in that case, one can embed that result into a
Multiverse, and the cycle repeats.


I think I'm coming around to the view that neither of the above steps are valid - but one could equally say they are as valid as each other.

Not sure I see which steps you are talking about. The MGA is a
reductio ad absurdum from comp + physical supervenience.

The step of folding a Multiverse into single universe by
dovetailing, followed by reembedding the single universe back into
another multiverse. It is what we've been discussing, but is _not_
part of either MGA nor Maudlin.

Indeed. We just show that comp + physical supervenience are contradictory. But you are the one introducing a multiverse to object on the validity of the reasoning, and my point is that such a move does not work because multiverse, if needed, can be emulated in single computation, where we can do Maudlin's or MGA reasoning again.





...snip...


This is not
different than the comp or quantum immortality argument. The fact
remains: the physical activity in one normal branch missing the
register is the same as the physical activity in some branch not
missing it, for the same particular computation.

In all branches, or just special ones? If all branches, then the
register is totally unnecessary.

In this case the same computations, with the same inputs are done in
all branches.


But then, this is not a Multiverse. By definition, a multiverse's
branches will be distunguished by the inputs.

That depends on many things. If I take a lift, I will take a lift in the vast majority of my consistent extensions.






If just a special pair of branches,
then Maudlin's argument shows that supervenience must occur across
more branches than those two.

There is no more branches. We are now simulating all the branches in
a single reality. If that is not possible, then comp is already
false.


We're talking past each other here...

...snip...

OK, but then to make your argument you have to shift toward a multi-
multiverse, given that we have come back to a single universe. And
the argument can continue: I will just simulate that
multi-multiverse in a single branch on a single classical universal
machine. If that is not possible, then comp is false. If that is
possible, then MGA will apply again.


As I said above, in simulating the multiple branches of the Multiverse
by dovetailing, we are no longer instantiating consciousness.

I don't follow you on this. The first person is not even aware of the giant delays brought by the dovetailing procedure. The UD (at first concrete) does instantiate consciousness (and "at time" with the physical supervenience thesis). It is all what we need for getting the (epistemological) contradiction.





(Just
like all blocks of marble are not David.)

It is different. There is no supervenience of experience in this case.




This step is basically
invalid. It does not imply COMP is false, though.

Of course, if you can think of another way of simulating a multiverse
within a single universe, I wil naturally reconsider...

The way of simulating it changes nothing. It is the point of the first six step of UDA.







But this doesn't answer the why question. I could imagine that you
might feel that Multiverses are otiose, so would prefer a derivation
of their existence from something "simpler" - eg arithmetic of the
whole numbers.

Not at all. It is the idea that there is primary matter which is
otiose, or epistemologically contradictory. It is physicalism which
is show wrong. The multiverse is shown to be emergent from a numbers
multi-dream. Physics becomes a branch of machine's theology.




That's fine and dandy - but the Multiverse is not otiose - it is far
less of an impost than a single reality.

Yes. We agree on this. It is the main theme of this list. The
question is not about their existence, but their primitivity.


Actually, I have no problem with this. I am quite persuaded by Kant's
concept of the unknowable noumenon to apply Laplace's "Sire, je n'ai
besoin de cet hypothese" to the whole issue of primitive reality. I
don't need the MGA to conclude that. But others seem to require a
primitive "something" to exist, and are most upset at
immateriality.

Yes. The idea is to explain that the belief in a primitive physical reality makes no sense once we bet on digital mechanism. Comp makes physical reality emerge already from the additive and multiplicative number relation. It explains what it is a complete coupling consciousness/physical-reality which is emerging, and consciousness has a key role in that emergence. (Not human consciousness, but universal number consciousness).




I'm happy to work out what is the limit of our
knowledge and move on from there. If any universal system can
reproduce phenomena, then that suffices.

I agree. But it took me 30 years to convince die hard materialist. So it is not obvious. That's why we need a proof/argument.



It is a meaningless question
to ask "which universal system" - it could be all of them, or even nothing
at all.

We need at least one universal system.
UDA (including step 8) shows that the laws of physics are independent of the choice of the universal system, because physics (below our subst level) is a sum of the works/dreams of all universal machines. This gives, together with self-referential correctness, enough precise constraints (by UDA) to (begin) the derivation of the laws of physics (AUDA).





Coming back to the MGA - it would be interesting to know whether your
movie graph construction escapes the multiverse embdding move. Assume
your filmed graph has had most of its nodes removed, but by
coincidence, a supernova sends a blast of photons that exactly
reproduce the graph in one of the MW branches. In the vast majority of
branches, the filmed graph is a dead as a doornail.

Does this render the graph conscious in that single lucky branch?

Yes (in case of comp + physical supervenience). But no branch is single. All branch have 2^aleph_zero counterparts. Branches can be rare relatively to other, in measure term, but all have the power of the continuum, in the first person sense (despite everything is countable in the (big) third person picture). This is trivial, because for all programs, the UD, stupidly enough, re-execute it by dovetailing it dumbly on all initial segment of all real numbers (which is a continuum in the first person perspective).




I think this question bears on the issue of Boltzmann brains too,
something we haven't discussed enough here.

The UD makes this irrelevant, or trivial. We are in infinities of digital Boltzmann computers. We belong to infinite layers of universal dreams. Matter is a first person plural mind construct. Boltzmann idea can be see as a precursor of the UD type of argument. It is Church thesis and the notion of universal machine which makes that idea precise.

Best,

Bruno

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 everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/everything-list?hl=en.

Reply via email to