Re: Multiverse Morality

[My gmail seems to sometimes produce bad indentation --I'll write this one
in Emacs.]


On Wed, Aug 28, 2013 at 6:22 PM, hibbsa <hibbsa-/[email protected]> wrote:
...

> On that matter - the tense that you are raising here with your chaos
> example of macroscopic amplification - my basic position has been
> that I am saying exactly the same thing that MWI itself
> says. Namely, that the multiverse represents all the possible
> outcomes, in the same way that, within a one-universe model - a
> statistical model of the possibilities would say the same thing. As
> such, the multiverse could be conceptualized as a statistical
> model. Albeit one that cannot be derived for real.

You have it just right; [quantum] probability in the single universe
we see is the same as measure in the multiverse.  I'd only say that
the multiverse *can* in fact be derived for real; that derivation is a
straightforward application of the SWE.

> In that context, what I am challenging, is whether the multiverse -
> in context of the divergence from some ancestor world, can ever
> produce novelty that, were that novelty to have occurred in the
> ancestor world, it would have amounted to a statistically
> significant change in that ancestor world.

OK, let's unpack that.  We have an ancestor world A.  (By "world" I
presume we're talking about a bounded region of spacetime.)  It
produces several successors: B, C, D, and so on.  (Note that the time
bounds of these are later than A's time bounds, otherwise A wouldn't
be their ancestor.)  Let's take B, which we know to be quite different
from A.  I think what you're saying is C, D and the others are very
much like each other, but B is different (it has "diverged"
macroscopically from its siblings).  Of course none of them are much
like A, since they're all later and things have presumably changed over
time.
So the multiverse produced novelty
in the form of B.  So far, so good.  Can you walk through the
counterfactual part ("were that novelty to have occurred...") in this
example?  Are you asking how that novelty could get from B (a small
fraction of worlds) to become a large fraction of worlds later on, so,
say, the grandchildren of B would come to dominate over the
grandchildren of C, D, etc.?  Or something different?  Feel free to rework
my example to show your point.

> But the problem I have with applying that sort of theory within the
> context of emergence, is that....what if there are 100 or 100 fully
> distinct emergent levels in our universe.

I disagree with this.  I think each concept (entropy, intelligence,
macroscopic objects, etc.) has its own emergent properties and layers
of emergence, and they're quite fuzzy; there's no way to slice that
across all concepts so there are 100 "fully distinct emergent levels".

> So far as I know, there are no properly worked through variants on
> chaos theory, that describe the effect in terms of that emergence.

This could be a good area for research perhaps?  Chaos theory itself
is quite well researched and worked through mathematically.
Modern communication theory for instance would be impossible without
detailed models of chaotic behavior.  And of course chaotic behavior
is strongly emergent; fractals are the obvious example, but so is
period-doubling and various kinds of attractors.  But I'm sure there's
more to be figured out.

> I don't even think there are yet any hard scientific models of
> emergence that have made it all the way to mathematics.

Isn't that exactly what thermodynamics is?  A model of how quantum
random behavior leads to stable (emergent) chemistry and physics?

And on the intelligent-behavior-emergence front, there's a lot of a-life
work
that could be seen as relevant (producing "intelligent" behavior
by evolving random genotypes for instance -- there's a lot of work
done around that kind of thing.)

> In which case, it's hard to see how it can currently be regarded as
> reliable, robust, reasoning, to speak of such effects within a
> context of emergence at all.

You may have something different in mind from standard emergent
properties here.

> Beyond that, as I mentioned to you in a private mail a few weeks or
> months back, the issue I am raising here is not primarily about MWI
> itself, but about Deutsch's specific variant of MWI, which heavily
> involves the concept of fungibility. Emergence isn't directly
> linked, but he does describe a personal theory about emergence in
> BoI. So the issue I'm raising is really about whether these three
> concepts all work together properly, or whether there is a problem.

> For example, can you or anyone, explain how worlds can be fungible -
> which Deutsch describles as being literally in the same dimensional
> 'place'

Well, I don't know about Deutsch (I can't make it through BoI, it's
too loosely argued) but fungibility is quite a standard term in
physics.  It just means completely indistinguishable.  Two photons are
fungible if they can be swapped without any effect.  In the MWI, an
Electron which can be spin-up or spin-down has an infinite (or at
least huge) number of spin-up little-e electrons and the same number
(really equal measure) of spin-down little-e electrons.  All the
spin-up ones are fungible with each other, all the spin-down ones are
fungible with each other, but the spin-up ones have a different
property from the spin-down ones so they are not fungible with each
other.

The canonical example is a Bose-Einstein condensate.  Google that for
more.

Macroscopically I think the concept of fungibility gets fuzzier, but
it's certainly possible to say "...in all the worlds in which I do the
Schroedinger's Cat experiment..." and mean that all those worlds,
which may be microscopically variant, are FAPP fungible up to the
point of the photon being emitted (in some of them).  There's no real
difference between them, you can't tell which one you're in, they all
have the cat, the vial, the experimenter, etc.

But of course that macroscopic "FAPP" fungibility doesn't mean
literally fungible in the QM sense.  Of course there are many
macroscopic fully-fungible regions of spacetime, but at the macro
level physicists usually don't care about where particular air
molecules are.  It's another form of emergence I guess: what counts as
fungible depends on the question.  (Just like two dollar bills are
fungible when considering their value, but not when considering their
history, pattern of folds, or serial numbers.)

> ....and not just at the quantum level but at all levels. How
> does that work, using Deutsch's description of emergence which
> explicitly rejects a purely bottom-up determinism in terms of
> emergences.

> The implication being, that macroscopic levels - supposedly fungible
> - can potentially be influenced by top-down effects, which
> presumably in some possible instances result in changes at the
> macroscopic level. Which presumably would violate fungibility. Which
> so far as I can see, would have to be see as a refutation of the
> concept of fungibility itself, unless Deutsch has an explanation how
> this can nevcer happen.....which in turn maintains consistency with
> his explanation of emergence itself.

I'm afraid I get lost in these last two paragraphs -- I'd need to
understand Deutsch's concepts better I guess.


-- 
Gary


[Non-text portions of this message have been removed]