On 07-05-2016 08:28, Brent Meeker wrote:
On 5/6/2016 10:51 PM, smitra wrote:
On 07-05-2016 02:36, Bruce Kellett wrote:
The use of the relative orientation angle theta is intrinsically
non-local. That angle cannot be obtained by local means in the above
derivation. The equation for |psi> derived above shows the full
coherent wave function as evolved from the initial state according
Schrödinger's equation. There is nothing else -- no more worlds or
dopplegangers than the four explicitly shown. The observers can only
differentiate into one of these four worlds. And that is correct --
it
is in agreement with experience. But it is still non-local.
It is wrong to invoke this angle in this way in the MWI. While it
leads to the correct answer, one has to consider that the evolution of
the state vector is still due to local dynamics. It's therefore a
trivial fact that there cannot be any non-local effects here.
The illusion of a non-local effect comes from cutting corners in the
derivation by assuming that there exists a macroscopic Alice here with
some polarizer setting and a macroscopic Bob over there with some
other polarizer setting and then we can can compute the correlations
by just applying the usual formalism. And then we make hidden
assumptions based on the classical behavior of Alice, Bob and the
polarizers as they are macroscopic. That sounds reasonable, it also
yields the correct answer but it's still wrong as a description of the
physical situation according to the MWI.
A correct MWI derivation must involve working with a wavefunction that
evolves under unitary time evolution.
But that wavefunction is a function of different points in space, some
of which are spacelike separate. The wf dynamically evolves the
probabilistic location of the singlet particles to their interaction
with the polarizers and detectors. The interaction at the polarizers
changes the wf at other locations. In the usual formalism this change
is instantaneous, i.e. spacelike. If it's not instantaneous, as Rubin
argued, then it must propagate within the forward lightcone and the
reduced wf only describes the correlation in the part of spacetime in
which the forward lightcones of Alice and Bob's measurments overlap.
Yes, but this leads to only apparent non-local effects that are all due
to trivial "common cause effects".
If you do that you're just going to re-derive the same old result, but
using a much more cumbersome formalism. But that cumbersome formalism
then does falsify your claim that the MWI is non-local.
The crucial point where your analysis is faulty is when you invoke the
angle in an ad hoc way. The angle arises from the setting of the
polarizers, we can e.g. assume that the polarizers were set a priori
to some settings and that information was known globally. But then
there is no issue with non-locality. You can also assume that Alice
and Bob decide to choose the polarizer settings later, but then the
evolution of Alice and Bob leading up to their choices must be
included in the dynamics. If we are to assume that Alice cannot even
in principle know what Bob's setting is, then that means that the
physically correct state will be a superposition of many different
polarizer settings for both Alice and Bob.
That doesn't follow. Suppose the polarizers are set according to the
detection or not of a photon from distant stars who are opposite on
another on the celestial sphere. Alice can see how her polarizer is
set and Bob can see his, so there's no need to postulate a
superposition; but their seeing of the settings are spacelike.
That doesn't matter, my point is that you need to account for the entire
system. If you invoke photons emitted by some distant star, you need to
include these details in your description and then see if you still get
some strange non-local effect, where "strange" means that it isn't the
sort of trivial common cause effect.
While you can project out the subspace where Alice chooses some angle
and finds some particular result and then claim that if Bob had chose
that same angle two of the four outcomes would mysteriously have
vanished, there isn't anything on Bob's side that makes him make that
same choice. Invoking that he'll do so amounts to just planting the
information that exists on Alice side to Bob's side, that's then not a
non-local effect at all.
But when the results are compared Alice and Bob will be able to sort
out which results went with which polarizer settings. That's how the
correlation is seen. No one claims that this can used to communicate
FTL. Only that the interactions are spacelike and violate Bell's
inequality.
Yes, but it's only in single World theories that there is a problem
here. In the case bith have chosen the same settings, Alice knows that
there were 2 possible outcomes for her and Bob also has two possible
outcomes. They know that there are no local hidden variables, therefore
what each of them found was not predetermined, at least not locally.
However, since Alice knew what Bob must have found, this means that the
information about Bob's result also appeared at Alice's site. The way
this information appeared there is not described by the (local) dynamics
(because collapse is assumed to be random) and there is a non-trivial
non-local aspect to this.
But this effect is completely absent in the MWI, because there is no
random selection of certain sectors of Hilbert space, there is only the
local dynamics at play which at most can yield apparent non-local
effects which are trivial common cause effects.
The branching structure of the MWI is only an effective description,
it's not some fundamental postulate, this is where Bruce goes wrong.
It's quite similar to the debates in physics about irrevesibility in
thermodynamics, paradoxes involving Boltzmann's H-theorem etc. etc. What
goes wrong in such debates is that people tend to take certain effective
description as the truth, basically replacing some "FAPP truth" by "the
truth" and then arguing on the basis of that and then arriving at some
paradox.
Saibal
--
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 https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
