On 11/11/2017 10:57 PM, Bruce Kellett wrote:
On 12/11/2017 5:39 pm, Brent Meeker wrote:
On 11/11/2017 9:56 PM, Bruce Kellett wrote:
On 12/11/2017 4:04 pm, Brent Meeker wrote:
On 11/11/2017 6:47 PM, Bruce Kellett wrote:
On 12/11/2017 4:34 am, John Clark wrote:
On Fri, Nov 10, 2017 at 7:08 PM, Alan Grayson
<agrayson2...@gmail.com>wrote:
>>
That's not the measurement problem, its determining if
how and why observation effects things.
>
Not to split hairs, but why we get what we get in quantum
measurements, and how measurement outcomes come to be what
they are, are the same problem IMO.
The measurement problem is not the ability or inability to
predict exact outcomes,
the measurement problem is defining what is
and
what
is not a measurement and
finding the
minimum properties a system
must
have to be an observer. Nondeterminism is not a problem and there
is no inconsistency at all regardless of what turns out to be true
;
if some effects have no cause and true randomness exists then
that's just the way things are are
and
t
here is no resulting paradox and no question that needs answering.
The title of this thread is about the consistency of Quantum
Mechanics, but far more important than QM is the ability of ANY
theory to be compatible with experimental results, and one of
those experiments shows the violation of Bell's Inequality. And
that violation tells us that for ANY theory to be successful at
explaining how the world works AT LEAST one of the following
properties of that theory must be untrue:
1) Determinism
2) Locality
3) Realism
Is Many Worlds deterministic? Yes in the sense that it just
follows the wave function and that is deterministic, it's only
the collapse of the wave function that is nondeterministic and
that never happens in Manny Worlds.
Is Many Worlds Local? Some say yes but I would say no
because those other worlds are about as non-local as you can get,
you can't get there even with infinite time on your side. But
even if I'm wrong about locality Many Worlds would still be in
the running for a successful theory because it is certainly not
realistic.
I would agree with you that the many worlds account is non-local.
The problem that MW faces is that the separate worlds split off
when measurements are made at either end of the EPR experiment
must somehow be made to match up appropriately when the two
experimenters communicate. This requires coordination of separate
worlds, which, as you say, is about as non-local as you can get.
The problem becomes particularly apparent if you consider an EPR
experiment with time-like separation. Let Alice prepare an EPR
pair in her laboratory, then measure the spin of one of the pair
in some defined direction. She then takes the other member of the
EPR pair down the corridor to her partner, Bob, and gets him to
measure the spin projection in the same direction. If the two
particles are independent, then both measurements give 50/50
chances for up/down. After Alice measures her particle, she splits
into Alice_up and Alice _down according to her result. Both copies
then go to Bob's laboratory, which by then has also split
according to Alice's result. So Alice_up meets Bob, but when he
measures his particle, he still has 50/50 chances of either
result. Unfortunately, the only result that is consistent with
spin conservation is that if Alice got 'up', he must get 'down',
and vice verse (remember that the measurements are aligned by design).
Since Alice_up can't meet a Bob_up, there must be a non-local
influence that determines Bob's result according to which Alice he
meets. This is not removed be assuming no collapse and many worlds.
But Bruno's model assumes infinitely many worlds; some in which
Alice sees up and Bob sees down and others in which Alice sees down
and Bob sees up..."influence" doesn't really appear in the model
because it's kind of block multiverse and there's some rule
(conservation of angular momentum) that means up-up and down-down
don't appear. I think this is also true of t'Hooft's
super-deterministic model because in that model there's nothing
special about the event of Alice's measurement that needs to be
communicated. The idea of influence propagating from an event
derives from the idea that Alice had "free-will" and so her choice
had to be communicated from the measurement event.
That does not address the scenario I have outlined. In the time-like
case, Alice_up meets Bob with a spin state that can result in either
up or down; similarly Alice_down meets Bob with a spin state on
which Bob's measurement can result in either up or down with 50% of
each. There are only two worlds involved at that stage.
No, that's the point. There are infinitely many "worlds" involved
from the beginning. There's no splitting. It's all predetermined.
That is simply not true. There are, by construction, only two worlds
in this scenario before Bob makes his measurment, and that splits him
(along with the Alice beside him) into two more.
True? This is a hypothetical model. In Bruno's model there are
infinitely many worlds "by construction".
The question is, how does Bob with Alice_up not get an up result,
contradicting conservation of angular momentum.
Because each world obeys conservation of angular momentum.
The world in which Alice and Bob both get an 'up' result does not obey
conservation of angular momentum. How do you exclude that world?
Exactly the problem with Bruno's model. He says it recovers things like
linearity and superposition - but beyond that it's not clear that it can
recover QM. He calls this "the white rabbit problem" as though it will
be confined to a few peculiarities.
Similarly, how does Bob with Alice_down not get a down result. Since
the measurement axes are explicitly aligned in this case, the
'up-up' and 'down-down' observations are forbidden. Appealing to an
infinity of worlds is not going to help.
They don't really need to be infinite, just very numerous so that
when we repeat some experiment for which the Born rule predicts 1/pi
or other irrational number, we won't get results in our finite number
of tests that are inconsistent with it.
You are not getting it, Brent! There is only one EPR pair made in this
scenario.
No, Bruce, you're not getting. It your stuck in Everett/Dewitt version
of many worlds in which there is branching from one to two to many.
Imagine an ensemble of worlds which "splits" into two ensembles when
Alice makes her measurement.
Alice and Bob each measure their separate particles, and get either up
or down, with 50% each way. Once Alice has measured and takes her 'up'
result to Bob, he has to make a separate measurement. According to AM
conservation he can't get 'up' also (in this world) since Alice has
presented him with an 'up' result. What prevents him getting 'up'?
That is still a 50% chance, after all, according to the wave function.
The point of this scenario is that the /only/ possibility for Bob
after Alice brings him an 'up' result is 'down'. What stops the 'up'
possibility for his measurement?
't Hooft's superdeterministic model simply says that in this case
the particles are originally produced with spins in the previously
agreed measurement direction. In other words, the 'previously
agreed' direction was determined from the time of the big bang.
Maudlin in his Facebook discussion with 't Hooft makes it clear that
he thinks this is not a well-formulated position. It is not a matter
of freedom of the will in choosing setups and orientations, because
these can be chosen according to the digits of pi after the
10,000,000th.
They can only be chosen that way by physically computing and choosing
that number; events determined since the Big Bang.
Or anything else, and the initial conditions at the big bang could
not have covered all possibilities - at least not in any believable way.
But we can't test all possibilities. Alice and Bob can only do the
experiments determined by the past state of the universe, i.e. those
determined by the Big Bang. I don't know what's "unbelievable" about
that - it's what Laplace et al once believed about the world.
Superdeterminism is a red herring here. 't Hooft explicitly rejects
many words, so his arguments do not apply to the case I am presenting,
which is developed in an explicit no-collapse, many worlds context.
I understand that. I just noted it as another no-collapse model.
If many worlders are to explain the time-like case I have outlined,
they are going to have to work quite hard to avoid the notion of
some influence at a distance.
In Bruno's model the "influence at a distance" is determing which
world you're in.
If that means anything at all, it is still non-local because Bruno has
to rule out the worlds in which angular momentum is not conserved; he
has not shown how he can do this. If it is simply that you cannot find
yourself in a world in which AM is not conserved, then that is just an
unabashed appeal to magic, since such worlds have not been shown not
to exist.
Right. They haven't been shown not to exist, or even be rare, in the
plenum of Bruno's Everything Computable. It has been shown empirically
that we never experience one.
Brent
--
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 everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.