How can one event affect another instantly over a distance

[Sherlock]

So, I'll repeat my question that you didn't answer. :-)
What happens to these local hidden variables
when we incorporate these individual measurement
events into a correlational context involving other individual
measurement events at spacelike separations from these?
Do the hidden variables just vanish (along with local
reality)? Or is it simply that they aren't determining the
joint results?

If the lhv's simply aren't a factor in determining the joint
results, then isn't it incorrect to say that these setups
show that lhv's don't exist, or that there is no locally
realistic behavior occurring in these setups, or that lhv
descriptions of any setup are therefore ruled out?

And I'll repeat my answer: There are no local hidden variables in QM.

This isn't what I asked.

All LHV theories are incompatible with all of the predictions of QM.

No, some qm formulations can be supplemented with
lhv info, and some can't. What's the difference between
those that can and those that can't?

If you want to postulate a LHV which mimics SOME of the predictions of QM, no one is disputing your ability to do that. But since such a theory makes erroneous predictions about some experiments (such as Aspect), it is not likely to find much acceptance among scientists.

It's a matter of supplementing qm formulations with lhv
info. In some cases this would improve qm predictions
(eg. individual results), and in some cases (composite
setups) including lhv's as determining parameters
reduces the accuracy of predictions. Why?

Entangled systems are merely a tool that enables us to realize Bell's Theorem (i.e. that LR is incompatible with QM). It is not a boundary condition, i.e. that the world is local realistic everywhere EXCEPT
entangled systems.

We see that lhv's apply (determine outcomes) in some setups
and not in others. Am I to suppose that there are no
lhv's existing in the composite setups simply because they
don't determine the outcomes, or is there more to it than
that?

 

[DrChinese]

...We see that lhv's apply (determine outcomes) in some setups and not in others.

There are no known such situations within the realm of QM - there couldn't be, because such would violate the HUP.

You are free to contradict that with an actual example. An example would be something which displays the actual local hidden variables for us to see, not something which is a hypothetical abstraction.

 

[Sherlock]

There are no known such situations within the realm of QM - there couldn't be, because such would violate the HUP.

You are free to contradict that with an actual example. An example would be something which displays the actual local hidden variables for us to see, not something which is a hypothetical abstraction.

If we could see them they wouldn't be hidden variables,
would they? :-)

So, the whole discussion is about hypothetical abstractions ...
ie., what would happen if we supplemented some formulation
or other with hidden variable information?

And, we see that wrt some formulations it would help, and
wrt other formulations it wouldn't.

The HUP has nothing to do with more accurately predicting
detection patterns given some inferred additional information
about submicroscopic processes that's otherwise hidden from us.

This is what's happening when random individual
detections are combined to produce predictable joint
results.

 

[DrChinese]

If we could see them they wouldn't be hidden variables,
would they? :-)

So, the whole discussion is about hypothetical abstractions ...
ie., what would happen if we supplemented some formulation
or other with hidden variable information?

And, we see that wrt some formulations it would help, and
wrt other formulations it wouldn't.

The HUP has nothing to do with more accurately predicting
detection patterns given some inferred additional information
about submicroscopic processes that's otherwise hidden from us.

This is what's happening when random individual
detections are combined to produce predictable joint
results.

Well, I can't allot weight much to a theory that explains nothing new, predicts nothing new, is not falsifiable (even when experiments such as Aspect DO falsify it), applies only in occasional spots and appears to do nothing other than satisfy perceived dissatisfactions with QM. This is why Bell's Theorem is so useful. I don't even need to consider the idea of this theory further because the entire class of LHV theories are ruled out.

If I was really smart, I'd get you and ttn talking to each other... You advocating Local HV theories as being "proven", and ttn advocating Non-local HV theories as being "proven". Then I would just side-step outta here.

 

[Sherlock]

Well, I can't allot weight much to a theory that explains nothing new, predicts nothing new, is not falsifiable (even when experiments such as Aspect DO falsify it), applies only in occasional spots and appears to do nothing other than satisfy perceived dissatisfactions with QM. This is why Bell's Theorem is so useful. I don't even need to consider the idea of this theory further because the entire class of LHV theories are ruled out.

If I was really smart, I'd get you and ttn talking to each other... You advocating Local HV theories as being "proven", and ttn advocating Non-local HV theories as being "proven". Then I would just side-step outta here.

I'm not advocating lhv theories as being proven. I just think that
some important points are being overlooked. Lhv's (not lhv theories,
just lhv's) can exist and still not be relevant in some setups. So,
where they're not relevant you just don't use them. That's all.
This doesn't rule out lhv theories in general. It doesn't mean
that lhv's don't exist. There's still some real stuff happening
between emitters and detectors and we use what we can
infer about it to develop better, more complete, descriptions
of physical reality.

 

[NateTG]

Now I'll ask you the question that I asked DrChinese.
What happens to the lhv's in the composite systems?
Do we conclude that they don't exist in either individual
or composite systems? Or that they exist in one but not
the other?

The problem with 'nice' LHV theories for explaining composite systems is that they don't provide a mechanism for the HUP. Since I'm not particularly interested in the QM nuts and bolts I can't be certain of this, but Bells theorem looks like it rules out any 'hidden local realistic' theory that assigns a value to the chance of correlating non-commuting measurements, for example, measuring spin direction along a couple of different axes.

Before discussing them, I will warn you that this type of model is not AFAIK well received in mainstream physics. However, there are hidden variable theories that do not assign values to the correlations of non-commuting measurements, and hence are not invalidated by Bell's theorem + Aspect et al., but that involves unmeasurable sets. Moreover, it's clear that models of this type that make identical predictions to the 'wave equation' model can be constructed.

 

[NateTG]

You are free to contradict that with an actual example. An example would be something which displays the actual local hidden variables for us to see, not something which is a hypothetical abstraction.

Ah, but Science works by falsification, not by demonstration, and I can suggest an experiment that could falsify the notion that spin state can be completely explained by LHV theories:

This is only a thought experiment, but consider the following:
Let's say we have an entangled positron source, and an entangled electron source, separated by two light seconds, covered so that only pairs that send one member towards the other source are emitted. So the set up might look something like

   ______         ______
     E+             E-
   ______         ______

So, we have the source on the left emitting positrons, and the source on the right emitting electrons, and sending them into the middle where they anihillate pairwise.

We can time the anihillation, and measure the spin orientations (along the up-down axis only) of the particles that are sent out the outside ends of the apparatus.

If the anihillation occurs readily for all of the particle-antiparticle pairings, and the spins correlate then we have correlating measurements that cannot be drawn back to a single (non-hidden) event since they are (or at least could be) separated from the creation of the other particle by more than ct. This cannot be explained by any local hidden variable theory.

 

[ZapperZ]

I suppose at this point I should add that there is an obscenly comprehensive review of hidden variable theories written by Marco Genovese[1]. It is a 78-page review of the theories and experiments on the EPR-type issues, and contains 504 references! It took me a week to actually finish reading the damn thing, and I need to go over it again. It covers the historical development of the field along with practically all the important theoretical and experimental results in this field, except for the 3 that I have mentioned that was recently published (see PF Blog).

There are some strange sentences in the article (a more accurate proof-reading might have made it better), but I still highly recommend this for anyone wishing to understand this area of physics. I think other than Special Relativity, this is one area of physics that has a lot of misunderstanding.

Zz.

[1] M. Genovese, Phys. Rep. v.413, p.319 (2005).

 

[Locrian]

It may be true that this argument is repeated once a month, as NateTG noted, but I learn something new each time.

 

[Sherlock]

Ah, but Science works by falsification, not by demonstration, and I can suggest an experiment that could falsify the notion that spin state can be completely explained by LHV theories:

This is only a thought experiment, but consider the following:
Let's say we have an entangled positron source, and an entangled electron source, separated by two light seconds, covered so that only pairs that send one member towards the other source are emitted. So the set up might look something like

   ______         ______
     E+             E-
   ______         ______

So, we have the source on the left emitting positrons, and the source on the right emitting electrons, and sending them into the middle where they anihillate pairwise.

We can time the anihillation, and measure the spin orientations (along the up-down axis only) of the particles that are sent out the outside ends of the apparatus.

If the anihillation occurs readily for all of the particle-antiparticle pairings, and the spins correlate then we have correlating measurements that cannot be drawn back to a single (non-hidden) event since they are (or at least could be) separated from the creation of the other particle by more than ct. This cannot be explained by any local hidden variable theory.

The simpler, optical Bell tests can't be explained by lhv
theory either -- unless you don't assign a value to the
hidden parameter. Then it's not an lhv theory, but a
local hidden whatever (some mysterious constant
relationship between members of each pair?) theory. :-)

Anyway, I'm curious. Regarding your thought experiment,
how would you correlate anihilations to coincidences? Or,
would you just compare the counts per unit of time, or what?

 

[Sherlock]

The problem with 'nice' LHV theories for explaining composite systems is that they don't provide a mechanism for the HUP. Since I'm not particularly interested in the QM nuts and bolts I can't be certain of this, but Bells theorem looks like it rules out any 'hidden local realistic' theory that assigns a value to the chance of correlating non-commuting measurements, for example, measuring spin direction along a couple of different axes.

Before discussing them, I will warn you that this type of model is not AFAIK well received in mainstream physics. However, there are hidden variable theories that do not assign values to the correlations of non-commuting measurements, and hence are not invalidated by Bell's theorem + Aspect et al., but that involves unmeasurable sets. Moreover, it's clear that models of this type that make identical predictions to the 'wave equation' model can be constructed.

This sounds like the way I've been thinking about it. Do you
happen to have any references handy?

 

[NateTG]

The simpler, optical Bell tests can't be explained by lhv
theory either -- unless you don't assign a value to the
hidden parameter. Then it's not an lhv theory, but a
local hidden whatever (some mysterious constant
relationship between members of each pair?) theory. :-)

Actually, Bell's theorem only eliminates 'nice' lhv theories. To avoid confusion, I will refer to the local hidden theories that it does not eliminate as local hiden monster theories (lhm). The experiment I described may be able to falsify these lhm theories because it has a larger separation between the measurements than a traditional EPR experiment.

Anyway, I'm curious. Regarding your thought experiment,
how would you correlate anihilations to coincidences? Or,
would you just compare the counts per unit of time, or what?

I was thinking that you control the emitters so that the events are sparse. Then count per unit time should work. I expect that a macroscopic count per unit time would be theoretically nice. However, the net spin of a bunch of particles is going to be about the square root of the number of particles or less, which means that if you deal with large numbers of particles and anihillations there should be more 'noise'.

From my perspective the experiment has the bigger problem that I have absolutely no idea what sort of prediction conventional models make for the experimental results - so it may be a non-experiment in that the predictions made by the theories do not differ.

 

[Sherlock]

From my perspective the experiment has the bigger problem that I have absolutely no idea what sort of prediction conventional models make for the experimental results - so it may be a non-experiment in that the predictions made by the theories do not differ.

A more detailed diagram of what you have in mind
is necessary (at least for me).

My intuitive prediction is that local hidden monster
(I prefer local hidden constant) formulations won't be
falsified. :-)

If I get time, I'll do some homework about how your
proposed setup might work.

 

[ohwilleke]

Let me look at QM v. Special Relativity from a slightly different perspective.

The basic deal in entanglement is that when you locally learn facts about half of a set of entangled particles, you know something about the other half of the set of engtangled particles, regardless of their distance.

In advance, you can't know what data will be found when you collapse the wave function on either set of particles. And, viewed alone, each set of particles will comply with QM predictions.

To use the Copenhagen interpretation (or as it has been called in this thread) OQM, the reason that you can't know in advance what data will be found when you collapse the wave function on either set of particles is that this data doesn't exist yet. Until there collapse happens, any result is possible. A set of particles does not have a deterministic pre-ordained state upon collapse.

OK, so enough of OQM for a moment. We'll come back to that.

Now, one of the most basic elements of SR is that light in a vacuum travels at speed c. It has been assumed that one can infer from SR the stricter condition that information also does not travel in excess of speed c.

It seems to me that one way to reconcile SR and OCM is to violate neither of these propositions, but instead to violate the implicit, but mathematically unnecessary assuption of SR that information and light always go forward in time.

The way that you would do this is to stick to OCM. The state of one half of a set of entangled particles does not exist until you collapse the wave function. So, how does the other part of the set of entangled particles end up corollated?

Maybe, at the moment that one collapse happens, that information goes backwards in time (at a speed not greater than c) to the point of entanglement, and then goes forward in time from there to the other entangled particles, communicating the information to the second set without the message ever having traveled faster than c.

This looks like FTL, but it isn't. One of the keystones of entanglement is that it only happens to particles that have been physically local at some point in time, making this backward to forward in time communication possible.

Unlike a traditional hidden variable theory, nothing that has yet happened makes it possible to determine the final set of either part of the set of entangled particles.

But, only events within the light cone of the initial entanglement (the light cone of the global system if you will) can influence either result.

Does that make sense? Where have I gone wrong?

 

[NateTG]

This sounds like the way I've been thinking about it. Do you
happen to have any references handy?

I'll warn you again, that this is non-standard stuff. I'm also going to warn you that it involves non-measurable sets (if you don't know what this means, you might want to look into it a bit google unmeasurable sets, and banach-tarski for a taste).

That said, you might check out this thread:
https://www.physicsforums.com/showthread.php?t=30947&page=1&pp=15

 

[DrChinese]

It seems to me that one way to reconcile SR and OCM is to violate neither of these propositions, but instead to violate the implicit, but mathematically unnecessary assuption of SR that information and light always go forward in time.

Maybe, at the moment that one collapse happens, that information goes backwards in time (at a speed not greater than c) to the point of entanglement, and then goes forward in time from there to the other entangled particles, communicating the information to the second set without the message ever having traveled faster than c.

This looks like FTL, but it isn't. One of the keystones of entanglement is that it only happens to particles that have been physically local at some point in time, making this backward to forward in time communication possible.

This explanation makes AT LEAST as much sense as the other interpretations. Probably a lot more, since all physical laws are otherwise time symmetric. If the future were to exhibit SOME influence on the past, then it would appear to those in the past as being random (uncaused).

There are some articles out there which hint at this. See http://www.arxiv.org/abs/quant-ph/0507269 for example.