Non-local preparation in entanglement swapping experiments

[DrChinese]

Do you actually have any other concrete criticisms to the other thread...

Out of respect for you, but without in any way touching on it further: That thread was closed for a reason. I fully support ending further discussion of that paper.

I have literally a hundred bookmarks of papers denying Bell, asserting Local Realism under a variety of names or guises, and/or asserting physics already disproven by experiment. It is not helpful to pull such references out and discuss each and every one individually as if it is generally accepted science. It's not fair to readers and participants.

 

[iste]

Out of respect for you, but without in any way touching on it further: That thread was closed for a reason. I fully support ending further discussion of that paper.

I have literally a hundred bookmarks of papers denying Bell, asserting Local Realism under a variety of names or guises, and/or asserting physics already disproven by experiment. It is not helpful to pull such references out and discuss each and every one individually as if it is generally accepted science. It's not fair to readers and participants.

But the paper doesn't deny Bell. The paper is about deriving Bell correlations that violate Bell's inequalities. The paper explicitly asserts that what they are saying doesn't imply realism. They are giving a mechanism which is Bell non-local, violates Bell inequalities, has no local hidden variables - all they are saying is that you can violate Bell inequalities with a mechanism that doesn't imply FTL signalling. This is not a denial of Bell or standard quantum mechanics or local realism.

This is yet another example of you just failing to understand what you are reading.

I think it is therefore very relevant to know what other objections you have considering that the paper doesn't say what you say it does.

 

[pines-demon]

We can say "there is entanglement without interaction in any sense", or deny it, but all the resulting interpretations have to coincide in explaining the same set of experimental data that make up QM, right?

Yes the predictions are the same. Generally, in entanglement swapping particles 1 and 4 do not interact locally (being in the same place), however different interpretations consider that particles can interact nonlocally (or not).

 

[pines-demon]

If you can’t flip the order in some reference frame, they cannot be made to look simultaneous either.

Sure, I am not denying that, I just found simultaneity more important for describing the particles as coexisting in response to another user.

It’s a question of distance apart in spacetime. I gave you a specific example. Why don’t you address that one, and show some reference frame where the A and B events are simultaneous?

Again I was not necessarily addressing your comment, but it is a general statement that works for your case if events are spacelike.

Edit: ah you were talking about the $x<1$m-3ns experiment? sure that is no longer spacelike.
Edit2: this remark about spacelike events is important because if the events cannot be made spacelike, at least in principle, means that there could be some loophole where both particles interact in some way.
Edit3: If events are spacelike and action-at-a-distance exists, then it does not matter when the two events happened. If the events are not spacelike, you can posit that some ordinary force imposes the correlations.

 

[javisot20]

I assumed that two particles that act as if entangled without any interaction or influence in any sense don't need to share any type of information to acting like this (is this a reasonable assumption?)

These particles show quantum correlations when all the coincidences and causalities happen for this to happen, without having interacted directly in any sense. Finding two finance agents who, without having ever interacted, show quantum correlations in their way of acting seems complicated, but quantum particles are simpler and more fundamental than finance agents.

 

[pines-demon]

I assumed that two particles that act as if entangled without any interaction or influence in any sense don't need to share any type of information to acting like this (is this a reasonable assumption?)

These particles show quantum correlations when all the coincidences and causalities happen for this to happen, without having interacted directly in any sense. Finding two finance agents who, without having ever interacted, show quantum correlations in their way of acting seems complicated, but quantum particles are simpler and more fundamental than finance agents.

Two particles that never ever have interacted in any way can behave as if entangled, that is no mystery. The mystery is that we can produce many of them and show that the pairs in the ensemble behave as if entangled (even if the pair of particles have never interacted locally). This requires measurements in different angles, and in principle there is no mechanism for the particles to know which angles are going to be measured. We perform as many experiment as to show that this is not just some statistical coincidence.

 

[Morbert]

Sorry, Bell's Theorem is fully accepted in the physics community. There is no amount of hand waving that will justify his position - as of 2015 he rejects both Bell and loophole free Bell tests. By his own words.

He doesn't reject Bell's theorem at all. He (and many others) reject the claim that Bell's theorem necessarily implies Einsteinian nonlocal influence.

And as for the need to consider modern theory and experiment: "Following this somewhat lengthy introduction to the circuit in Fig. 1 let us now use it to search for genuine quantum properties which might be counterparts of, or at least resemble in some way, the mysterious λ that plays a central role in discussions of Bell’s inequality in the literature." In the various entanglement swapping experiments since 2008 I have repeatedly cited (and you have commented on), the final entangled pair are created without having existed in a common light cone. Consequently, there can be no hidden variables λ in the first place. (And why he would ask us to assume there are hidden variables is something of a mystery to me anyway - it's a form of circular reasoning.) So it seems he does condition his reasoning on what you called "coincident pasts of subsystems" which is in fact λ.

Consistent histories rejects hidden variables. The point is without λ, you don't have to infer nonlocal influences.

 

[gentzen]

He doesn't reject Bell's theorem at all. He (and many others) reject the claim that Bell's theorem necessarily implies Einsteinian nonlocal influence.

Consistent histories rejects hidden variables. The point is without λ, you don't have to infer nonlocal influences.

Yes, it doesn‘t reject Bell‘s theorem. However, it cannot go beyond statistical interpretations. So exactly the opposite of what DrChinese wants.

 

[Morbert]

Yes, it doesn‘t reject Bell‘s theorem. However, it cannot go beyond statistical interpretations. So exactly the opposite of what DrChinese wants.

Could you expand on this? For example
https://journals.aps.org/ppf/pdf/10.1103/PhysicsPhysiqueFizika.1.195

In a theory in which parameters are added to quantum mechanics to determine the results of individual measurements, without changing the statistical predictions, there must be a mechanism whereby the setting of one measuring device can influence the reading of another instrument, however remote. Moreover, the signal involved must propagate instantaneously, so that such a theory could not be Lorentz invariant.

Taking this to be Bell's theorem, Consistent Histories does not challenge it. According to Griffiths, Consistent Histories offers a realistic interpretation of QM without the need for parameters that "are added to quantum mechanics to determine the results of individual measurements".
[edit] Sorry I misread "Yes, it doesn‘t reject Bell‘s theorem" as "Yes, it does reject Bell‘s theorem"

 

[Morbert]

Yes, it doesn‘t reject Bell‘s theorem. However, it cannot go beyond statistical interpretations. So exactly the opposite of what DrChinese wants.

Consistent histories doesn't limit us to statistical interpretations if we adopt a Bayesian interpretation of probabilities.

 

[gentzen]

Consistent histories doesn't limit us to statistical interpretations if we adopt a Bayesian interpretation of probabilities.

Maybe ask yourself first what a realistic interpretation should provide. If that is missing, then Bayesian probabilities cannot fix that either.

Also note that not being able to provide this is not necessarily a drawback. It would even be incompatible with „Copenhagen done right“, if CH could provide this.

 

[DrChinese]

He doesn't reject Bell's theorem at all. He (and many others) reject the claim that Bell's theorem necessarily implies Einsteinian nonlocal influence.

Consistent histories rejects hidden variables. The point is without λ, you don't have to infer nonlocal influences.

My head is spinning.

I didn’t see the non-hidden variable mechanism that would then need to exist in CH. (We see that in MWI. We see that in retrocausal type explanations.) On the other hand, in your post #69: you say CH is realistic, but denies hidden variables. I am not sure how it can be realistic, which implies a pre-existing and determinate outcome for measurements at all angles independent of a setting elsewhere.

And he does accept a form of “proper” nonlocality. But I am very open to better understanding what is being presented, because it doesn’t seem to fit together as I read it.

 

[Morbert]

Maybe ask yourself first what a realistic interpretation should provide. If that is missing, then Bayesian probabilities cannot fix that either.

By realistic, Griffiths means measurements reveal pre-existing properties of the microscopic system. This would be in contrast with Roland Omnes's position, where macroscopic data is still ultimately what is real.

 

[gentzen]

By realistic, Griffiths means measurements reveal pre-existing properties of the microscopic system.

Griffiths only talks about statistical properties. This is not what most people (including me) mean by realistic. In fact, I claim that CH itself can only talk about statistical properties. This is a nontrivial claim, and it could be wrong. But not in the way Griffiths argues against it, by simply ignoring the issue.

(It could be wrong in Aharonov‘s way, where you actually get rid of the statistics.)

 

[DrChinese]

By realistic, Griffiths means measurements reveal pre-existing properties of the microscopic system. This would be in contrast with Roland Omnes's position, where CH logic is still ultimately only about macroscopic data.

OK, that is pretty much as good a definition as I could compose. The only think I might add is the caveat "the setting of one measuring device cannot influence the reading of another instrument".

So if it is realistic, according to Bell, there must be a nonlocal (instantaneous) mechanism of influence (per your quote of Bell). This is standard deduction from Bell, well-accepted and discussed in thousands of papers. If you deny this, you deny Bell. You can't say "I follow Bell" but then say Bell doesn't apply. If it doesn't apply, you simply don't follow Bell. You can't have your cake and eat it too.

In fact: We know there are no Alice/Bob datasets featuring pre-existing and immutable values in which a) there are perfect correlations at same angles; and b) Bell inequality violations at CHSH angles. I can't understand the point of asserting otherwise. And certainly not after decades of heavy study.

 

[DrChinese]

Griffiths only talks about statistical properties. This is not what most people (including me) mean by realistic.

Help me out: How does use of the word "statistical" change anything? (I am not arguing against you, just trying to understand how that can matter to Griffiths or anybody.)

We must also consider perfect correlations in Bell tests, so the "statistic" in these is 100%... i.e. an element of reality per EPR. There are no hypothetical datasets that reproduce this behavior. Any "realistic" explanation must be able to handle the cases of perfect correlation, as well as CHSH type correlation.

 

[gentzen]

We must also consider perfect correlations in Bell tests, so the "statistic" in these is 100%... i.e. an element of reality per EPR. There are no hypothetical datasets that reproduce this behavior. Any "realistic" explanation must be able to handle the cases of perfect correlation, as well as CHSH type correlation.

CH can handle this 100% case. It is the other cases where it only talks about statistics. And if it could reduce those other cases to 100% cases, then my claim would be wrong. But Griffiths does not even try, or acknowledge the issue.

 

[DrChinese]

CH can handle this 100% case. It is the other cases where it only talks about statistics. And if it could reduce those other cases to 100% cases, then my claim would be wrong. But Griffiths does not even try, or acknowledge the issue.

My point lies in the 100% cases (perfect correlations) where there is no common light cone. That's what we can't get a straight answer about. I've seen how Griffiths "hand waves" away Bell Inequalities, but there aren't any with perfect correlations between remote systems.

 

[gentzen]

My point lies in the 100% cases (perfect correlations) where there is no common light cone. That's what we can't get a straight answer about.

Why not?

I've seen how Griffiths "hand waves" away Bell Inequalities, but there aren't any with perfect correlations between remote systems.

I haven‘t seen him „hand wave“ them away. Griffiths has occasionally weak spots, but nothing serious, especially compared to Goldstein who simply is wrong about CH.

CH is a consistent interpretation, it does not need the support by mumblings of famous physicists like Murray Gell Mann. And the same applies to Griffiths: he has presented CH in easily accessible form multiple times, and that is great. But CH is not dependent on an exegesis of his words, comparable to what is needed to understand Bohr‘s opinion on QM.

 

[DrChinese]

Why not?

I can't answer that, I'm the one asking.

I have provided the most specific examples possible for it to be explained. And I've provided papers by Nobel laureates detailing how there is no common interaction via any form of Einsteinian causality. All I ever see is stuff like "Bell's Inequality doesn't apply" or "we don't know anything until the results are brought together using classical signals". Both of these explanations are meaningless of course.

 

[jbergman]

I can't answer that, I'm the one asking.

I have provided the most specific examples possible for it to be explained. And I've provided papers by Nobel laureates detailing how there is no common interaction via any form of Einsteinian causality. All I ever see is stuff like "Bell's Inequality doesn't apply" or "we don't know anything until the results are brought together using classical signals". Both of these explanations are meaningless of course.

There is a section on the GHZ experiment in Griffith's article, https://plato.stanford.edu/entries/qm-consistent-histories/#GreeHornZeil .

 

[Morbert]

There is a section on the GHZ experiment in Griffith's article, https://plato.stanford.edu/entries/qm-consistent-histories/#GreeHornZeil .

I also discussed it in an earlier thread here

 

[Fra]

(Trying here once more to put the finger on something in a compact way)

In fact: We know there are no Alice/Bob datasets featuring pre-existing and immutable values in which a) there are perfect correlations at same angles; and b) Bell inequality violations at CHSH angles. I can't understand the point of asserting otherwise. And certainly not after decades of heavy study.

IMO, one possible presumed "point" of ("bells theorem does not apply", and we need the classically communicated bsm data) that is implicit in how I've trided to argue on this also in previous threads is this:

• We know after decades that we (anyhthing in the macroscopic environment) have no way of knowing/reading this potentially pre-existing immutable variable that would solve (a), without destroying the "entanglement" and thus the interference pattern. It does not follow logically, that this implies it is not known to the two entangled subsystems, does it? If so, how does it follow? Without physical inquiry? It just means there is no relation to the environment. This is a fallacy IMO.
  
  
• Bells ansatz (and this is at the heart of the divisibility issue that was discussed in the other thread) ASSUMES that a the variable that potentially only the two entangled "subsystems" might know, can still be used to to imagine a hypothetical sample space (that we KNOW can't be sampled! as it destroys entalgnment!!) and divide the process into a partition, where we further assumes that the total results is an average over a process that is assumed to be as if, the hidden variable was known (but just escaped the theorist data). This ansatz IMO follow from no sane logic. It is a simple possibility yes, and the one going into bells ansatz. But the point I tried to make, and I would say also behind the objection of Barandes, is that this is also a fallacy, and the one that creates the confusion. One might argue that this "ansatz" is part of some "realism assumption" that bell had, so that he felt that if this divisibility assumption is not valid, then it does not meet his notion of "realistic". That may well be, but I think it is a limited view on reality in that case. Give or take the definition of "realism" - we still seek an explanation, that we do not have.

/Fredrik

 

[gentzen]

I can't answer that, I'm the one asking.

I feel like having a deja-vu:

I don’t think I know it better than you, I am the one asking.

What exactly is your question? All I can see is an assertion which raises questions regarding your familiarity with CH.

I have provided the most specific examples possible for it to be explained.

It is unclear to me what you expect from CH. What would it mean for you to "explain the most specific examples"? CH is most of all a formalism, which can often be applied to model a given physical situation or to quantitatively describe some actual physical experiment. This is most useful for situations that initially seem paradoxical. The advantage of CH over Copenhagen is that it is more rigorous, but its intuitive explanatory value is often not significantly different.

 

[DrChinese]

There is a section on the GHZ experiment in Griffith's article, https://plato.stanford.edu/entries/qm-consistent-histories/#GreeHornZeil .

Not much to that, it's just the same hand wave as he does to dismiss Bell. That is: he is denying the very realism he is asserting. He just uses his own modified definitions to make them different than usage by everyone else. If you say that a quantum object has objective properties at all times, then Bell and GHZ give that statement a mathematically precise definition. This is well accepted in the scientific community.

Of course, anyone can deny this formulation. But it doesn't really accomplish anything, because you are out on an island with a handful of fellow occupants.

 

[DrChinese]

It is unclear to me what you expect from CH. What would it mean for you to "explain the most specific examples"? CH is most of all a formalism, which can often be applied to model a given physical situation or to quantitatively describe some actual physical experiment.

I gave the experiment (or this, and this) many times. How does CH/Griffiths explain perfect correlation of photons with no common past without invoking nonlocality? How much more specific can I be?

Walk through a specific case a la the Ma paper. Photon 1 is measured |L>, photons 2 & 3 distantly measured indistinguishably |VV>. a) How does distant photon 4 end up as |L> (which is a mutually unbiased basis to V/H)? b) And why would that NOT work if the 2 & 3 photons are distinguishable?