Nonlocality - fact or fiction?

[confusedashell]

I just wonder to the doubters after so many experiments proving nonlocality, what type of "Proof" would be "proof" enough for the people out there who still think a local interpretation could ever describe reality?

 

[ahrkron]

A macroscopic one, involving daily-life-size objects... which pretty much precludes any QM effects to show clearly,

The reason locality is so strongly ingrained in our interpretation of the world is that it works so well for our interaction with all of the objects in our usual surroundings.

 

[confusedashell]

Well, according to every person who advocates nonlocality there is no macroscopic nonlocality

 

[akhmeteli]

I just wonder to the doubters after so many experiments proving nonlocality, what type of "Proof" would be "proof" enough for the people out there who still think a local interpretation could ever describe reality?

The last time I checked, and it was not long ago, there were no "experiments proving nonlocality". For some mysterious reason, each and every one of them had one or more "loopholes", such as "fair sampling", "locality", etc. If my information was wrong or something radically new has happened since then, please advise. Meanwhile, independent of whether I "doubt" nonlocality or not, or whether I "still think a local interpretation could ever describe reality", nonlocality has not been proven experimentally.

 

[confusedashell]

I thought Bell in essence proved that if any theory are to fit the emperical evidences of Quantum Theory it has to be nonlocal?
So either you must believe quantum theory is wrong, or accept nonlocality?

 

[akhmeteli]

I thought Bell in essence proved that if any theory are to fit the emperical evidences of Quantum Theory it has to be nonlocal?

Not really. As far as I understand, it is believed that Bell proved that a local theory cannot be compatible with predictions of quantum theory. Predictions, but not existing empirical evidence. In other words, experimental results incompatible with local theories have not been obtained yet. And I suspect there are reasons to doubt such results will ever be obtained. I am not an expert in the Bell inequalities, but my reasoning (it follows nightlight's posts) is as follows. Unitary evolution under quantum theory does not seem to suggest nonlocality. To prove nonlocality, you need the theory of measurements of quantum theory (projection postulate). However, you can include your instruments (and the observer, if you wish) in the system, and this extended system, according to quantum theory, will undergo unitary evolution and thus would not suggest nonlocality. So attempts to prove nonlocality experimentally may well be an uphill battle. Again, I am not an expert, so the above reasoning may be faulty.

So either you must believe quantum theory is wrong, or accept nonlocality?

If what I wrote above is correct, it seems that the only thing you need to reject nonlocality is to reject the projection postulate. However, this postulate is, strictly speaking, at odds with quantum theory anyway (as the latter suggests unitary evolution). The situation reminds me the situation with classical mechanics (though replacing it with quantum mechanics does not change the following) and thermodynamics. Mechanics underlies thermodynamics, but it cannot provide irreversibility, however thermodynamics with its irreversibility still gives reasonably correct experimental predictions.

 

[jtbell]

As far as I understand, it is believed that Bell proved that a local theory cannot be compatible with predictions of quantum theory.

Make that "a local and realistic theory." Here "realistic" means that variables such as position or spin orientation have definite (although possibly "hidden") values at all times, even before a measurement of those variables.

Predictions, but not existing empirical evidence. In other words, experimental results incompatible with local theories have not been obtained yet.

As far as I know, the experimental evidence so far is consistent with the predictions of QM and therefore inconsistent with the predictions of any local realistic theory (per Bell's Theorem). Therefore, a viable theory to replace QM must be either non-realistic or non-local, or both.

 

[strangerep]

One thing always bugs me about the proof(s) of Bell's theorem...

Take (for example) eq(5) on p36 of "Speakable and Unspeakable in QM":

<br /> P(\hat a, \hat b) ~= \int d\lambda ~ \rho(\lambda) A(\hat a, \lambda) B(\hat b, \lambda)<br />

where "\lambda" denotes all the hidden variables.

My problem is that such an integral is only defined for finite-dimensional sets
"\{\lambda\}", or sets of countably-infinite dimension.

This relates to an elementary theorem in general topology which says that finite
product spaces, and product spaces of countably-infinite dimension, are
metrizable, but product spaces of uncountably-infinite dimension need
not be.

Function spaces are generally uncountably-infinite, so maybe this is
why Bohmian mechanics got as far as it did - its "hidden variables" are
such a function space.

So it seems to me that Bell's theorem doesn't go through in such cases,
and therefore says nothing about local theories with an uncountably-infinite
number of hidden variables.

Or am I missing something?

 

[akhmeteli]

Make that "a local and realistic theory." Here "realistic" means that variables such as position or spin orientation have definite (although possibly "hidden") values at all times, even before a measurement of those variables.

Well, both I and OP were too lazy to mention that :-). Another reason, as far as I was concerned, was that THAT Bell :-) believed nonlocality is immanent to quantum theory as well.

As far as I know, the experimental evidence so far is consistent with the predictions of QM and therefore inconsistent with the predictions of any local realistic theory (per Bell's Theorem).

I am afraid this is a non sequitur. Experimental evidence consistent with the predictions of QM is not "inconsistent with the predictions of any local realistic theory (per Bell's Theorem)" as long as it does not break the Bell inequalities. And as far as I know, no existing experimental evidence breaks those inequalities.

 

[Doc Al]

I am afraid this is a non sequitur. Experimental evidence consistent with the predictions of QM is not "inconsistent with the predictions of any local realistic theory (per Bell's Theorem)" as long as it does not break the Bell inequalities. And as far as I know, no existing experimental evidence breaks those inequalities.

The entire point of the Aspect and other EPR-Bell type experiments is that they confirm the predictions of QM in regions where the Bell inequalities are broken.

 

[Demystifier]

Experimental evidence consistent with the predictions of QM is not "inconsistent with the predictions of any local realistic theory (per Bell's Theorem)" as long as it does not break the Bell inequalities. And as far as I know, no existing experimental evidence breaks those inequalities.

You misunderstood something.

 

[peter0302]

Show me information transferred faster than the speed of light and then I'll believe in non-locality. Until then, what you call the "loopholes" look to me like symtoms of the fundamental problem, i.e., that locality is real and that realism is wrong.

 

[akhmeteli]

The entire point of the Aspect and other EPR-Bell type experiments is that they confirm the predictions of QM in regions where the Bell inequalities are broken.

I agree, that was the entire point of those experiments. The problem is, strictly speaking, they failed to make their point. Abner Shimony (and it seems he knows what he is talking about and he is no fan of local realistic theories of QM) wrote the following in his article on the Bell theorem in the Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/bell-theorem/#3 ):

"The incompatibility of Local Realistic Theories with Quantum Mechanics permits adjudication by experiments, some of which are described here. Most of the dozens of experiments performed so far have favored Quantum Mechanics, but not decisively because of the “detection loophole” or the “communication loophole.” The latter has been nearly decisively blocked by a recent experiment and there is a good prospect for blocking the former."

Therefore, as I said, until recently (the article is marked "Copyright 2004"), there have been no experiments demonstrating (without additional, more or less arbitrary assumptions, such as the "fair sampling" assumption) a violation of the Bell inequalities. Again, maybe something radically new has happened since then that I am not aware of. In such case I would appreciate if somebody more knowledgeable than I could enlighten me.

 

[akhmeteli]

You misunderstood something.

Certainly, I value your opinion and would appreciate if you could explain.

 

[ZapperZ]

I agree, that was the entire point of those experiments. The problem is, strictly speaking, they failed to make their point. Abner Shimony (and it seems he knows what he is talking about and he is no fan of local realistic theories of QM) wrote the following in his article on the Bell theorem in the Stanford Encyclopedia of Philosophy (http://plato.stanford.edu/entries/bell-theorem/#3 ):

"The incompatibility of Local Realistic Theories with Quantum Mechanics permits adjudication by experiments, some of which are described here. Most of the dozens of experiments performed so far have favored Quantum Mechanics, but not decisively because of the “detection loophole” or the “communication loophole.” The latter has been nearly decisively blocked by a recent experiment and there is a good prospect for blocking the former."

Therefore, as I said, until recently (the article is marked "Copyright 2004"), there have been no experiments demonstrating (without additional, more or less arbitrary assumptions, such as the "fair sampling" assumption) a violation of the Bell inequalities. Again, maybe something radically new has happened since then that I am not aware of. In such case I would appreciate if somebody more knowledgeable than I could enlighten me.

I posted this just 2 days ago:

https://www.physicsforums.com/showpost.php?p=1698898&postcount=67

Zz.

 

[akhmeteli]

I posted this just 2 days ago:

https://www.physicsforums.com/showpost.php?p=1698898&postcount=67

Zz.

Thank you very much for the reference.

For the benefit of others: as far as I could understand, the experiment in the cited paper (I guess there is a version in arxiv as well) demonstrates the Bell inequalities with the detection loophole closed. The locality loophole is still there big way though: to close it, they need to increase the spatial separation from 1 m to 15 km. Thus, it looks like the general conclusion still stands: there has been no experimental proof of nonlocality so far (44 years since the Bell's article), and I don't hold my breath waiting for such proof. And not just because of experimental difficulties or because local realism is any relative of mine, but because, for reasons outlined in my post #6 in this thread, such proof would mean that unitary evolution of quantum theory is not universal. And this may be too big for me to swallow.

 

[peter0302]

Well, there've been other experiments in which the distances were sufficient to close the locality loophole. I know you want both loopholes closed in one experiment but I think you can syntehsize the two results and deduce what would happen...

 

[ZapperZ]

Thank you very much for the reference.

For the benefit of others: as far as I could understand, the experiment in the cited paper (I guess there is a version in arxiv as well) demonstrates the Bell inequalities with the detection loophole closed. The locality loophole is still there big way though: to close it, they need to increase the spatial separation from 1 m to 15 km. Thus, it looks like the general conclusion still stands: there has been no experimental proof of nonlocality so far (44 years since the Bell's article), and I don't hold my breath waiting for such proof. And not just because of experimental difficulties or because local realism is any relative of mine, but because, for reasons outlined in my post #6 in this thread, such proof would mean that unitary evolution of quantum theory is not universal. And this may be too big for me to swallow.

At some point, you need to step back and look at the body of evidence here.

There's a whole slew of experiments that have closed the locality loophole. All of them involved photons.

There's another slew of experiment that have closed the detection loophole. All of them involved some form of particles, from protons, neutrons, muons, etc.

Now, you are telling me that, somehow, each one of the still "open" loophole conspires to make themselves be THE factor that can still account for the apparent violation of Bell inequality. Don't you just find that to be rather a very unlikely coincidence?

There's another glaring aspect of this. All, and I mean 100%, of the experiments on Bell inequality all make the same claim of the violation. Now, one would think that someone who isn't convinced of this, or who is claiming that such-and-such a loophole is responsible for the apparent violation, would at least be able to conduct his/her own experiment, show the data, and argue conclusively that others who have performed the experiment have erroneously analyzed the data, and that the <insert favorite loophole here> loophole is there, in the data. Now, can you find me such an experimental report? I have found none. Don't you find this rather curious? Why is that?

I will tell you why. In the detection loophole, for instance, even without a 100% efficiency in photon detection, people who do such experiments have to first of all learn about the behavior of their detectors. Everyone who depends on any form of photodetectors have to do this, including high energy experimentalists. We need to know how these instruments behave, what they can do, and more importantly, what they can't do. We need to know when the data we have is reliable, and when we are over-reaching.

So when experiments involving entangled photons are performed, even without a 100% efficiency, we have an excellent idea of the performance of the detectors to say with reasonable confidence of what the actual data are. To me, it is why you have never, ever seen such experiments that contradict the conclusion of violation of local realism so far. It is because once you learn and understand the behavior of such detectors, you'd never pay attention to the weak "detection loophole" argument. The argument against the validity of a data set can only be made by experts not only in the physics, but also in the detection scheme. When Talayerkhan claimed to detect fusion in his bubble fusion experiment, his detractors were not someone who have no clue on the experimental method he was doing. In fact, many of them were world-renowned experts in neutron detections, and they pointed out exactly where the device he's using and the method he adopted can easily produced faulty results. This then threw a lot doubt in the data and subsequently the conclusion (I haven't yet mentioned the fact that others who tried to reproduce the experiment did not get the same result). I have never seen that done with any of the Bell experiments done so far. Considering that there have been plenty of such experiment, and with 100% agreement on the conclusion, I find the lack of contradicting experimental results to be a very obvious shortcoming of those who claim otherwise.

So to me, the continued stubbornness in proclaiming that local realism is still valid because this loophole is still open, or that loophole is still open, has nothing to do with not having convincing experiments. That's like saying Evolution isn't true simply because there are still "gaps" in our knowledge, or that QM isn't right because it still can't be reconciled with GR. It rings hollow because of what they all CAN do already, whereas the alternative have done nothing. All the experiments have produced ONE very convincing argument in favor of violation of non-local realism per the Bell theorem. That is what all these papers have argued and concluded. The paper that I had recently cited simply tried to start hammering down the last nail in the coffin - by being the first to attempt at closing BOTH locality and detection loophole simultaneously. The alternative, being local realism via the non-violation of Bell inequality, have ... er ... zero experimental evidence!

Zz.

 

[akhmeteli]

Well, there've been other experiments in which the distances were sufficient to close the locality loophole. I know you want both loopholes closed in one experiment but I think you can syntehsize the two results and deduce what would happen...

Earlier in this thread you wrote the following:

Show me information transferred faster than the speed of light and then I'll believe in non-locality. Until then, what you call the "loopholes" look to me like symtoms of the fundamental problem, i.e., that locality is real and that realism is wrong.

I am somewhat at a loss trying to understand your point of view: so are "loopholes" symptoms of a fundamental problem or "you can synthesize the two results" and thus eliminate the issue of loopholes?

 

[akhmeteli]

At some point, you need to step back and look at the body of evidence here.

There's a whole slew of experiments that have closed the locality loophole. All of them involved photons.

There's another slew of experiment that have closed the detection loophole. All of them involved some form of particles, from protons, neutrons, muons, etc.

Now, you are telling me that, somehow, each one of the still "open" loophole conspires to make themselves be THE factor that can still account for the apparent violation of Bell inequality. Don't you just find that to be rather a very unlikely coincidence?

I don't know anything about any conspiracy. There is just no experimental evidence of violation of the Bell inequalities (the assumptions of the inequalities require sufficient spatial separation). What am I supposed to do? Consider the entire body of experimental evidence demonstrating no violations of the inequalities and conclude that the inequalities are violated? With all due respect, this is no easy task. The OP asked: how local realists can be so obstinate? It does not matter whether I am a local realist or solipsist or whatever. I just tried to explain that local realists' life is not as hard as OP believed. I hope my explanation was not absolutely useless as it looks like some part of it was new for the OP.

There's another glaring aspect of this. All, and I mean 100%, of the experiments on Bell inequality all make the same claim of the violation. Now, one would think that someone who isn't convinced of this, or who is claiming that such-and-such a loophole is responsible for the apparent violation, would at least be able to conduct his/her own experiment, show the data, and argue conclusively that others who have performed the experiment have erroneously analyzed the data, and that the <insert favorite loophole here> loophole is there, in the data. Now, can you find me such an experimental report? I have found none. Don't you find this rather curious? Why is that?

I readily admit that I don't know much about experiments testing the Bell inequalities. But my understanding is there is a consensus among knowledgeable people (including you), no matter what they think of the interpretation of QM, on what has been demonstrated experimentally and what has not (actually, you are not saying that I misrepresented the experimental situation. If you believe I did, please advise). So I don't see any need to perform any special "local realism inspired" experiments. I hope the bulk of the existing experiments were conducted by knowledgeable people, and I don't care whether they were local realists, positivists, solipsists, whatever. There is no disagreement on what experimental results were obtained. But everybody is free to draw their own conclusions, as long as such conclusions are consistent with the established experimental data.

I will tell you why. In the detection loophole, for instance, even without a 100% efficiency in photon detection, people who do such experiments have to first of all learn about the behavior of their detectors. Everyone who depends on any form of photodetectors have to do this, including high energy experimentalists. We need to know how these instruments behave, what they can do, and more importantly, what they can't do. We need to know when the data we have is reliable, and when we are over-reaching.

So when experiments involving entangled photons are performed, even without a 100% efficiency, we have an excellent idea of the performance of the detectors to say with reasonable confidence of what the actual data are. To me, it is why you have never, ever seen such experiments that contradict the conclusion of violation of local realism so far. It is because once you learn and understand the behavior of such detectors, you'd never pay attention to the weak "detection loophole" argument. The argument against the validity of a data set can only be made by experts not only in the physics, but also in the detection scheme. When Talayerkhan claimed to detect fusion in his bubble fusion experiment, his detractors were not someone who have no clue on the experimental method he was doing. In fact, many of them were world-renowned experts in neutron detections, and they pointed out exactly where the device he's using and the method he adopted can easily produced faulty results. This then threw a lot doubt in the data and subsequently the conclusion (I haven't yet mentioned the fact that others who tried to reproduce the experiment did not get the same result). I have never seen that done with any of the Bell experiments done so far. Considering that there have been plenty of such experiment, and with 100% agreement on the conclusion, I find the lack of contradicting experimental results to be a very obvious shortcoming of those who claim otherwise.

So to me, the continued stubbornness in proclaiming that local realism is still valid because this loophole is still open, or that loophole is still open, has nothing to do with not having convincing experiments. That's like saying Evolution isn't true simply because there are still "gaps" in our knowledge, or that QM isn't right because it still can't be reconciled with GR. It rings hollow because of what they all CAN do already, whereas the alternative have done nothing. All the experiments have produced ONE very convincing argument in favor of violation of non-local realism per the Bell theorem. That is what all these papers have argued and concluded. The paper that I had recently cited simply tried to start hammering down the last nail in the coffin - by being the first to attempt at closing BOTH locality and detection loophole simultaneously. The alternative, being local realism via the non-violation of Bell inequality, have ... er ... zero experimental evidence!

Zz.

Look, you know a lot about photodetectors, I know next to nothing about them. Does this mean that I am not in a position to have my own opinion on the violations of the Bell inequalities? I don't know if people who first raised the issue of the detection loophole knew a lot about photodetectors. What I know is that this issue is generally recognized as such. Otherwise why all these attempts to close the loophole? Why publish the results of such attempts in PRL? Are you telling me I must accept the fair sampling assumption just because I don't know anything about photodetectors? Am I supposed to believe in god just because I have not read the bible or the quran? Then in which god am I supposed to believe - the god of the bible or the god of quran? I just know that the fair sampling assumption is not generally recognized, so I am free to accept or to reject it. I choose to reject it, and I am left with no experimentally observed violations of the Bell inequalities.

OK, forget about local realism for a moment. In my posts I was trying to explain why I don't admit (as long as it is possible without a contradiction with experimental data)that there is any experimental evidence of violations and why I don't expect any such evidence to appear. You might have noticed that my reasoning had little to do with local realism. Furthermore, I actually swore by quantum theory. You see, you cannot prove the Bell inequalities without the projection postulate (please advise if I am wrong). The projection postulate introduces irreversibility, whereas unitary evolution of the quantum theory allows no such thing. You cannot have it both ways - unitary evolution and the projection postulate. You have to choose. I choose the unitary evolution, which is quantum theory, pure and simple. You may say: but we have to accept both mechanics (classical or quantum), where there is no irreversibility, and thermodynamics with its irreversibility. Yes, but we understand that the irreversibility of thermodynamics is just a very good approximation, as the underlying mechanics does not allow any rigorous irreversibility. You need some "manual" interference, however subtle, to obtain irreversibility. And it seems that the analogy between mechanics vs. thermodynamics, on the one hand, and unitary evolution vs. projection postulate, on the other hand, may be deep enough, as arXiv:quant-ph/0702135 (Phys. Rev. A 64, 032108 (2001), Europhys. Lett. 61, 452 (2003), Physica E 29, 261 (2005)) demonstrates, using a rigorously solved model, how the projection postulate evolves as a result of thermodynamic irreversibility.

 

[Doc Al]

You see, you cannot prove the Bell inequalities without the projection postulate (please advise if I am wrong).

Why do you say this? The projection postulate is part of (some interpretations of) quantum mechanics; I don't see where it's used in deriving Bell's inequalities.

 

[Hans de Vries]

I just wonder to the doubters after so many experiments proving nonlocality

One can, in principle, never prove non-locality with entanglement experiments. One
can however disprove specific hidden variable theories.Step by step.

1) Correlations are the result of Entanglement.

2) The entanglement takes place within the light cone of both particles.

3) The theory is local if the information is carried from the entanglement to the
correlation detector in the form of hidden variables.

4) To prove non locality one has to disprove all possible hidden variable theories.
Known theories as well as unknown theories.


Bell proposed a specific hidden variable theory. One that uses the polarization angle as
the only hidden variable and which relies on the assumption that Malus law depends only
on the polarization angle.

The experiments have proven beyond reasonable doubt that Bell's proposed hidden
variable theory is wrong.Regards, Hans

 

[Demystifier]

Otherwise why all these attempts to close the loophole? Why publish the results of such attempts in PRL?

That is a good question. If it is very unlikely that there is a true loophole, i.e., if it is very unlikely that actually QM is wrong and nature is local, then why new experimental confirmations of the things that we "already know" are regularly published in journals such as PRL and Nature? I really don't understand it. Does anybody?

Just for the record, I do think that QM is correct and that nature is nonlocal. Although I admit that there are possibilities for the loopholes, I do not think that these possibilities are serious. However, I do not understand why the experiments that confirm that over and over again are so important.

 

[Hans de Vries]

Just for the record, I do think that QM is correct and that nature is nonlocal.

What about all the people who believe that both QM and SR are correct?

(Of course both statements are somewhat political.. unnecessary I think here,
It's like "Those who do not agree with my personal opinion are dissidents" ) Regards, Hans.

 

[ZapperZ]

I don't know anything about any conspiracy. There is just no experimental evidence of violation of the Bell inequalities (the assumptions of the inequalities require sufficient spatial separation). What am I supposed to do?

You are supposed to write a rebuttal to all those tons of published papers and argue why their claim of violation of Bell inequalities are not valid, rather than argue them here. Pick up any of these papers, and there are clearly words and phrases to the effect that they claim they observed such violations.

I readily admit that I don't know much about experiments testing the Bell inequalities. But my understanding is there is a consensus among knowledgeable people (including you), no matter what they think of the interpretation of QM, on what has been demonstrated experimentally and what has not (actually, you are not saying that I misrepresented the experimental situation. If you believe I did, please advise). So I don't see any need to perform any special "local realism inspired" experiments. I hope the bulk of the existing experiments were conducted by knowledgeable people, and I don't care whether they were local realists, positivists, solipsists, whatever. There is no disagreement on what experimental results were obtained. But everybody is free to draw their own conclusions, as long as such conclusions are consistent with the established experimental data.

Er.. no. The issue with the detection loophole is that one has to "select" the data and ignores what are called "background" and "missing counts". For some who are arguing that the detection loophole could contribute to the apparent violation of Bell inequality, this "background" and "missing counts" are the crutch that they rest on. This is extremely weak, because if it is true, then even high energy physics will have a problem. But the statistics are what saved both of them. And this is why no one from the "other camp" have even attempted to duplicate the experiment to argue that the detection loophole is what is causing all this. I find that a very glaring absence.

Look, you know a lot about photodetectors, I know next to nothing about them. Does this mean that I am not in a position to have my own opinion on the violations of the Bell inequalities? I don't know if people who first raised the issue of the detection loophole knew a lot about photodetectors.

You can raise it as much as you like, but if you are arguing that there are no inequalities due to a detection loophole, then you'd better know what "detection" means! The fact that no experiment has ever been done to show exactly where the detection loophole rears its ugly head to influence the outcome should mean something to you.

Are you telling me I must accept the fair sampling assumption just because I don't know anything about photodetectors? Am I supposed to believe in god just because I have not read the bible or the quran? Then in which god am I supposed to believe - the god of the bible or the god of quran? I just know that the fair sampling assumption is not generally recognized, so I am free to accept or to reject it. I choose to reject it, and I am left with no experimentally observed violations of the Bell inequalities.

Yet, you have no reluctance to reject them, even when you just admitted that you're no expert in them. What gave you such powers?

Do you also reject the evidence for superconductivity, the presence of quasiparticles in Fermi Liquid Theory, the fractional charges in 2D conductors, the giant magnetoresistance effect, the CP-violation in Kaon decays, etc.. etc? Or are you already experts in those fields as well and have the knowledge to know not to reject them? If you are not, then I don't see you complain about the validity of those phenomena, which means that you depend on experts in those fields to proclaim their validity that allow you to use some of them in your modern electronics. How come you aren't bringing up their validity based on the "philosophy" you are living by? Why only pick on this?

OK, forget about local realism for a moment. In my posts I was trying to explain why I don't admit (as long as it is possible without a contradiction with experimental data)that there is any experimental evidence of violations and why I don't expect any such evidence to appear. You might have noticed that my reasoning had little to do with local realism. Furthermore, I actually swore by quantum theory. You see, you cannot prove the Bell inequalities without the projection postulate (please advise if I am wrong). The projection postulate introduces irreversibility, whereas unitary evolution of the quantum theory allows no such thing. You cannot have it both ways - unitary evolution and the projection postulate. You have to choose. I choose the unitary evolution, which is quantum theory, pure and simple. You may say: but we have to accept both mechanics (classical or quantum), where there is no irreversibility, and thermodynamics with its irreversibility. Yes, but we understand that the irreversibility of thermodynamics is just a very good approximation, as the underlying mechanics does not allow any rigorous irreversibility. You need some "manual" interference, however subtle, to obtain irreversibility. And it seems that the analogy between mechanics vs. thermodynamics, on the one hand, and unitary evolution vs. projection postulate, on the other hand, may be deep enough, as arXiv:quant-ph/0702135 (Phys. Rev. A 64, 032108 (2001), Europhys. Lett. 61, 452 (2003), Physica E 29, 261 (2005)) demonstrates, using a rigorously solved model, how the projection postulate evolves as a result of thermodynamic irreversibility.

Then you are not having a problem with the experiments, but with Bell theorem itself. This is a separate issue. However, considering that the rigorous formulation and update to Bell has been done, and this includes both the GHZ and the Leggett's inequalities, I'd say that unless you are willing to stick your neck out and publish your own version of why it doesn't do what it says it is doing, then I would say talk is cheap. I would then turn around and apply your own philosophy and ask you why in the world I would accept what you have to say when you have shown nothing. Apply your own principle to what you are doing and you'll realize that what you claim has even less of a leg to stand on than what we know about Bell theorem.

Zz.

 

[Demystifier]

What about all the people who believe that both QM and SR are correct?

They are not realists, i.e., they do not believe that there is objective reality existing even without measurements.

 

[Hans de Vries]

They are not realists, i.e., they do not believe that there is objective reality existing even without measurements.

I completely disagree here. Most of the people who believe that both QM and SR are
correct do not even consider a world without objective reality. Regards, Hans.

 

[Demystifier]

I completely disagree here. Most of the people who believe that both QM and SR are
correct do not even consider a world without objective reality.

Maybe you are right that most people think that way, but is their view logically consistent? Do they understand the Bell theorem? I don't think so.

I was referring to those whose view is logically consistent, and you may be right that they do not represent the majority.

 

[Fra]

Most of the people who believe that both QM and SR are
correct do not even consider a world without objective reality.

Just to add my vote :)

I don't believe in a fundamentally objective reality, or rather do I see any way of firmly and objectively defining this objectivity, because all the tools at hand are subjective, so even my definition of objectivity is bound to be subjective. But paradoxally the notion of some kind of objectivity is essential because it's the reference we relate back to, it warrants stability, and saves us from chaos. So I think that the process of "emergence of en effective objectivity" is physically interesting. But the idea that there exist some universal, constant objectivity that is of any use to an observer ignorant about it is alien to me.

I think neither QM nor SR are correct as in the fundamentally objective sense.

It strikes me that this obsession with locality becomes different in an information view. If you like I do, consider the observers image of reality, as the relevant microstructure under examination, then locality or non-locality seems to just be related to the way information is stored and encoded on this image or "screen" through which we are bound to interact with the uknown(=realiy?).

I think that non-local interactions are almost synonymous to improbable correlations in the spacetime structure(yet more fog). It could be that spacetime is by construction balanced so as to minimze non-local things. So SR becomes a result of self-organisation of the image on the screen. Persistent and confident non-local interaction should certainly somehow call for a deformation of spacetime itself, shouldn't it?

/Fredrik

 

[Hans de Vries]

Do they understand the Bell theorem?

In post #22 I reasoned that non-locality can not be proved by the failure of Bell's
specific type of hidden variable theory.


Regards, Hans.

 

[Fra]

As I see it in addition to the association of spacetime as a kind of information geometry, is that I don't see how an observation of some apparently non-local causation can be distinguishable from a normal correlation which are explained as statistical fluctuations?

Violations of any based on experience concluded expectations are not in contradiction to anything as long as the confidence in the violation are low relative to the confidnce in the expectations. And in any case, a rational decistion maker, will certainly dynamically update his expectations if violations are frequent. So one might suspect that violations are to be only transient anyway?

If you add to this the idea that all decisions are made on incomplete or limited information, then we seem doomed to play this game. There are only two options, play and risk beeing wrong, or refuse to play and dissipate ;)

/Fredrik

 

[Demystifier]

In post #22 I reasoned that non-locality can not be proved by the failure of Bell's specific type of hidden variable theory.

First, even if you are right, this is not what most people think (we were talking about the majority, right?).
Second, I disagree with you. I think that the Bell result disproves ALL local hidden variable theories.
Third, there are even more direct proofs that QM and local hidden variables are incompatible. The Hardy proof is the simplest one.

 

[Hans de Vries]

First, even if you are right, this is not what most people think (we were talking about the majority, right?).
Second, I disagree with you. I think that the Bell result disproves ALL local hidden variable theories.
Third, there are even more direct proofs that QM and local hidden variables are incompatible. The Hardy proof is the simplest one.

A majority of Quantum Field Theorists does not believe in causal relations outside the
lightcone and the textbooks teach accordingly, Peskin and Schroeder, Weinberg...

Regards, Hans.

 

[Demystifier]

A majority of Quantum Field Theorists does not believe in causal relations outside the
lightcone and the textbooks teach accordingly, Peskin and Schroeder, Weinberg...

Is that supposed to be in contradiction with something that I said?

Do I really need to say that quantum nonlocalities cannot be used to send instantaneous human-controlled information?

 

[Fra]

A short conceptual question for Demystifier:

I'm interested to follow your reasoning and your idea to try to connect to string theory but what about this.

In the Bohmian context and "hidden variables", do you consider these "hidden degrees of freedom" to be objective? Or how about the notion where the distinguishable degrees of freedom is subjective (meaning observer relative)? Does that such a thought make any sense at all from your point of view?

/Fredrik

 

[Demystifier]

A short conceptual question for Demystifier:

I'm interested to follow your reasoning and your idea to try to connect to string theory but what about this.

In the Bohmian context and "hidden variables", do you consider these "hidden degrees of freedom" to be objective? Or how about the notion where the distinguishable degrees of freedom is subjective (meaning observer relative)? Does that such a thought make any sense at all from your point of view?

Fra, in the Bohmian context the hidden degrees of freedom are objective. I am not sure that I understand the rest of your question, but just ask yourself what would be an analogous question in classical (not quantum) physics. Whatever your answer in the classical case might be, essentially the same answer can be applied to the Bohmian version of quantum physics as well.

 

[Hans de Vries]

Is that supposed to be in contradiction with something that I said?

Do I really need to say that quantum nonlocalities cannot be used to send instantaneous human-controlled information?

local is inside the light-cone.
non-local is outside the lightcone.

First you say that the viewpoint that both QM and SR are correct is logically inconsistent
and then there is the familiar statement above...

Like: Yes, well, ok, SR is not really violated because of ... Regards, Hans

 

[ueit]

I think that the Bell result disproves ALL local hidden variable theories.

What about superdeterminism? It is a logical possibility, accepted by Bell as such.
A short description about how I think a superdeterministic theory might look like.

Assume that the universe is a sort of deterministic computer game (say chess).
Assume that what we call "entangled" particles are bishops that were at some time on adiacent squares. This will ensure that whenever such bishops are observed they will sit on different colors.

Of course, that cannot explain EPR-type results, but here the superdeterministic idea comes in. There is another constraint that has not been taken into account, the observation itself. Assume that only a king is conscious and can "measure" the bishops. But the king itself cannot move anywhere, its motion is also constrained by the rules of chess. So, it is possible that when this suplimentary set of constraints (those that deal with the act of measurement itself) are accounted for, the EPR correlations might arrise.

So, I think that non-locality cannot be proven because superdeterministic theories are not ruled out by Bell.

 

[ZapperZ]

What about superdeterminism? It is a logical possibility, accepted by Bell as such.
A short description about how I think a superdeterministic theory might look like.

Assume that the universe is a sort of deterministic computer game (say chess).
Assume that what we call "entangled" particles are bishops that were at some time on adiacent squares. This will ensure that whenever such bishops are observed they will sit on different colors.

Of course, that cannot explain EPR-type results, but here the superdeterministic idea comes in. There is another constraint that has not been taken into account, the observation itself. Assume that only a king is conscious and can "measure" the bishops. But the king itself cannot move anywhere, its motion is also constrained by the rules of chess. So, it is possible that when this suplimentary set of constraints (those that deal with the act of measurement itself) are accounted for, the EPR correlations might arrise.

So, I think that non-locality cannot be proven because superdeterministic theories are not ruled out by Bell.

But this is now similar to pre-Bell times where people are arguing things based on simply a matter of tastes simply because each argument cannot be falsified. So unless you have a "super-Bell Theorem" that can make such tests, what you just said here cannot be verified either. This makes any kind of rational discussion on which one is valid or not rather meaningless.

Zz.

 

[ueit]

But this is now similar to pre-Bell times where people are arguing things based on simply a matter of tastes simply because each argument cannot be falsified. So unless you have a "super-Bell Theorem" that can make such tests, what you just said here cannot be verified either. This makes any kind of rational discussion on which one is valid or not rather meaningless.

Zz.

No, I simply pointed out that by rejecting the so-called "statistical independence assumption" Bell's theorem cannot be derived anymore, therefore non-locality cannot be proven using this theorem.
In other words Demystifier should prove that for any possible hidden variable theory the constraints imposed by the theory over the act of measurement itself average to zero. While this is usually assumed to be true it might not be so.

 

[JesseM]

No, I simply pointed out that by rejecting the so-called "statistical independence assumption" Bell's theorem cannot be derived anymore, therefore non-locality cannot be proven using this theorem.
In other words Demystifier should prove that for any possible hidden variable theory the constraints imposed by the theory over the act of measurement itself average to zero. While this is usually assumed to be true it might not be so.

Similarly, it might be true that the laws of physics allow opposite charges to sometimes attract and sometimes repel, but they also conspire to ensure that whenever an intelligent being is measuring them they are always attracting. But these sorts of ideas make nonsense of the idea of discovering the laws of physics via the scientific method, and they also seem to require that the laws of physics have some kind of high-level understanding of what it means for a system to be "measured", as opposed to being interacted with in some other way.

 

[akhmeteli]

Why do you say this? The projection postulate is part of (some interpretations of) quantum mechanics; I don't see where it's used in deriving Bell's inequalities.

Maybe I should have said "Bell theorem", rather than "Bell inequalities", to avoid any misunderstanding. In the Bell theorem, first, the Bell inequalities are derived based on some assumptions, and, second, one proves that these inequalities can be violated in quantum mechanics. The projection postulate is used in this second part of the Bell theorem, when probabilities are calculated.

 

[ueit]

Similarly, it might be true that the laws of physics allow opposite charges to sometimes attract and sometimes repel, but they also conspire to ensure that whenever an intelligent being is measuring them they are always attracting. But these sorts of ideas make nonsense of the idea of discovering the laws of physics via the scientific method, and they also seem to require that the laws of physics have some kind of high-level understanding of what it means for a system to be "measured", as opposed to being interacted with in some other way.

OK, I've used the idea of a conscious observer in my example but this is not necessary at all. You can replace him with an atom or whatever particle the entangled particles may interact with. The relevant properties of that particular atom (position, spin, whatever), on which the measurement result depends, are constrained by the laws of physics. It is not obvious for me that the effect of those constraints averages to zero for every conceivable theory. I agree that it is a nice assumption from a practical stand-point but I don't see it as necessary true.

 

[Demystifier]

So, I think that non-locality cannot be proven because superdeterministic theories are not ruled out by Bell.

I agree. But I don't take superdeterminism seriously, even though I admit that it is a logical possibility.

After all, strictly logically, NO PHYSICAL LAW CAN BE PROVED EVER. But science is not pure logic. When we say that science has proved something, we do not mean it in a strict logical sense.

 

[Demystifier]

local is inside the light-cone.
non-local is outside the lightcone.

First you say that the viewpoint that both QM and SR are correct is logically inconsistent
and then there is the familiar statement above...

Like: Yes, well, ok, SR is not really violated because of ...

If there is reality beyond measurements, then this reality is NOT described by standard quantum theory. You see (unexisting) inconsistencies because you mix standard quantum theory with hypothetical reality. The former obeys the principles of relativity, the latter, if exists, violates them. Don't mix apples with oranges!

 

[Fra]

After all, strictly logically, NO PHYSICAL LAW CAN BE PROVED EVER. But science is not pure logic. When we say that science has proved something, we do not mean it in a strict logical sense.

I think this a good focus. We can not be absolutely confident in anything, not even the laws of physics. I agree completely.

Now what conclusions do we make of this?

My personal action is to turn my attention to the emergence of the laws in the first place, and consider this to be a physical process (as oppose to just some human stuff), and try to ask myself it the confidence in the laws of nature can somehow subjectively be rated. Objective ratings are when a group of observers happens for whatever reasong to agree and the subjective ratings correlate.

I think that is what is going on. And the limited confidence in any law in the first place suggest that observed violations of this law are consistent with observation as long as the information associated with theo observations are not significant to distort the laws to the given confidence.

Thus I think even the laws of physics are in motion. At least the observations of laws are in motion, and that's all I have on my table anyway.

/Fredrik

 

[akhmeteli]

You are supposed to write a rebuttal to all those tons of published papers and argue why their claim of violation of Bell inequalities are not valid, rather than argue them here. Pick up any of these papers, and there are clearly words and phrases to the effect that they claim they observed such violations.

Why should I do it? What's the point? Other people did it, raised the issue of the detection loophole and so on. Furthermore, the arguments of those people were accepted by knowledgeable people. Let me quote Shimony again: "Most of the dozens of experiments performed so far have favored Quantum Mechanics, but not decisively because of the “detection loophole” or the “communication loophole.”" Just think about it, if there was just one repeatable evidence of the Bell inequalities violation, how could it be indecisive? The problem is people typically do not substitute the measured probabilities (correlations) in the Bell inequalities, but first adjust these probabilities using the fair sampling assumption. You call it a violation, I don't. Shimony is not quite happy about loopholes, you are not quite happy (otherwise why would you mention the recent paper on the Bell ineq. violations in the other forum, unrelated to this discussion, if you believe that the evidence of violations is overwhelming as it is?), and yet you want me to be quite happy?
So I don't write rebuttals because others did it better than I could, and I argue here because not everybody here knows the experimental situation as well as you do.

Er.. no. The issue with the detection loophole is that one has to "select" the data and ignores what are called "background" and "missing counts". For some who are arguing that the detection loophole could contribute to the apparent violation of Bell inequality, this "background" and "missing counts" are the crutch that they rest on. This is extremely weak, because if it is true, then even high energy physics will have a problem. But the statistics are what saved both of them. And this is why no one from the "other camp" have even attempted to duplicate the experiment to argue that the detection loophole is what is causing all this. I find that a very glaring absence.

Again, what's the difference what camp an experimental physicist is in, as long as he/she is not a fraudster? Experiments without loopholes are too difficult for the experimental state of the art, so why do you demand that people from the "other camp" do something that nobody can do now? I don't know any other experimental way "to argue that the detection loophole is what is causing all this" but to perform an experiment without loopholes. There is a theoretical way, though: as far as I know, there are local realistic models where the Bell inequalities are not violated, but the fair sampling assumption does not hold. Existence of such models proves that the inequalities "violated" in existing experiments are a far cry from the true Bell inequalities.

You can raise it as much as you like, but if you are arguing that there are no inequalities due to a detection loophole, then you'd better know what "detection" means!

Actually, I don't need to. Because you know what "detection" means, and that's enough for me. Indeed, you know the experimental situation, and, being honest and knowledgeable, you are not trying to state that the Bell inequalities were violated in such and such experiment. You know that it is not so. You are just saying that some manipulations with raw data, such as using the fair sampling assumption, are a necessary evil, that after such manipulations the inequalities are indeed violated, and that eventually violations without loopholes will be demonstrated. However, you do know that no experiments have demonstrated true Bell inequalities (with appropriate spatial separation). If I am wrong, and you believe that there are such experiments, please advise.

The fact that no experiment has ever been done to show exactly where the detection loophole rears its ugly head to influence the outcome should mean something to you.

Maybe it should. But it does not. Maybe the fact that no experiment has demonstrated genuine violations should mean something to you. But it does not. Life is tough.

Yet, you have no reluctance to reject them, even when you just admitted that you're no expert in them. What gave you such powers?

As I did say that "I am not an expert in the Bell inequalities", I assume that by "them" you mean the Bell inequalities, not something else. I am not sure I reject the Bell inequalities, I do agree that they can be derived under some assumptions. I just believe, based on opinions of experts, such as Shimony, that genuine Bell inequalities have not been violated in experiments so far. And I don't feel you disagree. Do I really need any magic powers for such belief? I also doubt there will ever be any such violation demonstrated. But I cannot be sure that won't happen.

Do you also reject the evidence for superconductivity, the presence of quasiparticles in Fermi Liquid Theory, the fractional charges in 2D conductors, the giant magnetoresistance effect, the CP-violation in Kaon decays, etc.. etc? Or are you already experts in those fields as well and have the knowledge to know not to reject them? If you are not, then I don't see you complain about the validity of those phenomena, which means that you depend on experts in those fields to proclaim their validity that allow you to use some of them in your modern electronics. How come you aren't bringing up their validity based on the "philosophy" you are living by? Why only pick on this?

As I said, I rely on experts in many areas, including the Bell inequalities. However, you're right, I am not happy with the Copenhagen interpretation (or any other, to be precise) and feel that there is a glaring contradiction between unitary evolution and the projection postulate. Actually, some results of my own research (and they are not philosophical at all) strongly influence my opinions in this area.

Then you are not having a problem with the experiments, but with Bell theorem itself. This is a separate issue. However, considering that the rigorous formulation and update to Bell has been done, and this includes both the GHZ and the Leggett's inequalities, I'd say that unless you are willing to stick your neck out and publish your own version of why it doesn't do what it says it is doing, then I would say talk is cheap. I would then turn around and apply your own philosophy and ask you why in the world I would accept what you have to say when you have shown nothing. Apply your own principle to what you are doing and you'll realize that what you claim has even less of a leg to stand on than what we know about Bell theorem.
Zz.

I am not sure I have problems with the Bell theorem. As far as I understand, it just states that certain assumptions imply certain inequalities, which are violated in quantum mechanics. I don't see any holes in the proof. However, I have problems with the projection postulate of quantum mechanics (and I said why), which, as far as I can understand, is used in the proof. This is the reason I am not sure any violations will be demonstrated experimentally. Although I can be dead wrong.

And I fully agree with you, I have shown nothing. I am not sure I said anything original. I just said that so far no experiments have demonstrated violations of the genuine Bell inequalities. And I don't have to worry whether you'll accept it or not for the simple reason that you don't seem to dispute that. If I am wrong, and you do dispute it, please advise. You are saying that deficiencies of the existing experiments are not important, and you may well be absolutely right. I am not sure though. Anyway, I don't see how I can "stick my neck" and what I can "publish", as I said nothing new. I can just say "don't kill the messenger". I said pretty much the same as Shimony, but maybe my words were less diplomatic. Why should not I call a spade a spade?

 

[ZapperZ]

Because Shimony's opinion doesn't reflect what the majority of people in this area accept as valid.

You still haven't addressed the two facts that I mentioned earlier. The FACTS were : a set of experiments that closed the detection loophole (but not the locality loophole) claimed to violate the Bell inequality, and the set of experiments that closed the locality loophole (but not the detection loophole) violates the Bell inequality. I asked you if, knowing how these experiments work and how such things are detected, that you think it is simply mere coincidence that they both arrive at the identical conclusion even when they not only use different entanglement/objects to detect, but also different loopholes that were possibly left open. Remember, the "detection" loophole has a different set of statistics that has nothing whatsoever to do with the "locality" loophole. Unless you've never done any experiment in your life, changing two different, independent conditions should not give the same type of results!

And I don't buy this argument that you have no need to the details of the experiment. In fact, I would say that your ignorance of the experiment IS the source of this disagreement. The knowledge of what a photodetector can and cannot do is vital in the degree of confidence in the result. I will put it to you that you have placed your life and the lives of your loved ones on knowledge with the SAME degree of confidence as what we get out of the photodetectors used in these experiments.

It is a FACT that there are no Bell-type experiments being conducted has ever proclaim that these loopholes were responsible for the apparent Bell violation. As an experimentalist, when I look at the body of evidence, and the lack of even ONE experiment to cast a doubt on the conclusion, then there is an overwhelming evidence for the validity of that conclusion. You throw around the word "proofs" as if we have "proofs" in physics. Find me something in physics that has the "proof" that you accept. Again, you have picked on these experiments, while ignoring the fact that other parts of physics have the same "baggage".

You never did tell me whether you accepted all the various phenomena that I listed. Are you experts in those areas as well so much so that you know the intricate details to know that they are valid? If not, then how come you don't complain about, say, the validity of the experiments in superconductivity? why are you sitting back and accepting the conclusions from the experts on this, but not for the Bell-type experiments?

For every Shimony, there are dozens of Zeilingers. Why you accept one, but ignore the others, I have no idea.

Zz.

 

[confusedashell]

Just out of curiosity ZapperZ, which interpretation are you proponent of, if none, which you think are on the right track?

 

[Demystifier]

Just out of curiosity ZapperZ, which interpretation are you proponent of, if none, which you think are on the right track?

Apparently, he didn't voted here:
https://www.physicsforums.com/poll.php?do=showresults&pollid=978

By the way, confusedashell, why didn't you voted for the Bohmian?