How do you understand EPR & Bell's Theorem?

[Blue Scallop]

This is related to vanhees71 statements (see bottom). I'll explain.

First. The idea is very simple. As summary: Einstein showed that if reality was objective and quantum theory complete, then there had to be nonlocal effects. But since nonlocal effects can violate relativity, then there had to be hidden variables (insert the EPR arguments here about measurement position and momentum of the two entangled particles which I don't want to detailed again)

Now enter Bell's Theorem. I read in a bell book "Bell wanted to see what the quantum world would be like if local hidden variables really existed - and here the word "local" is important. Local hidden variables refer to physical quantities which locally determine the state of an object inside an imaginary surface. By contrast, nonlocal hidden variables could be instantaneously changed by events on the other side of the universe. Assuming that any hidden variables are "local" is the assumption of local causality. Using this assumption, Bell derived a mathematical formula, an inequality, which could be checked experimentally. The experiment has been done independently at least half a dozen times, and Bell's inequality - along with its central assumption of local causality - was violated. The world, it seemed, we as not locally casual!"

Of course it doesn't mean we can send signal faster than light. Bell's equality violation is due to very detailed nature of the Clauser and Aspect experiments which showed that when the polarization of A is changed, somehow B is correlated. If wave function is ontic or objective, then there may be non-local hidden variables. If wave function is epistemic, then the correlations are in the equations yet there are still distant correlations via the equations. This is in contrast to the arguments about blue and red socks being determined from the beginning. For this message. let's assume the wave function is epistemic and the correlation occur via the equation even light years apart.

Now enter Vanhees71. I've been analyzing his thoughts for many days. It's important because he is one of the most powerful physicists around in the sense that he represents the rest of the physicists who don't care about interpretations and only lab results. There is nothing wrong with this.. but I think there is big possibility he hasn't really understood the exact setup for the Clauser and Aspect experiments.. so he argued there was correllation but it occurred in the initial preparation just like the blue and red socks arguments.

This is very important to scrutinize because vanhees71 has been describing it for 7 years and the purpose of physicsforums is to educate.. but somehow some of us get quite confused by his arguments that don't conform to the mainstream idea of Bell's experiments. Ok here's the proof. I read all his messages about Bell's Theorem and the following summarized his views. Vanhees71 states:

"It is very important to distinguish between nonlocal interactions (which are by construction not present in standard relativistic QFT) and longranged correlations, which are a specific consequence of any quantum theory, including relativistic QFT. It is not that measurement of one photon's polarization in an entangled two-photon state that causes the correlation but the preparation of the entangled two-photon state in the very beginning of the experiment. Although the single-photon polarizations are maximally indetermined, the 100% correlation between them is there all the time due to the preparation procedure in an entangled state. There is no non-local interaction between the single photons and the polarizers and photo detectors at A's and B's far-distant locations whatsoever! Consequently, according to local relativistic QFT it cannot be nonlocal interactions caused by the local measurements at the local detectors. This shows that these experiments do not contradict standard QED and thus there's no need for nonlocal interactions!"

Reference: Msg 21 of https://www.physicsforums.com/threads/collapse-of-a-wave.845613/page-2#post-5309489

But Clauser and Aspect experiments showed changing A can somehow affect B when comparing the notes later.. as if the correlation is in the equations (that is, even if wave function is not objective and there is no non-locality.. no problem with this as the correlations can occur in the equation).. this means even when light years apart.. the statistics of each can be affected what one does in the other end.. while Vanhees71 arguments is they never do and the long distant correlation is in the initial preparation..

Am I wrong or Vanhees71 wrong? If Vanhees71 is right, and the correlation is due to the initial preparation.. but doesn't this go against the results of the Crauser and Aspect experiments? Please tell me what is the case. Thank you.

 

[PeterDonis]

Assuming that any hidden variables are "local" is the assumption of local causality.

No, it isn't, it's the assumption of locality. In quantum field theory, "causality" has a subtly different meaning: it means, roughly speaking, that measurements at spacelike separated events must commute, i.e., their results cannot depend on the order in which they are performed. All known quantum field theories are consistent with this requirement. But note that this requirement does not prohibit correlations between spacelike separated measurements that violate the Bell inequalities. Only the "locality" assumption does that--i.e., only the locality assumption is shown to be false by experiments confirming that the Bell inequalities can be violated.

 

[PeterDonis]

Clauser and Aspect experiments showed changing A can somehow affect B when comparing the notes later

No, they didn't. They only showed that the correlations between the measurements violate the Bell inequalities. The measurements still commute--the results don't depend on the order in which you make them. That means, if you think about it, that one measurement can't be "affecting" the other in the way you are thinking.

 

[Nugatory]

But Clauser and Aspect experiments showed changing A can somehow affect B when comparing the notes later..

They did not. They showed that the result of a measurement on one member of an entangled pair cannot have been determined solely by preexisting properties of that particle.

 

[Blue Scallop]

No, they didn't. They only showed that the correlations between the measurements violate the Bell inequalities. The measurements still commute--the results don't depend on the order in which you make them. That means, if you think about it, that one measurement can't be "affecting" the other in the way you are thinking.

I know there is no way to send signal if the data is random because the end is still random.. I read "At first you might think that by changing A's polarizer we have directly influenced the number of errors produced at B. Hence by altering A's polarizer to various settings in a sequence of moves, B could, by observing the alteration in the number of errors produced at B, get a message from A - a telegraph that would violate causality. But no information can possibly be transmitted from A to B using this device because holding a single record of events at either A or B would be like holding the message of a top-secret communication in a random code - you can't ever get the message. Because the sequences at A and B are always completely random there is no way to communicate between A and B. That is how real nonlocality is avoided by the God that plays dice; He is always shuffling the deck of nature."

My point is. There are still correlations even if nature uses randomness encryption to make sure no information is transferred, right?

 

[PeterDonis]

I know there is no way to send signal if the data is random because the end is still random.

That's true, but we don't really know why the data is random. To say that something is "always shuffling the deck" is somewhat optimistic since we don't actually know that a shuffling process is what is happening underneath the random results we see. We just know that they're random, and that the results don't depend on the order of measurements.

There are still correlations

Yes. And the correlations can contain information--they just can't contain information that gets propagated faster than light.

Another way of looking at this is that, if Alice and Bob are both making spacelike separated measurements on entangled particles, they can use the measurements as part of a process that transmits information, but the measurement results alone can't be all of the process. They also need to communicate which measurements they made--for example, which direction they oriented their polarizers if the entangled particles are polarized photons--and that communication has to take place via an ordinary slower-than-light channel. Then that information can be put together with the measurement results to decode a message that one transmits to the other.

 

[Blue Scallop]

That's true, but we don't really know why the data is random. To say that something is "always shuffling the deck" is somewhat optimistic since we don't actually know that a shuffling process is what is happening underneath the random results we see. We just know that they're random, and that the results don't depend on the order of measurements.
Yes. And the correlations can contain information--they just can't contain information that gets propagated faster than light.

Another way of looking at this is that, if Alice and Bob are both making spacelike separated measurements on entangled particles, they can use the measurements as part of a process that transmits information, but the measurement results alone can't be all of the process. They also need to communicate which measurements they made--for example, which direction they oriented their polarizers if the entangled particles are polarized photons--and that communication has to take place via an ordinary slower-than-light channel. Then that information can be put together with the measurement results to decode a message that one transmits to the other.

Yes. I know it. But vanhees71 still emphasized the correlations are in the initial preparation only. Also he said:

"Bell's theorem is about local "realistic" deterministic theories and shows, in form of an inequality, that QT cannot be equivalent to such a theory, and that's its very point. It made the philosophical mumblings of Einstein and Bohr a scientifically testable statement, and it has now been tested to overwhelming precision in favor of QT."

Reference: msg 191 https://www.physicsforums.com/threads/collapse-from-unitarity.860627/page-10#post-5419662

He may be right above but he thinks it is the end result. He just ignored the correlations can occur long distant and not in the initial preparation only (this is the most important point). Hope vanhees71 can clarify later about this as this is what I read in all his messages and been bugging me for a week.

 

[PeterDonis]

He just ignored the correlations can occur long distant and not in the initial preparation only

You are making an incorrect distinction here. The "long distant" correlations are there because of the initial entanglement of the particles, i.e., because of the initial preparation. That's what @vanhees71 is saying. The experiments showing violations of the Bell inequalities don't show that there have to be additional "long distant" correlations occurring, over and above the initial preparation. They just show that the initial preparation cannot be a matter of setting classical hidden variables. It has to be something else, something that doesn't have any intuitive classical analog.

 

[Blue Scallop]

You are making an incorrect distinction here. The "long distant" correlations are there because of the initial entanglement of the particles, i.e., because of the initial preparation. That's what @vanhees71 is saying. The experiments showing violations of the Bell inequalities don't show that there have to be additional "long distant" correlations occurring, over and above the initial preparation. They just show that the initial preparation cannot be a matter of setting classical hidden variables. It has to be something else, something that doesn't have any intuitive classical analog.

If the initial preparation doesn't have a matter of classical hidden variables. Then the values are being adjusted on the fly by the equation when A made a measurement.. so the additional "long distant" correlations are the equations themselves giving the values (remember there is no hidden variables) so the observable can only produce one eigenstate or values only upon measurement..

 

[PeterDonis]

If the initial preparation doesn't have a matter of classical hidden variables. Then the values are being adjusted on the fly by the equation when A made a measurement

No, this is not correct. Neither one of these things is what is going on. That is what the experiments are telling us. If the initial preparation were setting classical hidden variables, then the Bell inequalities would not be violated. But if the values were being adjusted on the fly when distant measurements were made, then the predictions of QFT would be violated--but they aren't. So, as counterintuitive as it seems, neither one of those things can be what's happening.

 

[Blue Scallop]

No, this is not correct. Neither one of these things is what is going on. That is what the experiments are telling us. If the initial preparation were setting classical hidden variables, then the Bell inequalities would not be violated. But if the values were being adjusted on the fly when distant measurements were made, then the predictions of QFT would be violated--but they aren't. So, as counterintuitive as it seems, neither one of those things can be what's happening.

What I meant by values being adjusted on the fly is the equation giving the born rule and giving values.. but not real time but only during measurements.. ok..i 've been studying this reference in:https://www.physicsforums.com/threads/nick-herberts-proof.589134/page-6

'Step One: Start by aligning both SPOT detectors. No errors are observed.
Step Two: Tilt the A detector till errors reach 25%. This occurs at a mutual misalignment of 30 degrees.
Step Three: Return A detector to its original position (100% match). Now tilt the B detector in the opposite direction till errors reach 25%. This occurs at a mutual misalignment of -30 degrees.
Step Four: Return B detector to its original position (100% match). Now tilt detector A by +30 degrees and detector B by -30 degrees so that the combined angle between them is 60 degrees.'"

Are you familiar with the example? I think it's not far from Clauser experiment... Double error should produce less error.. so it's related to Bell's Inequality.. my point is.. there are correlations which you agree too but it uses randomness encryption and not real time...

 

[PeterDonis]

What I meant by values being adjusted on the fly is the equation giving the born rule and giving values.. but not real time but only during measurements

My comments still apply. You are thinking of "values being adjusted on the fly" as something that is happening. That's not correct. Saying that it only happens "during measurements" doesn't change that; it's still not correct.

 

[PeterDonis]

i 've been studying this reference

There are lots of proofs of Bell's Theorem. None of them change what I'm saying, because all of the proofs of Bell's Theorem are just different ways of showing that if the initial preparation were a matter of setting classical hidden variables, then the Bell inequalities could not be violated. But the Bell inequalities are violated in actual experiments. Finding a different proof of Bell's Theorem doesn't change that.

 

[PeterDonis]

it uses randomness encryption

What does this mean? What actual thing in the math are you referring to here?

(Yes, you probably can't answer that question, because you haven't been looking at the math, you've been looking at posts here on PF. That's my point.)

 

[Blue Scallop]

My comments still apply. You are thinking of "values being adjusted on the fly" as something that is happening. That's not correct. Saying that it only happens "during measurements" doesn't change that; it's still not correct.

What is the simplest paper or reference about details of the Clauser or Aspect experiments that you have encountered?

 

[PeterDonis]

What is the simplest paper or reference about details of the Clauser or Aspect experiments that you have encountered?

I don't have a good simple reference handy at the moment, unfortunately. It's been some time since I dug into the literature in this area.

 

[Blue Scallop]

There are lots of proofs of Bell's Theorem. None of them change what I'm saying, because all of the proofs of Bell's Theorem are just different ways of showing that if the initial preparation were a matter of setting classical hidden variables, then the Bell inequalities could not be violated. But the Bell inequalities are violated in actual experiments. Finding a different proof of Bell's Theorem doesn't change that.

Perhaps the best way to think about long distance entanglement and Bell's Inequality is to just think QM applies everywhere no matter if the systems are near or far. About the initial preparation. Well. The book says: "In measuring the polarization of the photon we put it into a definite state, but this alters the initial conditions of the experiment. This is identical to the problem we faced in the two-hole experiment with the electron. By observing with light beams which hole the electron went through we changed the detected pattern. Likewise, the very act of establishing the objective state of the photon alters the conditions under which Bell's inequality was derived. The attempt to experimentally verify the objectivity assumption has the consequence that the conditions of the experiment are altered in just such a way that we can no longer use the violation of Bell's inequality to conclude that nonlocal influences exist."

Point is. When A or B measured the polarization, it has effective of altering the initial conditions. So the correlations are not entirely due to the initial preparations only but the measurement of A and B can affect the final results too.

If anyone knows the clearest details about the Aspect or Clauser experiments. Please share it because I'd like to understand the original experiment itself. Thanks.

 

[zonde]

There are lots of proofs of Bell's Theorem. None of them change what I'm saying, because all of the proofs of Bell's Theorem are just different ways of showing that if the initial preparation were a matter of setting classical hidden variables, then the Bell inequalities could not be violated. But the Bell inequalities are violated in actual experiments. Finding a different proof of Bell's Theorem doesn't change that.

That's incorrect.
Nick Herbert's proof does not relay on "classical hidden variables" in any way. And Eberhard's proof does not relay on "classical hidden variables" as well.

 

[Demystifier]

Am I wrong or Vanhees71 wrong? If Vanhees71 is right, and the correlation is due to the initial preparation..

I think that the idea that "correlation is due to initial preparation" is meaningless if one does not answer the following question: Preparation of what?

And I think that vanhees71 cannot give a consistent answer to that question without implicitly assuming the existence of some hidden variables. But he refutes to claim their existence explicitly, because he knows that this would imply non-locality by the Bell theorem, and he cannot accept non-locality.

It does not mean that he does not have answers to all these questions. He does. But I think that his answers are not self-consistent. At some points he claims that QM only talks about statistical ensembles of systems, and not about single systems, while at other points he claims that QM says also something about single systems.

And all this does not mean that there are no consistent local interpretations of quantum correlations. There are plenty of them, as I reviewed in https://arxiv.org/abs/1703.08341 Sec. 5.3. But the vanhees71's interpretation is not one of them.

 

[Demystifier]

It's important because he is one of the most powerful physicists around in the sense that he represents the rest of the physicists who don't care about interpretations and only lab results.

If he just didn't care about interpretations, that would not be a problem. The problem is that he doesn't care about interpretations, but likes to talk about them.

 

[Blue Scallop]

I think that the idea that "correlation is due to initial preparation" is meaningless if one does not answer the following question: Preparation of what?

That's right. Preparation of what. I'm assuming the following:

I'm assuming the initial preparation is the entangled positronium photons. Maybe vanhees71 and peterdonis has other ideas of "initial preparation"? In the above case which is similar to Clauser experiment. Is it not changing the polarization at any end can change the results? What initial preparation was vanhees71 referring to when he said it all has to do with the initial preparation (like the blue and red sock analogy)?

And I think that vanhees71 cannot give a consistent answer to that question without implicitly assuming the existence of some hidden variables. But he refutes to claim their existence explicitly, because he knows that this would imply non-locality by the Bell theorem, and he cannot accept non-locality.

It does not mean that he does not have answers to all these questions. He does. But I think that his answers are not self-consistent. At some points he claims that QM only talks about statistical ensembles of systems, and not about single systems, while at other points he claims that QM says also something about single systems.

And all this does not mean that there are no consistent local interpretations of quantum correlations. There are plenty of them, as I reviewed in https://arxiv.org/abs/1703.08341 Sec. 5.3. But the vanhees71's interpretation is not one of them.

 

[Demystifier]

That's right. Preparation of what. I'm assuming the following:

View attachment 211656

I'm assuming the initial preparation is the entangled positronium photons. Maybe vanhees71 and peterdonis has other ideas of "initial preparation"? In the above case which is similar to Clauser experiment. Is it not changing the polarization at any end can change the results? What initial preparation was vanhees71 referring to when he said it all has to do with the initial preparation (like the blue and red sock analogy)?

In the picture above I don't see actual photons. I see only some actual macroscopic lab gadgets plus some imagined photons. Is QM only about macroscopic gadgets? Or is it also about invisible photons? Tough questions for someone who wants to defend an instrumental interpretation of QM (like vanhees71) and be consistent.

 

[Blue Scallop]

If he just didn't care about interpretations, that would not be a problem. The problem is that he doesn't care about interpretations, but likes to talk about them.

The forces of nature like electromagnetism, weak force, strong force has no connection to quantum interpretations. The forces work the same even in the minimal interpretation or where you don't want to think of the underlying ontology.. how about quantum gravity and the most difficult of all, the relativistic quantum etc. theory of consciousness. Will they be related to any quantum ontology dynamics or are they are like the other forces of nature? Do you have any theorem or axiom that can state what is the case. Because if quantum gravity, etc. is guaranteed not to be related to any quantum ontology. Then I want to forget about interpretations and be minimal intepretation proponent like vanhees71 as I spent over a decade studying intepretations already and want to bypass it if the ontology won't help anything in the search for further area in physics.

 

[vanhees71]

I think that the idea that "correlation is due to initial preparation" is meaningless if one does not answer the following question: Preparation of what?

Preparation of what you prepare ;-)). In this case it's the preparation of a polarization-entangled two-photon state. The preparation procedure is to shoot a laser beam into some crystal and choosing photon pairs a long the directions fitting the phase-matching conditions.

And I think that vanhees71 cannot give a consistent answer to that question without implicitly assuming the existence of some hidden variables. But he refutes to claim their existence explicitly, because he knows that this would imply non-locality by the Bell theorem, and he cannot accept non-locality.

I don't understand, how you can come to that conclusion. All I'm using is standard QED, no hidden variables, and it's obviously sufficient to describe the empirical findings.

It does not mean that he does not have answers to all these questions. He does. But I think that his answers are not self-consistent. At some points he claims that QM only talks about statistical ensembles of systems, and not about single systems, while at other points he claims that QM says also something about single systems.

This I also don't understand. In this case you prepare each single photon pair in the discussed polarization-entangled state, i.e., the state of the system is completely determined. For the polarization of the pair as a whole that means it's determined, for other observables it means they don't have determined values, and you can make only probabilistic statements about the outcome of measurements of these observables. This is in accordance with standard QT, and there is no (obvious) contradiction in that.

And all this does not mean that there are no consistent local interpretations of quantum correlations. There are plenty of them, as I reviewed in https://arxiv.org/abs/1703.08341 Sec. 5.3. But the vanhees71's interpretation is not one of them.

Well, I thought my view is of the flavor 5.3.1, i.e., Copenhagen (but without collapse and without cut).

 

[Demystifier]

Well, I thought my view is of the flavor 5.3.1, i.e., Copenhagen (but without collapse and without cut).

But 5.3.1 is explained in more detail in Sec. 2.2, where it is pointed that reality exists and is non-local, but that is just not the subject of physics. And I don't think that you agree with that. You think that my claim that "reality exists and is non-local" is meaningless because it cannot be tested in a laboratory. I agree that it cannot be tested in a laboratory, but I don't think that it makes it meaningless. After all, that is based on a sound logic, and logic is meaningful even when it cannot be tested in a laboratory.

 

[Blue Scallop]

Preparation of what you prepare ;-)). In this case it's the preparation of a polarization-entangled two-photon state. The preparation procedure is to shoot a laser beam into some crystal and choosing photon pairs a long the directions fitting the phase-matching conditions.

I don't understand, how you can come to that conclusion. All I'm using is standard QED, no hidden variables, and it's obviously sufficient to describe the empirical findings.This I also don't understand. In this case you prepare each single photon pair in the discussed polarization-entangled state, i.e., the state of the system is completely determined. For the polarization of the pair as a whole that means it's determined, for other observables it means they don't have determined values, and you can make only probabilistic statements about the outcome of measurements of these observables. This is in accordance with standard QT, and there is no (obvious) contradiction in that.

I always wanted to ask you this since a month ago vanhees71. Are you completely familiar with the details of the Clauser and Aspect experiments? in this thread https://www.physicsforums.com/threads/bells-theorem-easy-explained.562942/ some basic was shared like http://quantumtantra.com/bell2.html

"Reduced to bare bones, the proof looks like this. A's move causes a mismatch of 1 quarter of the bits in the message. Likewise B's move causes a mismatch of 1 quarter of the total bits. Taking both moves together (and assuming a local reality) the most mismatch that can occur is 2 quarters.

1 + 1 = 2 (or less) is what a local reality predicts.

However quantum theory and quantum fact lead to an entirely different outcome. The combined moves of A and B result in a mismatch of 3 quarters of the total number of bits in the message.

1 + 1 = 3 is what nature actually produces.

No local reality can explain these facts. Therefore reality must be non-local. Furthermore this conclusion is not a supposition or speculation but a mathematical proof. John Bell found a way to ask a question about reality itself--not merely about theory or fact. And he obtained a clear and surprising answer: reality is non-local.

In this situation, a non-local reality means that what happens at Miss A's SPOT detector--whether this particular photon registers as "1" or "0"--cannot depend on causes in Anaheim alone but must somehow depend also on the setting of Mr B's distant detector in Baltimore. To explain results like these only a non-local reality will suffice.

So even though the facts are local--nothing we can measure at A changes when detector B is rotated--the nature of the strong correlations observed between A and B necessitate that the reality that underlies this experiment be non-local."

Note "non-local" is just generic words. If there are no properties before measurement.. then just use the word "correlations".. but the fact is that changing the polarizer at A or B ends can cause the results to differ.. so why do you keep saying it has everything to do with the initial preparation?

I hope someone can give other web sites with even more clearer example.. thanks.

Well, I thought my view is of the flavor 5.3.1, i.e., Copenhagen (but without collapse and without cut).

 

[Demystifier]

In this case you prepare each single photon pair in the discussed polarization-entangled state, i.e., the state of the system is completely determined.

Now you effectively claim that when something is known with certainty, then it is real. But EPR proved that this very same criterion of reality, plus assumption of locality, implies that QM is incomplete. Yet, you don't accept that QM is incomplete. Another inconsistency in my view.

 

[Lord Jestocost]

To say:

"It is not that measurement of one photon's polarization in an entangled two-photon state that causes the correlation but the preparation of the entangled two-photon state in the very beginning of the experiment. Although the single-photon polarizations are maximally indetermined, the 100% correlation between them is there all the time due to the preparation procedure in an entangled state."

is a fundamental misunderstanding of quantum theory. You cannot turn quantum ignorance (where the desired information simply doesn’t exist) into classical ignorance (where the desired information exists but is hidden) by peering at the preparation.

 

[Blue Scallop]

For students of physics from undergraduate to Ph.D. What years of study when the full details of the Aspect experiments were taught in school.. or none and the person needs to study this off curriculum?

 

[Demystifier]

For students of physics from undergraduate to Ph.D. What years of study when the full details of the Aspect experiments were taught in school.. or none and the person needs to study this off curriculum?

Like most of the coolest stuff in physics, this is out of standard curriculums.