Hidden Variables
Bell's theorem - for or against Hidden Variables? - Printable Version

+- Hidden Variables (https://ilja-schmelzer.de/hidden-variables)
+-- Forum: Foundations of Quantum Theory (https://ilja-schmelzer.de/hidden-variables/forumdisplay.php?fid=3)
+--- Forum: The Violation of Bell's Inequalities (https://ilja-schmelzer.de/hidden-variables/forumdisplay.php?fid=7)
+--- Thread: Bell's theorem - for or against Hidden Variables? (/showthread.php?tid=8)

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28


RE: Bell's theorem - for or against Hidden Variables? - Thomas Ray - 07-25-2016

(07-24-2016, 11:20 PM)secur Wrote: Thomas Ray wrote: To be frank, I was put off by Joy Christian's title, "Disproof of Bell's theorem" ...

Me too. But a rose will still smell as sweet, and an incorrect paper will still be as incorrect, no matter what name you give it Smile

Thomas Ray wrote: You say Bell's theorem introduced new physics ...

It set off the investigations (Aspect and other experiments) which demonstrated the peculiar type of non-locality inherent in QM. That's "new physics", compared to the old idea that no influence (however minimal) can be FTL. Admittedly it's still not entirely clear, due to still-extant "loopholes".

Thomas Ray wrote: By renouncing spacetime, Bell's theorem (and quantum theory based on it) has renounced relativity, and its proofs run in circles. If you don't believe it -- get Richard Gill to define a measure space for Bell-Aspect; get him to describe what happens if Planck's constant goes to zero.

I don't know in what sense Bell "renounces spacetime". If Planck's constant "goes to zero" we would get classical physics; but of course Planck's constant doesn't go anywhere: it is what it is. If Gill wants to address your comments it would probably help me understand what you're getting at.

Thomas Ray wrote: The problem is, relativity is solid -- LIGO is only the latest in a long line of spacetime validation.

SR is solid, it agrees with all experiments. GR much less so, although various aspects of it have, indeed, passed the test. But, pending discovery of "Dark Matter", it disagrees with many observations of star and galactic speeds, which grossly violate Newton's inverse square law. Finally both SR and GR are unsound philosophically. The fact that we can't - yet - detect an absolute reference frame in no way justifies the assertion that such doesn't exist. But I still don't see the relevance to Bell.

Thomas Ray wrote: Have you noticed the quantum theoretical fringe trying to do away with spacetime?

No, I haven't; but this brings up the most interesting aspect of this whole debate. I can't tell who's "fringe" and who isn't! The big surprise was that LM doesn't "believe Bell" either - he rejects non-locality. As I mentioned a few posts ago he called George Muller an "idiot" (i.e., his opinion differs). Who cares? Well, to me, LM represents "establishment physics". I've never seen him miss the party line in physics (or, indeed, anything else). So that indicates that you, Christian, FrediFizzx et al are not the fringe - evidently I, Gill, Schmelzer, Bell, Aspect, Zeilinger (etc) are! This has nothing to do with math, physics, science or logic. But as a sociological phenomenon it's very remarkable. Someone ought to write a book on this whole subculture of "fringe physics", there's a fascinating story to be told.

Concerning Hess and Philipp: I'd already noticed the discussion on page 5 (that Gill mentions above). It basically admits the problem.

From H&P: "Then, since the joint probability conditional on {R = m} depends on both settings, the marginal distribution of capital-lambda (a, t) for setting a conditional on {R = m} may be different. How can this be without instantaneous action at a distance? The answer is that if c would have been chosen, then over a whole sequence of measuring times all the settings would be different. ..."

This is not very convincing, to put it mildly.

But - in Hess's favor -, AFAIK experimenters still haven't rigorously excluded a time-dependent "loophole". Do we really know that entanglement will "work" in other settings than these typical photon-based experiments? In quantum cryptography, computing, and similar discussions, Alice, Bob, and their friends fly all over the universe with entangled particles in boxes, doing amazing tricks with them. But that's all fantasy. The sad truth is that in real labs, over distances of centimeters (not light-years), with the expenditure of millions of dollars, they still can't reliably demonstrate quantum computing. Maybe "Bell non-locality" is just an artifact of the particular experiments done so far? So Hess's main point - that time-correlation, in currently achievable experimental designs, has something to do with it - can't be dismissed so easily. AFAIK, IMHO, and pending further investigation.

secur,

I think I'll just deal with the speed of light postulate in your post.  

" ... the investigations (Aspect and other experiments) which demonstrated the peculiar type of non-locality inherent in QM. That's 'new physics', compared to the old idea that no influence (however minimal) can be FTL. Admittedly it's still not entirely clear, due to still-extant 'loopholes'."

That's why I don't believe new physics were introduced--that 'peculiar type of non-locality' was already restricted by Einstein's postulate, and the loopholes in a Bell-Aspect type of experiment will never be closed.  Events were called non-local because they lay in the future light cone (advanced wave solution) -- whether or not information is available from this solution was never in question, however, by the unity of spacetime and the indistinguishability of past and future events. ("All physics is local".)

"From H&P: 'Then, since the joint probability conditional on {R = m} depends on both settings, the marginal distribution of capital-lambda (a, t) for setting a conditional on {R = m} may be different. How can this be without instantaneous action at a distance? The answer is that if c would have been chosen, then over a whole sequence of measuring times all the settings would be different. ...'

"This is not very convincing, to put it mildly."

It is, if the results are pairwise correlated.  And to be consistent with special relativity they have to be pairwise correlated.

You have an open mind, secur, and my respect.

Make no mistake: Gill's rejection of local realism is a rejection of special relativity and the speed of light postulate.


RE: Bell's theorem - for or against Hidden Variables? - secur - 07-26-2016

Gill1109 wrote:

"Whether a theory is local or non-local depends, I think, on what you consider to be real. If you want to consider the outcomes of not-performed measurements as real, then QM is non-local. If you want to consider the wave-function as real, then QM is non-local. But if you accept only the reality of actual outcomes of performed measurements, and accept irreducible randomness as a fundamental part of reality, then it seems to me that QM is local.

But I am not a professional philosopher, nor a physicist, just a mathematician. I think we should not worry so much about locality and non-locality. Maybe it is time to forget some distinctions which used to be considered important. Perhaps the phenomena are trying to teach us that some old distinctions have less meaning than we thought. It seems to me that successful Bell experiments are teaching us that reality is non classical. Things apparently happen in these experiments which cannot be explained in a mechanistic way. QM allows some things which classically would have been thought to be impossible; but it also forbids other things. Reality is *different* from what we thought. Different from how evolution has programmed our brains to imagine reality. Right now I think we should reject local-realism but that the idea that one of the two (locality or realism) has to be rejected and the other can be kept is too simplistic. It's more useful to explore the possibilities offered by QM and maybe adapt our ideas of locality and realism accordingly."

Thanks Gill1109,

"Non-local" has different meanings in different applications. In the current context "local" means "within the past light cone". Modern Bell experiments are designed so that when Alice and Bob make their spin measurements their stations are non-local in one key sense: their detector settings can't be known at the other station, because they're spacelike separated. In fact no observer in the universe can know both those settings, when the detections are made. Those two angles are "non-local" with respect to each other.

To me the phrase "QM is non-local" means the following in this context. When we analyze and predict mathematically the results of the experiment, those two non-local variables must appear together in the same equation. In fact, we must use the cosine of the sum of the angles (or, the dot product of vectors representing the detector settings) to predict the correlations of Alice and Bob's two detections (or a series thereof). This happens nowhere else in physics! To analyze any other experiment, and predict its results - or a function of the results, like correlation coefficients, or moments - it's always sufficient to use only the information available in the past light cone. Except in this one case. Here we must use two variables that no possible single observer could have known, at the time of the measurement. This very peculiar and unique situation can reasonably be called "non-local".

What does one do when confronted with a paper which claims to reproduce the results of a Bell experiment, like Christian or Hess? One knows that somewhere hidden in the math, those two variables have been used together. Maybe a key subscript has been dropped, or maybe a limit is taken to two separate values at once, or whatever. When you find it, you find their mistake (or trick). So my definition of "non-local" is an operational, practical definition.

Since the term "non-local" makes people uncomfortable, we can call it "property X" instead: the mathematical analysis must use, in the same equation, variables which are not within each other's past light cones.

As we've seen, Property X can be rigorously defined just by looking at the math. But the question inevitably arises, what does it mean physically? One might suppose that some "influence" travels between Alice and Bob detector stations faster than light. But other explanations are available under the general heading of "non-realism". There's rejection of counterfactual definiteness, rejection of "free will" and determinism, "nature conspiracy" ideas, "consistent histories", and others I've never heard of. Ontological issues arise: is the wave function real? Is irreducible randomness real? Epistemological issues arise: is the "collapse of wave function" a change in real physical information, or just a change in our knowledge? As far as I know, it's a complete waste of time to debate these questions. Instead we should ask, what hypotheses can be formed, and what experiments performed, that might help decide the issue?

Here's an example. Suppose the hypothesis is that an FTL signal of some sort travels between the two detector stations, between the two entangled particles, to ensure they get the right correlations. If it can't be tested, then it's mere philosophy. But we could try to block the signal by shielding. What if we put a thick slab of lead, or a powerful EM field, between the two stations? Does the correlation fail? I'm almost certain it wouldn't. I bet you could put a billion light years, and a supernova or active quasar, between them: QM would still work as advertised. But in the unlikely event that shielding destroyed the correlation, you see, that would support the hypothesis that an FTL signal travels between them. Such an experiment would be real physics, and even with the expected null result, far more useful than any amount of ontology.

I offer this only as an example, no doubt you can think of more useful experiments. The main point I hope we can agree on: don't worry about philosophy, much less terminology. Instead concentrate on math and experiments.


RE: Bell's theorem - for or against Hidden Variables? - secur - 07-26-2016

@ Thomas Ray,

Thomas Ray wrote: That's why I don't believe new physics were introduced--that 'peculiar type of non-locality' was already restricted by Einstein's postulate, and the loopholes in a Bell-Aspect type of experiment will never be closed.  Events were called non-local because they lay in the future light cone (advanced wave solution) -- whether or not information is available from this solution was never in question, however, by the unity of spacetime and the indistinguishability of past and future events. ("All physics is local".)

BTW as said in my above post to Gill, I'm thinking it would be a good idea to avoid the word "non-local" because it causes contention - we can call it "property X" or something. I agree it's conceivable all the loopholes will never be closed; Gill says something similar in his 2015 paper (the "fifth position", that Bell didn't mention).

However IF we assume the Bell experiments really do demonstrate what they appear to demonstrate, then information from the future light cone is being used to produce the experimental correlations. How Nature can access those two detector setting variables is unknown. Your suggested answer involves considering the exact properties ("unity", "indistinguishability of past and future events") of what we casually refer to as "space". Christian and FrediFizzx have a similar approach, based on "spinor properties" of space. Unfortunately the papers which attempt to prove these ideas aren't convincing, as far as I know at this time. What I'd really like to see is an experiment that could help decide the issue.

Thomas Ray wrote: Gill's rejection of local realism is a rejection of special relativity and the speed of light postulate.

There are many possible explanations of the Bell results. My personal preference is for a (partial) rejection of speed-of-light hypothesis: an FTL influence exists but can't be used for signaling. But this is, again, just philosophy, unless experiments can be done (or at least imagined) capable of falsifying or supporting the hypothesis.

Thomas Ray wrote: You have an open mind, secur, and my respect.

Thanks! My overall impression is, contrary to some rumors of personal strife I've heard, that everyone here is doing their best to work out these issues and communicate their thoughts.


RE: Bell's theorem - for or against Hidden Variables? - Heinera - 07-26-2016

Thomas Ray wrote: " "QRC" assumes that no data are hidden. So one should look for hidden data to prove the assumption.  Not finding any hidden data, one should conclude that no data are hidden.   If you know of a way to prove this negative proposition other than by double negation, I would like to hear it."

QRC is a challenge with some rules; it does not assume anything.  The challenge is to write a computer model with a source, and two detector stations.  Between these three, the only communication allowed is that the source can send information to the two stations.  The stations cannot send information to eachother.  The settings of the detectors wil be chosen at random, outside the control of the program.  The results should be in statistical significant violation of Bell's inequality (e.g., they should closely match the QM predictions).

This is of course the operational version of a local realistic model.  According to Bell's theorem, it is impossible to construct such a model and have it replicate all the QM correlations.  If there is an error in Bell's proof on the other hand, QRC will have a solution.  In other words, anyone who thinks the QRC can't be won, also aknowledges Bell's theorem, because these two statements are logically equivalent.


RE: Bell's theorem - for or against Hidden Variables? - gill1109 - 07-27-2016

(07-26-2016, 06:47 PM)secur Wrote: To me the phrase "QM is non-local" means the following in this context. When we analyze and predict mathematically the results of the experiment, those two non-local variables must appear together in the same equation. In fact, we must use the cosine of the sum of the angles (or, the dot product of vectors representing the detector settings) to predict the correlations of Alice and Bob's two detections (or a series thereof). This happens nowhere else in physics! To analyze any other experiment, and predict its results - or a function of the results, like correlation coefficients, or moments - it's always sufficient to use only the information available in the past light cone. Except in this one case. Here we must use two variables that no possible single observer could have known, at the time of the measurement. This very peculiar and unique situation can reasonably be called "non-local".
It is more subtle than this, I think. There is a local realistic model which predicts that the correlation is half the cosine of the difference between the angles. There is no local realistic model which predicts that the correlation is the full cosine.

To predict the correlation between both observer's measurements we need to know both observer's settings, in either case. No mystery about that.


RE: Bell's theorem - for or against Hidden Variables? - jrdixon - 07-27-2016

Speaking of redefining "local realism". I show that local realism, with the added assumption that nature has a limited ability to forecast future physical interactions, can explain the violation of a Bell inequality here:
vixra.org/abs/1103.0089
And also can explain a puzzling "no signaling" violation found in a recent "loophole free" Bell test experiment here:
vixra.org/abs/1606.0097


RE: Bell's theorem - for or against Hidden Variables? - secur - 07-27-2016

(07-27-2016, 11:19 AM)jrdixon Wrote: Speaking of redefining "local realism". I show that local realism, with the added assumption that nature has a limited ability to forecast future physical interactions, can explain the violation of a Bell inequality here:
vixra.org/abs/1103.0089        
And also can explain a puzzling "no signaling" violation found in a recent "loophole free" Bell test experiment here:
vixra.org/abs/1606.0097

This can be called a "prediction loophole". It's not surprising that QM results could be obtained this way. It could also violate no-signaling, but it seems to me that the same prediction algorithm couldn't do both? That is, if it mimics QM results exactly, by imperfect predicting, that same prediction algorithm couldn't also produce a no-signaling violation? That's why you had to modify it in your second paper. Anyway, I think the Quantum Randy Challenge (QRC) is a good filter for this type of loophole. You couldn't use this idea to beat the QRC. That puts it in an "exotic" class of loophole, what I might call a "Nature conspiracy" loophole, which is beyond the main theme of the discussion here.

One thing I like about your paper is that you propose possible experiments to determine if your idea is actually implemented in Nature. Very few proposals do that.


RE: Bell's theorem - for or against Hidden Variables? - jrdixon - 07-27-2016

It is the exact same setup in my two papers. In the second paper I simply introduce some notation to discuss how the emitter might toggle its choice of emission configuration among the two configurations allowed under its guess. Both the Bell inequality and the no signaling equalities can be violated in this setup, without nonlocal effects and without signaling. Yes, as the Adenier and Khrennikov paper I cite says, both quantum mechanics and traditional local realism (without forecasts) cannot expalin the no signaling violation, unless somehow there was signaling despite the expected closure of that loophole.


RE: Bell's theorem - for or against Hidden Variables? - secur - 07-27-2016

(07-27-2016, 08:14 AM)gill1109 Wrote:
(07-26-2016, 06:47 PM)secur Wrote: To me the phrase "QM is non-local" means the following in this context. When we analyze and predict mathematically the results of the experiment, those two non-local variables must appear together in the same equation. In fact, we must use the cosine of the sum of the angles (or, the dot product of vectors representing the detector settings) to predict the correlations of Alice and Bob's two detections (or a series thereof). This happens nowhere else in physics! To analyze any other experiment, and predict its results - or a function of the results, like correlation coefficients, or moments - it's always sufficient to use only the information available in the past light cone. Except in this one case. Here we must use two variables that no possible single observer could have known, at the time of the measurement. This very peculiar and unique situation can reasonably be called "non-local".
It is more subtle than this, I think. There is a local realistic model which predicts that the correlation is half the cosine of the difference between the angles. There is no local realistic model which predicts that the correlation is the full cosine.

To predict the correlation between both observer's measurements we need to know both observer's settings, in either case. No mystery about that.

For one thing, a correlation of half the cosine is no good because it doesn't give the right results. At an angle of pi, for instance, the true correlation is -1, but this would give 0.

But one can easily imagine semi-correct correlation functions that can be produced under local realism. The key necessity is that correlations of smaller angles, less than pi/2, are not so strong as in real QM. For instance, a correlation of (1 - theta * 2/pi) would do it. (Substituting theta = 2*pi - theta when theta is between pi and 2*pi.) Something like this will give roughly the right correlations. At least for theta=pi it is, correctly, -1. And it works fine without any knowledge of the other detector's setting. But for small angles, near 0, the correlation is too weak. So sure, a wrong correlation can be achieved easily in a local-realism model. But so what? It doesn't match experiments.

Perhaps it's harder than I thought to nail down the definition of "Property X". Let me try it as follows. Consider the QRC. One way to beat it would be to communicate detector settings between the two programs simulating the two stations. This "cheating" would be equivalent to an FTL signal between Alice and Bob's detectors.

If a real-time simulation of an experiment requires incorporating an FTL signal like that, to correctly mimic Nature: then the experiment possesses "Property X".

Can you see any problem with that definition? It's meant to be equivalent to the one I gave before, but less ambiguous.


RE: Bell's theorem - for or against Hidden Variables? - secur - 07-27-2016

(07-27-2016, 01:38 PM)jrdixon Wrote: It is the exact same setup in my two papers. In the second paper I simply introduce some notation to discuss how the emitter might toggle its choice of emission configuration among the two configurations allowed under its guess.  Both the Bell inequality and the no signaling equalities can be violated in this setup, without nonlocal effects and without signaling.  Yes, as the Adenier and Khrennikov paper I cite says, both quantum mechanics and traditional local realism (without forecasts) cannot expalin the no signaling violation, unless somehow there was signaling despite the expected closure of that loophole.

I noticed that "toggling" mechanism, and supposed that made it different from the first paper. Can you explain, briefly, how you can possibly reproduce QM faithfully while also violating no-signaling, since QM does not do that? On the face of it, it sounds illogical. Apologies for not reading your papers more carefully, since I'm primarily interested in "traditional" local realism.