Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A Bell Theorem with no locality assumption?

  1. Aug 13, 2010 #1


    User Avatar
    Science Advisor
    Gold Member

    I am opening a new thread to continue discussion of some interesting ideas around EPR and Bell. Specifically, this is about the idea of realism, and whether it is tenable in light of Bell and other HV no-go theorems. Note: I usually use Hidden Variables (HV) and Realism interchangeably although some people see these as quite different. I also tend to use Realism as being an extension of EPR's "elements of reality" as a starting point for most discussions. After all, if a physical measurement can be predicted with certainty without disturbing what is measured... well, I would call that as real as it gets.

    charlylebeaugossehad thrown out a few ideas in another thread - especially around some papers by Charles Tresser. So I suggest we discuss around these:

    We prove here a version of Bell Theorem that does not assume locality. As a consequence classical realism, and not locality, is the common source of the violation by nature of all Bell Inequalities.

    In Bohm's version of the EPR gedanken experiment, the spin of the second particle along any vector is minus the spin of the other particle along the same vector. It seems that either the choice of vector along which one projects the spin of the first particle influences at superluminal speed the state of the second particle, or naive realism holds true i.e., the projections of the spin of any EPR particle along all the vectors are determined before any measurement occurs). Naive realism is negated by Bell's theory that originated and is still most often presented as related to non-locality, a relation whose necessity has recently been proven to be false. I advocate here that the solution of the apparent paradox lies in the fact that the spin of the second particle is determined along any vector, but not along all vectors. Such an any-all distinction was already present in quantum mechanics, for instance in the fact that the spin can be measured along any vector but not at once along all vectors, as a result of the Uncertainty Principle. The time symmetry of the any-all distinction defended here is in fact reminiscent of (and I claim, due to) the time symmetry of the Uncertainty Principle described by Einstein, Tolman, and Podolsky in 1931, in a paper entitled ``Knowledge of Past and Future in Quantum Mechanics" that is enough to negate naive realism and to hint at the any-all distinction. A simple classical model is next built, which captures aspects of the any-all distinction: the goal is of course not to have a classical exact model, but to provide a caricature that might help some people.

    We prove here a version of Bell's Theorem that is simpler than any previous one. The contradiction of Bell's inequality with Quantum Mechanics in the new version is not cured by non-locality so that this version allows one to single out classical realism, and not locality, as the common source of all false inequalities of Bell's type.
  2. jcsd
  3. Aug 14, 2010 #2
    Well ok, i'll start.

    I'm a layman so the math is well over my head, a bit like the moon, unless I'm not looking at it, in which case it doesn't exist, until i look at it, and then it does, and did when i wasn't looking at it. Didn't it?
  4. Aug 14, 2010 #3
    Those preprints where not published not rewritten to be more final so far, but someone posted the link to the paper that appeared on European J. of Phys. in the thread on action at a distance and EPR. One of the papers cited by DrC (thanks again for opening the thread)
    relates more to the issue of ANY is not ALL in QM in the sense that in Classical Mechanics, ANY observation doable means ALL observation doable (at once) while this is not true in QM: in forward time because of the UP, in backward time, because of the time reversed UP of Einstein, Tolman, and Rosen (ETP). ETP also tels us that a single observale can hardly have a definite value before measurement, the first PROOF of non-realism, but his does not apply to EPR particles where two conjugate observable can pre-exist measurement if one measurement is made on each of the 2 particles. Then one had ANY but not ALL, while for generic particle, one does not even have any observable. Notice than the EPR case, the values are retrodictive (before the measurement, something that was accepted by Bohr and Heisenberg, at least until ETP showed that problematic). In some sense, the ANY is not ALL issue is a Bell-EPR type of issue but does not belong PRECISELY to this thread (but who cares?).

    I have uploaded (after all, it's easy) two more recent things that I have about the title of the thread: a copy of the paper and a preprint that makes part of the paper more precise (in fact the Bell theorem part as the GHZ part remains untouched). People interested in this thread might need some time to go through these two papers (or at least the preprint as the paper has been out for a few weeks) before expressing opinions or asking questions.

    This being said it seems to me that one essential ingredient in the "Bell Theory with no locality assumption" story is the hypothesis: Effect After Cause Principle. It states that for any Lorentz observer no effect can be observed and then be changed by a later caused. Yet the effect may happen after the cause (if one assumes Non-Locality to hold true, for instance), or one can ALSO assume Locality in which case of course the COMBINED hypothesis is not much different from locality. Once the EACP is recognized as not being nor implying locality, and in fact being weaker an hypothesis than locality, it remains of course to see if the proof is fine. As this is not math but physics, the issue does not reduce to a formal correctness issue.

    Attached Files:

  5. Aug 14, 2010 #4
    The hardest math in there is trigonometry, a bit of proba, and simple considerations about convergence issues.Where is the block. As for the moon, that is another issue: should we really start on that here?
  6. Aug 15, 2010 #5
    Hi DrC! Nice topic. Thanks, and to CleBG, for the links.
  7. Aug 17, 2010 #6
    One so yearns for dots to be connected. For instance: how does Tresser relate to Leggett-Garg (specifically in re: the Gisin group and Zeilinger group experiments?). CleBG's remark about the Effect After Cause Principle feels so much like the Suarez (and Gisin) Before-Before stuff ...
  8. Aug 18, 2010 #7
    Could you provide a clear definition of that principle?
    Lets forget about the definition of Cause and Effect: they both are Events, and the definition of Event in QM is interpretation-dependent. But this is irrelevant now. Just assume that Cause and Effect are 2 events in the BlockTime.

    How you define the difference between Cause and Effect? For me, Effect is in the lightcone of Cause, nothing more. (In CTL it could be laso vice versa). But then (in flat spacetime) Effect is always after the Cause - just by definition!

    Finally, the part about "then be changed..." denies the blocktime postulating (even imaginary) possibility of a "butterfly effect".

    I would say, this "principle" is lyrics, not physics.
  9. Aug 18, 2010 #8


    User Avatar
    Science Advisor

    "In science one tries to tell people, in such a way as to be understood by everyone, something that no one ever knew before. But in poetry, it's the exact opposite."
    Paul Dirac
  10. Aug 18, 2010 #9


    User Avatar
    Gold Member

    Great quote Demystifier! And funny! :smile:
  11. Aug 18, 2010 #10


    User Avatar
    Gold Member

    Last edited by a moderator: Apr 25, 2017
  12. Aug 18, 2010 #11


    User Avatar
    Science Advisor

    Thanks! If you want more, visit my blog. :smile:
  13. Aug 18, 2010 #12


    User Avatar
    Gold Member

    Bin there, done that! :wink:

    (And all your quotes are great!)
  14. Aug 18, 2010 #13
    I'll answer to my best, as a physicist as I have a fair voice but lyrism is not my thing. Others may have better words to explain, and the question is nice but too bad to have it in an aggressive tone. While Chaos is really my main forte since the mid 1970's, I must say that I need explanation on what the butterfly effect is doing here. I'll answer to anon-aggressive version of the question/argument till the word "Finally".

    I am not a non-localist, but the direct effect of non-locality is that effects (values of spin projection observed by Alice, or inferred as existing at Alice's location as a consequence of the chosen realism hypothesis (and I am not a realism believer either) ) depend on a cause
    (the axis used by Bob to make his measurement) that is spatially remote. So by you position you refuse non-locality, and the paper also tries to recuse non-locality, but without starting from the opinion that locality is right.

    So one starts from 2 hypothesis:
    Effect After Cause Principle (formulated so that it is compatible with non-locality) and realism to get a Bell inequality incompatible with QM.The oddness that you see in the EACP are not new when compared to usual non-locality sort of consideration. Most people in awe of Bell Theorem tell that the way violation is avoided by Nature is that Locality is false. The goal of the mentioned paper (in the title of this thread) is not to tell people convinced of locality that only locality makes sense among locality and non-locality. The goal is to prove a Bell Theorem without using locality so that non-locality cannot save the day. Now I do believe that it is microscopic realism that is the issue among the hypotheses of Bell's inequality (and in all form of Bell's Theorem, with or without inequalities as GHZ). More particularly, I believe that microphysics does not allow two EPR-Bohm particles to have a total of at least 3 projections of the spin defined at once (or two projections on each side for a total of four direction in the CHSH form of Boole-Bell inequality (for the role of name of Boole, see the work of Pitowsky, for the stat theory behind that, see the papers of Fine in J. Math Phys and the paper in PRL at about the same time). With less than 3 projections making sense at once for the two EPR-B particles together, be it 0, one, or two, there is no Bell inequality: more precisely, there is no Boole type inequality incompatible with the correlations that are compatible with quantum mechanics.

    But it was a question worse rising: hope the answer satisfied most people troubled by the question. It has been known for a long time (relatively) that non-locality violates the intuition of time we learned from Special Relativity. Gisin considers that since speed of light as a limit is new to physics, it i not too much a big deal if we have to abandon that. The worse part is that some time ago, he still had the support of very bizarre sects (perhaps still does). Physics is hill: the trauma brought to that science by people who like Bell (but also other masters) defended non-locality of QM, and realism in the case of Bell is very troublesome: for me it is as if Algebra would be under the control who tell you that equation must have both sides multiplied by zero to hold true, or some other non-sense.
    The great Antony Leggett wrote, more than once, that among physicists, it is essentially consensual that out of the two hypotheses of Bell's Theorem, the one that is false is realism: statitics by people around Marland Scully tell us just the contrary. I would be troubled by one QM supporting classical microscopic realism in a major US university (or UK, or French, or German, or Italian, or Japanese (any country with some tradition of geat physics where no criminals came to power like in Argentina). Science is at risk: as it goes, this beautiful invention going back to the Greeks will disapear very soon: are physicists (who would be recognized as such by Einstein, Feynman, Bohr, etc...) an endangered specie??
    What is clear to me is that with the level of lack of professionalism in citations and quotation, another discipline, in particular one using less science, would already have fully disappeared.

    Sad, no? But do I have a pessimistic view?
    Or am I about right?
    And then what should we do?

    PS: Ask for specific uploads, and I'll provide what I can (i.e., what I have in soft form or what I know where to find in the www, but others have provided competent help in the threads that I have seen: uploading is a bit too hard for me to do it right away).
  15. Aug 21, 2010 #14
    Can you make precise what should be connected to what:papers and indications on what to look at said papers. For me Leggett-Garg is about realism consequences at the macroscopic level. Several people have participated to that debate following them. As for the massive production of the Switz and Austrian groups, they are so massive that pointers are unavoidable. The experiments are most often great, the theory often to par but Gisin has written many papers that are causes of confusion and the position of Zeilinger on realism and locality is hard for me to read so that I'd love to have pointers. I remember having seen things I id not like more that Gisin's but I was still much more new to the field. And when you say "Tresser relates", do you mean the person, or a particular point of the positions taken in the papers. Anyway, any mean to see better in the massive production that you have mentioned would be great, as well as understanding all of Leggett's contributions (alone or accompanied) to the subject.

    Someone else has raised the issue of comparing the EACP with things of Gisin but the allusion is both lacking precision and lacking a precise reference (or several as appropriate).
  16. Aug 22, 2010 #15
    I read the paper twice. I did not see any problems with it; which does not mean it is without problems!
    What bothers me is that 1. It seems to rule out dBB (which most HV nogo theorems do not). 2. It is four years old and is not a BIG DEAL. What am I missing?

  17. Aug 22, 2010 #16
    Not sure of what dBB means. In such precise matters, guessing would be inappropriate.
    The paper has just been published and went mostly unnoticed for a long time. What I got from the analysis of a very serious physicist whom I would not name here without his consent, is that in a first approximation, each time the paper was refused, any argument against it for one referee was mostly the main argument in favor for another referee who would also have his/her critics similarly matched. Frankly, the first version were very unclear to my opinion and even the version that got published still needed the corrections in the version for Pierre Coullet's 60th birthday where the EACP's compatibility with non-locality (and with locality, but this is not the issue, except to make clear that the EACP is neither of Locality ad Non-Locality) was more explicitly used and the effect of the other side on the pair is more carefully analyzed.

    Now, if the paper is as right as you seem to consider, AND if enough people know what is in there and agree with you appreciation, then the impact should be visible as this paper
    seriously support the viewpoint that it is classical realism and not locality that is the false hypothesis in Bell's theory, and that the old masters where right to support non existence of microscopic realism, on which the opinion of Einstein seems to have been dramatically miss-represented (in order to let people like Gisin and other present themselves as better than Einstein) if we believe the historical work of Jammer and Fine. Miss quotations and miss refereeing is rampant in this field as if it would be closer to politics than to science.
    One useful thing that Gisin did (beside some nice experiments) was to lacye when it became kosher to write about Bell'sTheory and non-locality: a paer of Ekert that is patently false (as I explain in a pre-print that cries for a co-author to help me). A joint friend has proposed to Ekert (or so he told me) to write a paper with me to explain the mistake he did when invoking Bell theorem in his method for QM-based cryptography. I never heard back from Ekert who wrote the paper that made Bell's theory Kosher: his paper is in fact more cited than the original paper on Q-cryptography (Bennett-Brassard), but Bennett, with Brassard and with and without Mermin, made serious crimes of lack of scholarly acceptable

    Well to be short(er) at last, in a subject dominated by supporters of Bell (who often ignore that they defend a strong supporter of realism) a paper posted for 4 years on arXive and always refused has little chance of making an impact. Let's see if a published paper, and hopefully more to go, change the status of all that so the we can come back to real issues and advance physics in what is probably its main sub-discipline, i.e., QM and more precisely micro-physics in the non-relativistic and then in the relativistic regimes, with then progress in all issues that could benefit from a better understanding of the basics.
  18. Aug 23, 2010 #17


    User Avatar
    Gold Member

    dBB = http://en.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theory" [Broken]

    (also called the pilot-wave theory, Bohmian mechanics, and the causal interpretation)
    Last edited by a moderator: May 4, 2017
  19. Aug 23, 2010 #18
    The most illuminating Zeilinger paper IMO is one he did back in 2005 with Brukner. "Quantum Physics as a Science of Information". SpringerLink sits on it unless you want to give them money. But fortunately the gods have smiled it's available in toto via Google books:


    Leggett-Garg incorporates a nonlocality assumption and tests for "realism" (counterfactual definiteness, whatever). Inferentially or by default that's what Tresser also does. Or not? Anyway, how are they different?

    Gisin indeed has become awfully hard to follow. He's hiking deep into the woods. As I read him (and this interpretation of mine may very well be of absolutely no value whatsoever) it's all somehow coming back to the measurement problem. Hopefully what he's up to will become clear in time.
  20. Aug 24, 2010 #19
    With Leggett this paper shares the essential point that it is realsim and not locality that is the false asumption in Bell's theory. The author hope to soon write down "proofs" of that in terms of new experiments proposals and a discussion on the basis of the old superposition principle. One tool will be revisiting the Welcher Weg discussion, with views that are not new to Dirac, but new to Einstein, Bohr, Wheeler, Feyman, and many others, including Zeilinger whose views on locality are not fully clear, are they?

    As for Gisin, he loves to explain why he is smarter than Einstein and is smart enough indeed to know that the paper that HE points out as THE start of the fashion for Bell theorem in PRL is a fraud.
    Like many other localists, he cannot be (also) stupid enough despite his remarkable technical expertise in theory (beside experiments) for not let one think that he is not deeply lacking scientific honesty. Look at who support him: not so long ago (and for now I did not check) he had the support of some strange sect. This stinks, for me and I do not see any of that behind the paper being discussed. My understanding is that within a few month, another paper by the same author will be posted on QuantPh Arxive on
    WW, superposition and taking the issue of local realism out of metaphysics into physics with application to interferences with or without delay. I am not privy of the title to look for, nor of the exact content. The Bell paper took about 4-5 years to be accepted (while long ago Pitowsky told the author that there were so many new ideas in there that the paper should appear anyway, or something of that sort).
    Last edited by a moderator: Apr 25, 2017
  21. Aug 24, 2010 #20
    That's what I thought, but I wanted to be sure: sonce that paper is anti-realist, it is as anti dBB as possible, siding strongly with Pauli and Einstein on that matter.
    Last edited by a moderator: May 4, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook