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) |

RE: Bell's theorem - for or against Hidden Variables? - secur - 08-05-2016
Here's the video of Gell-Mann I referred to, https://www.youtube.com/watch?v=f-OFP5tNtMY. Overall it's good, worth watching, but the part about Bell starts around 11:45. Paraphrased, Gell-Mann says: Gell-Mann paraphrased: Bell's word "non-local" confused everyone. What they mean is "if it were a classical experiment, then it must be non-local". But it's not classical, it's QM! EPR is no different than Bertlemann's socks ... there is no non-local influence between the two particles. ... He winds up insisting that (his and Hartle's) decoherent histories interpretation is correct, and if you don't agree, you're an idiot (he doesn't use that word but the implication is clear). I can illustrate the problem best by giving what he should have said: What Gell-Mann should say: As I've just explained my "Decoherent Histories" is, I believe, the right explanation of QM "weirdness". Therefore I don't agree with others. In particular, there is no need for a non-local "influence" to travel between the two particles if you accept that there are different histories involved. BUT, of course none of this is proven yet. In fact it may never be provable - although I do think in time my theory will be supported. Until then it's just a matter of opinion. If you want to think in terms of a non-local influence, I can't say you're wrong. According to all experimental knowledge we have today, that ontology works. You see the difference. Both statements have exactly the same content as far as logic, math, physics. But the first leaves no room for discussion. Either you agree with him, or you're an idiot. The second allows everyone to agree on almost all of physics, apart from a couple rather obscure points. They can work together to devise experiments to determine who's right, in a civilized fashion. If either side turns out wrong it's not because they're stupid, but Nature gave a different answer than they thought. Thomas Ray wrote: We are learning, though, how the non-linearity of time contributes to the illusion of entanglement and superposition. I finally see you're not just proposing an alternative explanation of Bell, but all of QM! Superposition - probabilistic wave function - is at the heart of all standard QM ontologies. But they're illusions, according to your view, due to non-linear time. I don't doubt the case can be made: if you abandon orthocausal, linear time, you can reproduce QM without superposition or FTL influence. Elitzur et al's paper, that jrdixon referenced, uses retrocausality to achieve QM effects. That paper is very clear. Do you agree with it? Is that what you're talking about? If so it's a matter of opinion, at this time. We should concentrate on devising experiments to distinguish the two ontologies. You admit that the standard way of looking at QM, with probabilistic wave function and superposition, is an "illusion" - which means, it DOES seem that way. So until it can be proven otherwise it's Ok that some people (like myself) think it IS that way. Conversely I'm happy to agree that retrocausal time is another valid possibility - as Elitzur et al show - although it seems less intuitive. We agree on the actual experimental results, that's the main thing. Let's look for more advanced experiments that can help resolve different ontologies. If one or the other turns out to be right it doesn't mean the "losers" were idiots, just wrong. Thomas Ray wrote: Have I told you how delightful it is to have a reasonable discussion? There's far too much salesmanship going on - due, I think, to the funding game. If we carefully distinguish between knowledge, vs. opinion, these problems disappear. If someone disagrees with what we KNOW, they're wrong: say so. If they disagree with our opinion, however, they're not wrong (until such time the opinion becomes fact, via experiment), instead they're interesting. RE: Bell's theorem - for or against Hidden Variables? - Thomas Ray - 08-07-2016
(08-05-2016, 06:05 PM)secur Wrote: Here's the video of Gell-Mann I referred to, https://www.youtube.com/watch?v=f-OFP5tNtMY. Overall it's good, worth watching, but the part about Bell starts around 11:45. Paraphrased, Gell-Mann says: Hi secur, I haven't watched the video, though I am familiar with Gell-Man's opinion ("flapdoodle") from *The Quark and the Jaguar*. Joy Christian says much the same, for different reasons. Nonlocality has no physical meaning or purpose. Again, that's why Richard Gill wants to discard it; he is tasked to discard nonlocality without throwing out Bell's theorem. "I finally see you're not just proposing an alternative explanation of Bell, but all of QM! Superposition - probabilistic wave function - is at the heart of all standard QM ontologies. But they're illusions, according to your view, due to non-linear time." Bingo. "I don't doubt the case can be made: if you abandon orthocausal, linear time, you can reproduce QM without superposition or FTL influence. Elitzur et al's paper, that jrdixon referenced, uses retrocausality to achieve QM effects. That paper is very clear. Do you agree with it? Is that what you're talking about?" I'm still digesting that paper (thanks, John Dixon). It's clear in its conclusions -- not so much in its mechanism (at least, not to me yet). The authors claim that a particle does not exist in an 'up' or 'down' state until the experimenter chooses; I have a problem with that. Time symmetry in special relativity does not allow observer choice, as that would constitute a privileged frame. The key to my conjecture is the absolute indistinguishability of past and future events, expressed as entropy production. The authors state, "More recently, Elitzur and Cohen studied a simple quantum interaction [22] that ends up with an event and a nonevent, together appearing to violate momentum conservation. They showed that the nonevent in question is due to a 'Quantum Oblivion' effect, where a very brief virtual interaction undergoes 'unhappening.' Oblivion, they argued, underlies quantum IFM, erasure and several other peculiar effects. Venturing further to theory, they proposed [22, 23] a retrocausal evolution that accounts for such self-cancellation, involving exchange of negative physical values between earlier and later events." In a paper which I am trying (unsuccessfully so far) to initiate a collaboration at my ResearchGate site, I said, "Entropy hides its origin, so entropy is a good candidate for the ‘hidden variable’ of E-P-R reference. The lamba+ or lambda- which alternates (changes orientation) covariantly, depends on an initial condition which can be undone—giving us an impression of four lambdas: lambda+ lambda- lambda+ lambda-, which are taken to be in superposition until one is measured—an act that somehow transfers information at a distance to its partner. However, the distance is zero, following from an inability to distinguish entropy origin, past or future. The initial condition is done and undone, locally and non-linearly, by a global wave function that does not depend on the observer—it is inherent in the structure of 4-dimension spacetime." Nevertheless, I think this (E-C-S) is a breakthrough paper. Best, Tom RE: Bell's theorem - for or against Hidden Variables? - secur - 08-07-2016
Thomas Ray wrote: The authors claim that a particle does not exist in an 'up' or 'down' state until the experimenter chooses; I have a problem with that. You don't give the quote; they don't say exactly that. In normal Bell experiment, Alice and Bob choose detector orientation settings (the angle), then record the direction of spin (up or down). Those directions should show correlation (violate Bell's inequality, or similar). But this ECS gedanken records the spin first, before knowing the angle - that's the clever part. Then, the angles must show correlations, similar to the way directions do in the normal experiment. So normal experiment, following forward time direction, takes the directions as given and looks for correlations in the angles; ECS does the opposite, reversing (so to speak) the normal time direction. After going through the SGM the particle exists in a superposed state of the up and down channels, or beams, or waves. It "chooses" which state it's in only when the experimenter looks. He does so by sending a single photon through the 3 up channels, and another through the 3 down channels; whichever gets absorbed corresponds to the up/down spin state. That's the key trick. In this way we learn the direction without learning the angle. So the detector settings remain in a superposed "entangled" state, and should show correlations. The particle is up/down superposed until the photon detection trick is done. If that's what you have a problem with, this is the standard Copenhagen way of looking at it. Just interpret the statement according to your preferred ontology. For instance Gell-Mann might say the particle does have a definite spin but we haven't yet determined which decoherent history it, and we, have taken, yet. Or you might want to ascribe the uncertainty not to the particle, but to the experimenter: the particle has a definite spin we just don't know it yet. Since people describe the same ideas using different ontologies, I've gotten used to translating to my preferred ontology (which is, essentially, Copenhagen). It's easy enough once you get the hang of it. No, the "controversial" part here is that one would think the particle must go through one of the 3 SGM's in order to have a definite spin; so when you detect that spin you collapse the whole wave function (to use Copenhagen language) and interference among the 3 different orientations should be lost. As they show - convincingly, I think - that's not the case according to QM math, no matter what ontology you use. Of course it would be nice to actually do the experiment and show their conclusions are correct. So, by reversing the normal time sequence in which the experimenter gains knowledge (of detector angle and spin) they illustrate a situation where retrocausality seems, arguably, more intuitive. I think it's pretty clever. Although of course it can still be explained with other ontologies, it may make them less attractive. In particular, as they show, Bohmian mechanics has to stretch a bit to accommodate this scenario. My questions about ECS paper are: ECS: "If, at the quantum realm, causal effects proceed on both time directions, then sufficiently delicate experiments should be able to reveal this dual nature. Indeed TSVF already boasts some verifications of this kind [8], and further surprising theoretical and empirical results can be expected." I found [8] and similar work by these authors and see no experiments that actually "reveal" TSVF. They're all similar to this one: retrocausality is emphasized, or suggested; but other interpretations still work. Copenhagen, in particular, handles these situations easily as far as I can see. So their implication of "proving" TSVF is not justified. I'd be happy to hear any arguments supporting this assertion (or perhaps "hint" is a better word). ECS: "We conclude with a brief comparison between these interpretations and their traditional alternatives, Copenhagen, Bohmian mechanics and the Many Worlds Interpretation." They analyze only Bohmian, no others. They refer to [25] for Copenhagen - but it doesn't exist yet!: ECS: [25] A. C. Elitzur, E. Cohen, Why treat a disease with a no-better remedy? Copenhagen, Bohmian mechanics, and time-symmetric interpretations of QM. Forthcoming. Sometimes authors do this sort of thing hoping the reader simply won't notice but I don't suspect them of that. I look forward to the forthcoming reference [25] because I don't see why these TSVF-type experiments should be a big problem for Copenhagen. Finally, the idea of entropy constituting the "hidden variable" is reasonable on the face of it, but I don't see how that would actually work. I'll take a look at your site for further information. RE: Bell's theorem - for or against Hidden Variables? - Thomas Ray - 08-08-2016
(08-07-2016, 11:32 PM)secur Wrote: Thomas Ray wrote: The authors claim that a particle does not exist in an 'up' or 'down' state until the experimenter chooses; I have a problem with that. secur wrote, "So normal experiment, following forward time direction, takes the directions as given and looks for correlations in the angles; ECS does the opposite, reversing (so to speak) the normal time direction." This assumes there is a normal time direction. I'm saying that one cannot distinguish directions (past & future). From the ECS abstract: "We propose a new setting where the question is reversed: "What is the orientation along which this particle has this spin value?" It turns out that the orientation is similarly subject to nonlocal effects. To enable the reversal, each particle's interaction with a beam-splitter at t1 leaves its spin orientation superposed. Then at t2, the experimenter selects an "up" or "down" spin value for this yet-undefined orientation. Only after the two particles undergo this procedure, the two measurements are completed, each particle having its spin value along a definite orientation. By Bell's theorem, it is now the "choice" of orientation that must be nonlocally transmitted between the particles upon completing the measurement. This choice, however, has preceded the experimenter's selection. This seems to lend support for the time-symmetric interpretations of QM, where retrocausality plays a significant role." This is still a linear solution. In all linear solutions, the assumption of superposition holds -- SGM rectifies the trajectory to a forward direction; this does not imply that the trajectory not measured is retrograde, and vice versa. Which raises the question of what a nonlinear solution would look like. It would look like Joy Christian's unfairly-maligned exploding ball experiment -- no rectifying mechanism involved. This now makes it possible to judge the particle orientation independent of the trajectory -- that is, " ... without in any way disturbing a system, we can predict with certainty (i.e., with probability equal to unity) the value of a physical quantity, then there exists an element of physical reality corresponding to this physical quantity." (~EPR) True retrocausality is manifestly local, as ECS question, "Along which Orientation does the Particle have this Spin Value?” Thing is, we don't know the orientation (past or future, or if you prefer, left or right) if we get to manufacture it. It's all hidden in the production of entropy -- we conventionally assume that entropy flow is in the direction of 'disorder' yet without locally ordered states this is meaningless. So Joy's claim that his measurement framework accounts for all correlations -- not just quantum -- is correct. ECS are making a run at weakening efforts of 80 years to eliminate the role of continuous spacetime from quantum theory. It only brings along a plethora of ad hoc assumptions. You've said so much of value, secur, and I've replied to only one point. But this is getting too long. Perhaps in the course of discussion, we can revisit these important points? All best, Tom RE: Bell's theorem - for or against Hidden Variables? - secur - 08-18-2016
Joy Christian's "Macroscopic Observability of Spinorial Sign Changes under 2pi Rotations" gives a clear exposition of his idea. It's better than the previous paper we discussed, with similar concepts, which directly addressed Bell. I'm not surprised it passed peer review. But the classical experiment he proposes is, I'm rather sure, going to have a classical result. The linear correlation of SO(3) will be demonstrated, not the cos correlation of spin space. Nothing in the paper gives any reason to think otherwise. The "exploding ball" is, of course, basically a macro replica of a typical Bell experiment. This sort of thing has been thoroughly studied for more than 100 years. They would have noticed long ago if SO(3) were not the correct space. Richard Gill, http://arxiv.org/pdf/1412.2677v3.pdf, says there are mistakes in the paper, but I can't see it. Probably I'm looking at a different version (v4) than the one his comment addresses. Even if there are mistakes I'm sure they're fixable. Christian has done a good job describing the situation, but it's a mystery to me why he thinks the correlation calculations would give a non-classical result (like QM spin). Richard Gill is probably right that most experimenters believe the result would be classical, that's why they're not interested in doing it. A suggestion: Christian could probably come up with a very simple, cheap, table-top experiment that accomplishes the same thing, if he thought about it. That would stand a better chance of being performed. Presumably this relates to your idea of non-linear time, and entropy as a "hidden variable"? Christian's paper doesn't put it in those terms, but I can believe there's a connection - although I don't, at the moment, get it. Sorry I took so long to respond. The following thought occurred to me. Suppose the experiment were done, and it proved Christian was absolutely right. The establishment would still ignore it! Those people's minds are utterly closed. So if we're doing this for our own edification and enjoyment, that's great, it's a fine hobby. But if we actually think it will make a difference we're wasting our time. The same applies to any "alternative" physics, such as Schmelzer's papers. We're making a key mistake here. We're thinking it's about "truth" - no, it's about funding and money. This discouraging realization has led me to do other things - like, enjoy the beautiful weather - for the last week. I wish I could be more positive. Let me reiterate, I think it's a good paper overall, with a clear discussion of the difference between SO(3) and SU(2) correlations. Talk to you later! RE: Bell's theorem - for or against Hidden Variables? - Thomas Ray - 08-18-2016
(08-18-2016, 03:06 AM)secur Wrote: Joy Christian's "Macroscopic Observability of Spinorial Sign Changes under 2pi Rotations" gives a clear exposition of his idea. It's better than the previous paper we discussed, with similar concepts, which directly addressed Bell. I'm not surprised it passed peer review. Hi secur, Truth has no place in science. Whatever facts support a theory consistently, are true subject to falsification (Popper). So when Popper appropriated Tarski's correspondence theory of truth for scientific method, he left this condition open. A theory is only as true as it can be. It needs to be understood that Joy has presented a measurement framework--not a theory--that supports spacetime as a real phenomenon. Anyway, Gill's and Weatherall's "not even wrong" criticism is a straw man. My draft attempt to correct it is attached. Let's continue this delightful conversation. All best, Tom RE: Bell's theorem - for or against Hidden Variables? - gill1109 - 08-22-2016
(08-18-2016, 03:06 AM)secur Wrote: Richard Gill is probably right that most experimenters believe the result would be classical, that's why they're not interested in doing it.That's not what I say in my paper. I say that any *smart* experimenter can see that the result *certainly* will be classical. Christian explains how the experimenter is to measure the local hidden variable lambda and then calculate the spin (+/- 1) in all directions of interest. In CHSH terms: for each particle pair, there will be outcomes of simultaneous measurements in directions a and a' for Alice, and in directions b and b' for Bob. All correlations will be calculated from measurements on the same N particle pairs. The CHSH inequality will hold as a matter of simple algebra. RE: Bell's theorem - for or against Hidden Variables? - secur - 08-22-2016
@Gill1109, If you feel I misrepresented your words, I apologize - I hate it when people do that to me. But really I think my paraphrase is close enough. The intention was just to tone it down a bit, remove the rough edge which could be considered insulting. It's not really inaccurate to call Christian's calculated variable "spin" but the word carries the connotation of QM spin. Perhaps "direction of angular velocity" or some such phrase would be better. Calling it "spin" helps him in his objective of conflating the two cases, classical and QM. That's exactly the confusion that needs straightening out here. Of course I expect CHSH inequality will hold in this classical case. However your demonstration of that is flawed it seems to me. I'm looking at http://arxiv.org/pdf/1412.2677v3.pdf, the paragraphs beginning with "Now pick any two pairs of directions a1, a2 and b1, b2." You consider the script-E terms arising from every combination of these: E(a1, b1) - E(a1, b2) - E(a2, b1) - E(a2, b2) (Adapting the notation to avoid hassling with LaTex) A little algebra shows this must be +-2, so when all such terms are averaged, it can't get up to 2*sqrt(2). Christian expects, apparently, it should follow the SU(2) cos correlation not the SO(3) linear correlation function. As I said, that's the only part of his paper that makes no sense; obviously the classical SO(3) correlation should hold. Anyway, you're assuming that all four combinations of the a's and b's will occur. That happens in a typical Bell experiment, allowing CHSH inequality to be derived easily. However in Christian's case he instructs us to pick only one (a, b) then use it with each of the experimental data points to produce the script-E terms. So you won't get these four terms to combine, because it's not symmetrical like a typical Bell experiment. This explanation is not very clear but I hope you see my point. Of course it doesn't invalidate your main conclusion but you can't use this straightforward CHSH proof, for this case. @Thomas Ray, Reading your "restaurant" paper, and considering the obvious fact that you're a reasonable intelligent person, I can come to only one conclusion. This is a parody! You're cleverly making fun of some of these papers that seem, sentence by sentence, to make sense but taken as a whole, don't. That's true not only of some "against-the-mainstream" papers, but also distressingly many "real" physics papers from people like David Deutsch, Sean Carroll and others of that ilk. They seem to make sense at first, but actually don't. IF that's your object - to send up this whole gang of (*** insulting term censored ***)'s - congratulations! Very amusing. However if you're actually trying to make sense ... well, I just don't get it. Thomas Ray wrote: "Truth has no place in science. Whatever facts support a theory consistently, are true subject to falsification (Popper). So when Popper appropriated Tarski's correspondence theory of truth for scientific method, he left this condition open. A theory is only as true as it can be. It needs to be understood that Joy has presented a measurement framework--not a theory--that supports spacetime as a real phenomenon. Anyway, Gill's and Weatherall's 'not even wrong' criticism is a straw man. My draft attempt to correct it is attached." "Truth has no place in science": properly understood, I like this very much! Yes, Joy has presented a "measurement framework" which would support his point very nicely, if the experiment agreed with his conclusion. As mentioned, I doubt very much it would. Apart from that it's a pretty good paper. You don't need to correct anyone's criticism, or explain Christian - to me, at least. AFAIK everything makes sense. Everyone's made some minor mistakes, no big deal. Our only real disagreement concerns the results of the exploding ball experiment. With luck someone will perform it and put the issue to rest. RE: Bell's theorem - for or against Hidden Variables? - Thomas Ray - 08-23-2016
secur wrote, "Our only real disagreement concerns the results of the exploding ball experiment. With luck someone will perform it and put the issue to rest." A non-arbitrary initial condition, without entanglement, will put the issue to rest. That is the point. Understand that, and everything else falls into place. RE: Bell's theorem - for or against Hidden Variables? - gill1109 - 08-23-2016
(08-22-2016, 09:29 PM)secur Wrote: Anyway, you're assuming that all four combinations of the a's and b's will occur. That happens in a typical Bell experiment, allowing CHSH inequality to be derived easily. However in Christian's case he instructs us to pick only one (a, b) then use it with each of the experimental data points to produce the script-E terms.In his experiment, you observe N particle pairs, get N values of lambda, and then compute any script-E term for any (a, b) you like. For every (a, b): the same N particle pairs, the same N values of lambda. |