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

A Fresh Look at EPR

  1. Jan 12, 2005 #1
    We have one big advantage over what Bohr and Einstein had in the 1935
    EPR debate. We can actually perform the experiment. Let's see what Nature decides.

    The Einstein point of view is that when the two photons are created, they both have a definite polarization that is negatively correlated with the other due to conservation of spin, but we do not know what they are. When one is measured, we then know the polarization of the other (it is the opposite polarization). Since both photons have a definite polarization from birth, there is no question of whether the measurement of one photon affects the polarization of the other. This is the core of Einstein's "element of reality" argument:
    "If, without in any way disturbing the 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."

    The QM point of view is that, after they are created, both photons exist in a state of superposition of all possible polarizations. Until measured, neither photon has a definite polarization. When one photon is measured, we now have the question of whether the polarization of the unmeasured photon is determined by the polarization of the measured photon. Here is Bohr's argument:
    "The criterion of physical reality proposed by EPR contains an ambiguity as regards the meaning of the expression "without in any way disturbing the system." Of course, there is in a case like that just considered no question of a mechanical disturbance of the system under investigation during the last critical stage of the measuring procedure. But even at this stage there is essentially the question of an influence on the very conditions which define the possible types of predictions regarding the future behavior of the system.
    Since these conditions constitute an inherent element of the description of any phenomena to which the term "physical reality" can be properly attached, we see that the argumentation of the mentioned authors does not justify their conclusions."

    What does Nature say? Bennet, Brassard and Ekert did the foundation work for a famous crypto-system that is now the basis for several commercial products. Bennet, Brassard and Ekert say this:
    "The EPR effect occurs when a spherically symmetric atom emits two photons in opposite directions toward two observers, Alice and Bob. The two photons are produced in an initial state of undefined polarization. But because of the symmetry of the initial state, the polarizations of the photons, when measured, must have opposite values, provided that the measurements are of the same type. For example, if Alice and Bob both measure rectilinear polarizations, they are each equally likely to record either a 0 (horizontal
    polarization) or a 1 (vertical), but if Alice obtains a 0, Bob will certainly obtain a 1, and vice versa."

    It appears that Nature agrees with Einstein.

    All the best
    John B.
     
    Last edited: Jan 12, 2005
  2. jcsd
  3. Jan 12, 2005 #2

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    I missed the part where you connected Einstein's view to an experiment. Did you use a quantum eraser to eliminate that from the post? :smile:
     
  4. Jan 13, 2005 #3

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    No, it is exactly the opposite !! This is what Einstein took as an example of an absurd prediction of QM, to illustrate how absurd QM was. And when you do the experiment, the "absurd prediction" is verified !
     
  5. Jan 13, 2005 #4
    Hmmm ... I'm afraid the problem here is that there are doubts regarding what really does happen in the experiments! When you use "entangled particles" in quantum key encryption, it is not, when you come to look into it, necessary that they obey QM. Ordinary correlation will do perfectly well so long as you have detectors able (when suitably orientated) to distinguish between vertical, horizontal and +-45 deg plane polarisation. No single orientation will enable you to actually find out what polarisation any particular signal has (this problem has always been with us: with polarising filters you need either a whole set of measurements or the use of crossed filters to find out the direction of polarisation of a beam). The QKD people cleverly get around this difficulty.

    But I fear that what they say about the results of actual experiments may be just an approximation to the truth! The experimental behaviour, when they use PDC sources, is good enough for their purposes. They have no interest in the kind of source Aspect used, and little, I fear, in with whether or not a valid Bell test is violated.

    With a source producing truly random polarisation directions, the statement that when you measure coincidences using parallel polarisers you always get perfect correlation is not quite true. For one thing, of course, most of the time when you get a count for one "photon" you get nothing registered for the other.

    No, as far as I know, what you actually get (with a genuinely "rotationally invariant" (RI) source) is consistent with Einstein's idea but you have to make proper allowance for the operating characteristics of your apparatus before this becomes clear. You have to allow for the fact that when a polarising cube is used maybe any losses are concentrated around those signals that are not almost parallel or orthogonal to the cube axis. In addition, you have to allow for the likelihood that when the initial signals are very weak, after passage through a polariser and consequent reduction in intensity by (approximately at least) cos^2 (angle) only those signals whose polarisation direction was within, say, 30 deg of one of the polariser axes have a significant chance of detection. [See http://en.wikipedia.org/wiki/Local_hidden_variable_theory] [Broken]

    But with PDC sources I have reason to believe that we very often get situations that are not rotationally invariant. The statistical properties of the outputs are quite different from the RI case. From the point of view of coincidence measurements, they are, in the extreme case, the same as sources in which the polarisation takes only one fixed direction. See:

    Thompson, C H, “Rotational invariance, phase relationships and the quantum entanglement illusion”, http://arxiv.org/abs/quant-ph/9912082

    Caroline
    http://freespace.virgin.net/ch.thompson1/
     
    Last edited by a moderator: May 1, 2017
  6. Jan 13, 2005 #5

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Ok, the claim was that this cryptothing clearly indicated that Einstein was right and QM wrong. That's a fully wrong statement.

    The experiments are fully in agreement with QM predictions (that means, the QM predictions of the experiment, including the QM prediction of the behaviour of all components in the experiment such as the source, the beam splitters, the polarizers, the detectors), and yes, there is still a possibility that one might, one day, find a theory, the way Einstein hoped, that also explains all of it, as the results do not rule that possibility out. In order to do so, one should need photodetetors which have a higher efficiency than 87%, and the current state of affairs is around 50%. As far as I know, no such theory is yet known, but you never know what might be invented in the future.

    Is that a correct statement of the current state of affairs, Caroline ?

    cheers,
    Patrick.
     
  7. Jan 13, 2005 #6
    Yes, it looks fair enough, but I believe I've read somewhere that the people doing the QKD work recognise themselves that works even if there is no entanglement. If we want to satisfy ourselves finally as to whether or not entanglement of separated particles is a fact of this universe, we should look at actual Bell test experiments and not rely on QKD.

    Caroline
     
  8. Jan 13, 2005 #7
    Hi Caroline

    Hi Caroline
    The EPR crypto-system is heavily time gated and coincidence checked, so they are definitely not dealing with raw photons. The commercial EPR crypto-systems behave exactly as Bennet, Brassard and Ekert say:
    "If Alice and Bob both measure rectilinear polarizations, they are each equally likely to record either a 0 (horizontal polarization) or a 1 (vertical), but if Alice obtains a 0, Bob will certainly obtain a 1, and vice versa."

    If it did not behave this way, it could not be a commercial product. If raw photons were being used, it just would not work. Between the time gating and coincidence detection, all of the wayward photons are culled out.

    The really important LR issue is that these commercial crypto-systems are quite deterministic and causal. The fact is that if I want to build a deterministic and causal quantum system, I can. The universe is NOT stochastic as von Neumann's "silly" proof would lead you to believe.

    Nice to see that you are still kicking and even having some fair successes - keep it up - LR *WILL* win out - All the best
    John B.
     
    Last edited: Jan 13, 2005
  9. Jan 13, 2005 #8
    You know, a great deal of this controversy could be eliminated by a higher quantum efficiency in detecting the photons, is that correct, Caroline?

    If so, then photodetectors are the wrong way to go. CCDs are currently being made with QE in excess of 90%. With CTE (charge transfer efficiency) of over 99.999%, and similar signal amplifier efficiency and accuracy, this indicates that it should be possible to achieve the level of measurement that would prove or disprove the case for Aspect once and for all. Do you agree?

    On edit: oops, forgot a http://www.calypso.org/~neill/cameras/hrcam/HRCamQE.html [Broken] to a private observatory's CCD specs. Note that this is not a government-funded observatory!
     
    Last edited by a moderator: May 1, 2017
  10. Jan 13, 2005 #9
    Einstein's Hypothesis

    Here is Einstein's Hypothesis:
    When the two photons are created, they both have a definite polarization that is negatively correlated with the other due to conservation of spin, but we do not know what they are. When one is measured, we then know the polarization of the other (it is the opposite polarization). Since both photons have a definite polarization from birth, there is no question of whether the measurement of one photon affects the polarization of the other.

    Does anyone have a different hypothesis?

    Regards
    John B.
     
  11. Jan 13, 2005 #10
    Proof by sale I like that :smile: lets hope the buyer gets what they pay for :biggrin: (ofcourse I understand they did, just couldnt resist)
     
  12. Jan 13, 2005 #11

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    If only. :smile: I think this is her mission in life.
     
  13. Jan 13, 2005 #12

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    I do. It is called QM. The description above is not compatible with what has been observed in the lab, and has nothing to do with cryptography.
     
  14. Jan 13, 2005 #13
    You've left out one very important piece of information: spins in separate planes are conjugate under uncertainty; if the spin in one plane is known, then the spin in other planes is indeterminate. Now, remember, it is not unmeasureable- it is undefined, i.e. it is an eigenstate of two eigenvalues, whose mixing angle is the angle of the planes.

    On edit: DrChinese got in in front of me a couple times. This was in response to JohnBarchak's question, "Does anyone have a different hypothesis?"
     
    Last edited: Jan 13, 2005
  15. Jan 13, 2005 #14
    The reason they get such neat answers is, I'm fairly certain, that their sources do not have rotational invariance, only "binary rotational invariance". Because they are interested in commercial applications and not pure theory, they have no incentive to look into this. They can happily set the detectors at selected angles and have no need to ask what might happen in between.

    Without seeing the actual experiments, I currently think it likely that there is just one stochastic element at work here: the system "chooses" randomly between two possible phase relationships. Of course, even this choice probably isn't truly random, but the factors causing the final decision are way beyond our reach. [See my new paper (not yet submitted for publication): http://freespace.virgin.net/ch.thompson1/Papers/Homodyne/homodyne.pdf ]

    Thanks! If you've got a moment, perhaps you could have a look in wikipedia? Certain people (including DrChinese) are of the opinion that the pages I put there last summer contravene wikipedia's Neutral Point of View policy. I can't really deny this yet I think they have every right to be there. See the discussion pages associated with:
    http://en.wikipedia.org/wiki/Clauser_and_Horne's_1974_Bell_test
    and
    http://en.wikipedia.org/wiki/Bell's_Theorem

    Caroline
    http://freespace.virgin.net/ch.thompson1/
     
    Last edited by a moderator: Apr 21, 2017
  16. Jan 13, 2005 #15
    As a matter of fact, I think one could get by quite happily with imperfect detectors if only the experimenters would read and take note of the message of Clauser and Horne's 1974 paper! The CH74 Bell inequality does not depend on the fair sampling assumption. See http://en.wikipedia.org/wiki/Clauser_and_Horne's_1974_Bell_test and various paper on my web site.

    You may well be right. I don't know all that much about CCDs. However, an alternative way of getting an "outcome" every time is to re-define what you are going to consider to be the outcome! This is, I think, done in a perfectly legitimate manner (I can't see that any of the necessary assumptions are violated) in the Sanchez et al proposed "loophole-free" Bell test. [Another chance to plug my new paper, which discusses this proposal! See http://freespace.virgin.net/ch.thompson1/Papers/Homodyne/homodyne.pdf ]

    BTW, DrChinese, you are quite right! I shall have difficulty knowing what to do with myself if I ever succeed in my mission. :smile:

    Caroline
     
    Last edited by a moderator: Apr 21, 2017
  17. Jan 13, 2005 #16
    Hi Caroline
    Show me anything in Wikipedia that is "neutral." If Bohr's views are in Wikipedia, then your views are totally appropriate. Your work conforms to real science far more than any of Bohr's work. Bohr did not believe in the scientific method - you do!
    All the best
    John B.
     
  18. Jan 14, 2005 #17
    Help me understand EPR

    I’d just like to understand the EPR debate. Allow me to diagram a test as I understand the debate.

    Our observers will still be Alice and Bob each in one of two houses on opposite sides of a resort. Carol and Ted are leaving the resort to visit both houses separately, and have agreed that only answer Yes or No questions about what happened at the resort and one of them shall always answer “Yes” the other always “NO”. Now during there visits it happens that Ted goes to Bob’s house while Carol arrives at Alice’s. Bob gets YES answers and later meets up Alice to learn that Carol gave NO answers to the same questions!! The resulting conflict clearly a part of Ted and Carol’s design.

    Einstein View :

    Carol and Ted simple agree in advance as they departed the resort who would say YES and who would say NO.

    QM View:

    Carol and Ted only agreed to answer the opposite from the other's. But would pick Yes or NO based on odd or even on the nearest clock when first is asked. But - When the first question was asked the other new how the respond without checking for Odd or Even based on some unknown FTL link between them. Note: At the time they are asked their first question the houses are Space-Like separated.

    Now I can see where EPR tests have shown that they answer as predicted. But I do not see how the tests are insuring the answers could have been different if questioned differently in order to show that the QM view is better.

    If you make this clear I’d appreciate it.
     
    Last edited: Jan 15, 2005
  19. Jan 14, 2005 #18

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    I would humbly recommend my first posts in this thread: Bell's THeorem and Negative Probabilities

    You will see that the scenario you describe last above has 8 permuatations per the Einstein view, 2 of which must be negative to agree with the predictions of QM. QM says there are only 4 permutations and all are non-negative.
     
  20. Jan 14, 2005 #19
    Thanks for the encouragement, John! I do hope DrChinese agrees.

    Caroline
     
  21. Jan 14, 2005 #20
    But nobody "understands" how entanglement works! If you ask me (which I know you didn't!) the reason is that entanglement simply does not happen. Einstein et al were right about the existence of hidden variables (though slightly wrong in that, in the actual Bell test experiments, the hidden variables don't completely determine the outcomes, only determine their probabilities). Bell's Theorem is also correct, so something has to be wrong and this, in my book, is quantum theory.

    Yes, I know the experiments are supposed to show that entanglement does happen, but once you understand the various "loopholes", the way is open for hidden variable explanations, based on very ordinary classical physics.

    See my wikipedia pages, e.g. http://en.wikipedia.org/wiki/Bell_test_loopholes and http://en.wikipedia.org/wiki/Local_hidden_variable_theory.

    I'm afraid I haven't tried to follow your analogy in any detail. The answer, I suggest, is that you can't expect anyone to be able to make this clear since you are trying to explain the impossible. The real experiments are another matter. Notwithstanding anything you may have read, there is nothing weird or impossible about them.

    Caroline
     
    Last edited by a moderator: Apr 21, 2017
  22. Jan 15, 2005 #21
    DrChinese
    I appreciate your at least looking at my question in detail I just get a little lost in the permutations of statistical scenarios that even include cos and sin’s of most explainations.
    Thus I’m tring to make as simple an example as possible, in the hope I can analyze it correctly. And your insight on 8 vs. 4 permutations may be where I need the most help , as I do not think I can name them. So I will try to start defining them and allow you to correct me.

    Alice and Bob are interested only in the answer not who gives it.
    We must evaluate both answers.
    We must test from one Primary Observer.
    The Three variables that generate 8 permutations are:

    A) Primary Observer is Alice PO is Alice=A+ PO is Bob=A-
    B) Answer received by Primary Yes = B+ NO=B- (odds 50-50)
    C) Answer reported to PO by other Yes = C+ NO=C- (odds 50-50)

    [1] A+ B+ C+ Never 0%
    [2] A+ B+ C-
    [3] A+ B- C+
    [4] A+ B- C- Never 0%
    [5] A- B+ C+ Never 0%
    [6] A- B+ C-
    [7] A- B- C+
    [8] A- B- C- Never 0%

    From the Einstein View with as answers preset at departure 1,4,5,or 8 couldn’t occur.

    Clearly my 8 permutations do not agree with yours as my results are nothing like the one you gave. So before I try to define the QM permutations can you detail what the 8 you referred to are.

    Also, when you say “2 must be negative” I’m assuming you mean “NOT” or never as in Zero probability. Instead of a “Negative Probability” as I think this example is too simple for any of those to appear - or do they?

    And if you would detail the 4 permutations for QM . I having a little trouble understanding how we can define the test such that the number of outcome permutations can be different based on QM or not QM.

    I don’t expect this analogy to resolve the ‘correct’ view of the test. Just do a better job of describing EPR in the simplest possible analogy to understand the two conflicting views QM and Non-QM and maybe the intent of the test being done.

    Thanks RB
     
  23. Jan 15, 2005 #22

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    Randall,

    I will take us through step by step. You don't need the sin/cos stuff to see the paradox. For sake of our discussion, you can ignore Caroline because she is trying to distract you. Let me know after which steps seem confusing to you and which steps seem OK. We can then focus on that step in more detail.

    1. You can see the 4 basic permutations easily with Alice (A) and Bob (B):

    A+ B+
    A+ B-
    A- B+
    A- B-

    These are agree upon by all, including QM.

    2. Suppose you believe that the measurements of Alice and Bob were not particularly fundamental to the outcome. Then you could have also had observers Chris (C) and Dan (D) and as many more as you like, and they could have made measurements and not affected the results of Alice and Bob.

    This is called the local realistic or hidden variable interpretation, and it is immenently reasonable. Follow so far?

    3. It is reasonable - in fact so reasonable that it took 30 years for Bell to see the mistake here.

    So let's forget Dan and the others, and focus on Chris. Chris makes - or could have made, according to 2 above, observations on one of the photons. I mean after all, the particular angles measured by Alice and Bob shouldn't have been the only angles testable should they? So what IF Chris could have made a measurement at some other angle? Then he would see a + or a -. If you THEN combine his results with the possible results of Alice and Bob, you get the following 8 permutations:

    [1] A+ B+ C+
    [2] A+ B+ C-
    [3] A+ B- C+
    [4] A+ B- C-
    [5] A- B+ C+
    [6] A- B+ C-
    [7] A- B- C+
    [8] A- B- C-

    So hopefully you see that all of the above have a chance of occurring between 0 and 100% and that they are the only permutations possible for Alice, Bob and Chris. Therefore, they total to 100%. Do you see this step?

    4. We have no know way to measure all 3 (A, B and C) simultanueously and directly by experiment. So we measure it indirectly, two at a time.

    Alice and Bob - we get correlations and non-correlations, each as a % adding to 100%

    Alice and Chris - we get correlations and non-correlations, each as a % adding to 100%

    Bob and Chris - we get correlations and non-correlations, each as a % adding to 100%

    We measure the A & C correlated cases, the A & B non-correlated cases and subtract the B & C correlated cases. When we do this, we end up with cases (2*[3] and 2*[6]).

    If it helps, let me know and I will re-explain how this adds up. This is the step that involves the adding and subtracting of a bunch of cases, but it does work out. I want you to be confortable with this.

    5. But you would definitely think that the lowest value you could get for (2*[3] and 2*[6]) would be zero. Do you see this?

    6. We arrived at 5 without any reference to QM whatsoever - QM does even need to exist as a theory to get this last result. Do you see this too?

    7. Now we simply conduct an actual experiment. Anyone can do it, anywhere, anytime although it is not something you can do at home. But the experiments have been run many times over the past 30 years. The results are always the same. The value you get for cases (2*[3] and 2*[6]) is always less than zero. The results are indisputably at variance with the reasonable assumption that Alice and Bob are not fundamental to the results - which we made in step 2 above. Therefore, our assumption was invalid. Therefore, LR fails.

    Conclusions

    8. As you can see this has nothing whatsoever to do with QM. On the other hand, in QM there is this thing called the Heisenberg Uncertainty Relations. It states that observations DO affect the results. Therefore it does not acknowledge our assumption 2. above as valid and specifically disclaims it. As a result, it has no paradox.

    9. As icing on the cake: the observed values in the experiments above exactly match the predictions of QM to 30 standard deviations, but that doesn't change the paradox at all. Also, the predictions of QM also just happen to match classical optics (but without the fallacious LR assumption of 2.).

    Let me know the step numbers you would like some additional clarification on.
     
  24. Jan 15, 2005 #23

    Nereid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Hi JohnB.

    I've not yet come across the concept of 'neutrality' wrt Wikipedia, or indeed physics. Would you care to let us know what you consider the key aspects of 'neutrality'?

    For myself, I am quite 'agnositic', and totally disloyal ... I can be 'bought' by good concrete, specific, testable predictions; by clear demonstrations of self-consistency and (more important) consistency with good experimental and observational results; by 'utility' (not in the economic sense; I mean can we use it to do stuff like make computers?); and by 'fecundity' - does it open the door to (a.k.a. give birth to) lots of interesting and wonderful ideas, which our telescopes and microscopes can help us tease apart?

    Call me a 'whore', call me a 'slut', I don't care ... just turn me on with the number of decimal places your theory matches the experimental results!
     
  25. Jan 16, 2005 #24
    Neutral point of view!

    For a statement of wikipedia policy on this, see http://en.wikipedia.org/wiki/Npov

    They seem effectively to take "neutral" to be synonymous with "dominant". I suggested that they could solve the problem in one of the pages currently under dispute (http://en.wikipedia.org/wiki/Bell's_Theorem, which I had, with considerabl labour, re-written) by having a short introduction followed by links to two alternative pages, one from the QM point of view, another from my local realist one. Perhaps there are other alternatives that others could put in as similar links. They did not think much of the idea! Their policy, it seems, does not allow "forks". Apparently http://www.wikinfo.org/ was set up in order to accomodate valuable pages rejected by wikipedia, and this does allow forks.

    I'm with you there, except that (a) I am unashamedly "local realist" and (b) am not too terribly bothered about producing general formulae that are useful or predict anything new. Instead, I'm content with general rules of operation, general approaches to the matter of finding of plausible explanations for actual experiments.
     
    Last edited by a moderator: Apr 21, 2017
  26. Jan 16, 2005 #25

    vanesch

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    There's a need here for a basic "course on the scientific method 101" :-)

    It is a good thing that you state, what I was trying to point out, so clearly yourself! It explains all the endless discussions we're having and the hopelessness to lead us somewhere in doing so. What you propose is not physical science, if you're not "terribly bothered about producing general formulae that are useful or predict anything new". THAT is namely the very purpose of science!

    Again, the purpose of science is not to be absolutely certain that there are no angels pushing the planets, but to reject theories which make WRONG predictions (and to encourage temporarily theories that make RIGHT and PRECISE predictions). The point is that a theory that predicts an outcome of 3.88230470+/-0.000001 and a corresponding measurement that gives you 3.88230488 has more chances to have some predictive value than a "theory" that predicts that the result must be somewhere between 0.1 and 502.8. This last theory has more value than one which says that the result must be 8.302913402349234 +/- 0.0000000001.

    The reason is that the first one "puts its money where its mouth is" :tongue:

    Really, if you do not understand this, then you have no understanding of what science is all about, according to the universally accepted definition of science, and explains also why you have such difficulties communicating with scientists. You're simply talking about something else (call it theology :-)

    cheers,
    Patrick.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook