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

Need Understandable Explanation Of Bell's Theorum

  1. Dec 19, 2005 #1


    User Avatar
    Science Advisor

    Every once in a while I try to stretch my feeble mind and pick up layman books on quantum physics as a change of pace from the many clinical/medical journals I have to constantly keep up with. I have a BS in chemistry and understand basic quantum mechanics but as much as I try, I cannot fully understand Bell's theorum...just when I think I've got it , it escapes my brain. A fellow cardiologist has tried to explain it to me (he used to be a physicist) but I don't quite understand his explanation. Would anyone take a gander at explaining it to a lowly non physicist? And I won't feel insulted if you baby the language a little.:biggrin:
    Last edited: Dec 19, 2005
  2. jcsd
  3. Dec 19, 2005 #2


    User Avatar
    Science Advisor
    Gold Member

    I would recommend my own humble page: Bell's Theorem: Overview with Lotsa Links.

    There is both text and plenty of links there, although some of the links are advanced. If you will get things started with a few specific questions, I will take a shot at an answer. :smile: And there are a few others here who will help too.
  4. Dec 19, 2005 #3


    User Avatar
    Science Advisor

    Since neither electron "knows" what its spin is until we measure it, then when we measure it, it settles on a spin. But this means that the other electron, no matter how far away it is, must now take on the opposite spin.
    This relationship is entanglement, no? I am trying to rephrase what I read of your excellent link. Once again, excuse my ignorance. This entanglement thing is what throws my brain into a tizzy (which is why I am not a quantum physicist.)
    Last edited: Dec 19, 2005
  5. Dec 19, 2005 #4


    User Avatar
    Gold Member

    If it didn't, you would either be a genius, or a liar.
  6. Dec 19, 2005 #5


    User Avatar
    Science Advisor

    So the theorum disproves "hidden variables" and causality? That probabilities is the complete explanation for particle universe? In other words, it only solidified and cemented quantum physics even against its onslaught by the greatest mind of our generation, Einstein?
    Last edited: Dec 19, 2005
  7. Dec 19, 2005 #6


    User Avatar
    Science Advisor
    Gold Member

    Einstein's attack was pretty good. But he didn't give quantum mechanics quite enough credit. Understandable really, I mean QM was only a few years old when EPR was written. And the ideas of QM seem so outlandish in many ways. But hey, so does relativity when you first think about it!

    Yes, the answer is that when one entangled particle takes on a particular spin, the other one immediately takes on an appropriate matching value (parallel or perpendicular, as the case may be).

    As to hidden variables and causality... the general thinking is that you sacrifice EITHER local causality (locality) OR hidden variables. It's a matter of choice. You will find proponents here who will argue one side or the other with a great deal of conviction. :smile:
  8. Dec 20, 2005 #7


    User Avatar
    Science Advisor
    Homework Helper

    The theorem actually also eliminates 'simple' probability-based theories because the assumptions that it makes about any local hidden state are so weak.

    Einstein isn't 'our' generation. He was born in the 19th century, and was well past his prime (as a physicist) by the time of the Manhatten project, 60 years ago.

    Bell's theorem does very little in terms of justifying quantum physics, since it's a mathematical proof. Really, if you want to look for things that motivate QM you're going to have to look at things like, Bohr predicting hydrogen energy states or the Stein Gerlach experiment. By the time that Einstein made his criticisms, QM was already well established, and had quite a number of strong experimental results.

    Really, that's why QM is so wierd. It's built on 'wierd' experimental results.
  9. Dec 20, 2005 #8
    Bell's Theorem is the answer to the question about the complementary of quantum theory, which mean: Can we to complete quantum mechanic by unknown variables named Hidden Variables (HV) or no? This question was firstly to ask by Einstein, Podolsky and Rosen (EPR) paper. Around this question there is many philosophy because J. Bell sense that the answer to this question must be experimental. I.e. he thinks that the experiment only can to answer to this question. I suppose that he think that it is possible to add HV in quantum mechanics. But the results of experiments on his calculation give another answer: HV is absent in quantum mechanic. Bell calculates the quantities S that later named Bell's observable. If the world is classical or Quantum Mechanic +HV = Classical Theory than S<=1. Or S<=1 named Bell's Inequalities. Experiments gives S=1.4. Than Bell's Inequalities is violet. It is mean that we don’t find HV. From this follow Bells’ Theorem: It is impossible to construct quantum mechanics with HV.
    But I think that it is not finish!!! This question will be discuss a long time!
    Peoples will be finding HV for understanding quantum theory by classical quantities. It is because world which we see is classical and everything what we see is classical too. Yes, the calculation is non classical but why we doesn't understand what its mean.
    Last edited: Dec 20, 2005
  10. Dec 21, 2005 #9
  11. Dec 22, 2005 #10


    User Avatar
    Science Advisor

    Thank you all for your help. :smile: I think I understand it a little better than I did before solicitting everybody's help!
  12. Nov 20, 2007 #11
    Question: Bells Theorum

    If Electron "A" moves clockwise and Electron "B" millions of light years away moves counterclockwise, what happens if Electron "A" ALSO begins to move from left to right in a pendulum-like motion while still spinning clockwise, would Electron "B" still spinning counterclockwise begin to instantly move in the opposite pendulum-like motion direction as well i.e., from right to left keeping time with Electron "A"?
  13. Nov 21, 2007 #12
    Bells Inequality

    I wrote the simplest (possible?) explanaition of Bell's here:
    http://www.ronsit.co.uk/weird_at_Heart.asp [Broken]

    but having just reads it I need to improve it - a bit
    of geek crept in.
    Last edited by a moderator: May 3, 2017
  14. Nov 26, 2007 #13


    User Avatar
    Science Advisor
    2018 Award

    Wavens, I think your oversimplified explanation of the delayed choice at
    http://www.ronsit.co.uk/SimpleDelayedChoiceExperiment.asp [Broken]
    is actually wrong.

    For a correct explanation see e.g.
    The point is that for any (of the two) choices, both slits are open all the time. The choice (that takes place after the photon passes the slits) refers to two different possible types of measurements.

    I would also like to explain where the solution of the delayed-choice paradox lies. The crucial sentence on the latter link above is:
    "Whatever the photon does, it presumably does it now when it passes through the slits."
    The point is that this sentence is wrong if read without the word "presumably". Namely, the photon does something important even after passing the slit.
    Last edited by a moderator: May 3, 2017
  15. Nov 26, 2007 #14
    Delayed Choice

    Thanks Demystifier, I will have a look at the DCE again with what you said in mind.
    Your explanation seems much better on first reading.
  16. Nov 26, 2007 #15


    User Avatar
    Science Advisor
    Homework Helper

    I always use a version I slightly adapted from the one presented in John Presskill's lecture notes (you can google it). Clear, relatively simple to understand intuitively, entertaining and it needn't involve any math. Great for parties.
  17. Dec 11, 2007 #16
    No, I don't think so
    The theorum disproves "hidden variables with locality", which Einstein prefer
    But Bohm's QM is not "hidden variables with locality", which I think is OK
    since you can't use EPR to send signals that are faster than the speed of light
    Last edited: Dec 11, 2007
  18. Dec 11, 2007 #17
    From special relativity we know that time stand still and distance is zero for light
  19. Dec 12, 2007 #18
    I looked at the prove and actually it does not exclude hidden variable theory. The prove assumes that the probability factorizes P(A,B)=P(A)P(B) which doesn't have to be true for a *dynamic* hidden variable theory. Then it seems that entanglement is actually non-existent. Hope someone knows what I mean. Anyone knows a prove which doesn't assume factorization?

    Answer to the initial question: Bell's theorem claims to prove that if you mess around with one particle, another distant might be influenced by it.
  20. Dec 12, 2007 #19
    Perhaps an alternative: don't think about particles

    Quite a few assumptions are needed to derive Bell inequalities, but broadly speaking Physicists agree that classical particle models they are willing to use satisfy those assumptions. A nice presentation of the assumptions needed is given by A.G.Valdenebro, "Assumptions Underlying Bell's Inequalities", in http://arxiv.org/abs/quant-ph/0208161v2" [Broken],
    which appears in journal form in European Journal of Physics.

    As mentioned in another Reply, if someone is willing to use a nonlocal dynamics, de Broblie-Bohm trajectories can be used to model experiments which violate Bell inequalties. Physicists generally don't want to. To me also, the mathematics of de Broglie-Bohm models is quite ugly. That is a criterion.

    If, instead of thinking in terms of particle properties, we think in terms of fields, there are possibilities for classical modeling that do not satisfy the assumptions needed to derive Bell inequalities, but that are mathematically not as ugly (whether they look attractive enough to Physicists has yet to be seen, however).

    Suppose we see something weird happen in daily life; we would assume that either something weird must have happened in the past, or else that some weird coming together of non-weird things must have happened now. Taking that commonsense approach, a classical model for the weird correlations that violate Bell inequalities in experiments can be constructed, if we're willing to put weird correlations into the past. This is known technically as the conspiracy loophole. It is rarely discussed, because classical particle models that go this way just don't look right to a Physicist. I agree with that assessment. There are well-known field theories that have nonlocal correlations in thermal equilibrium states, however, for which violation of the no-conspiracy assumption is natural, so Bell inequalities can't be derived for them. If you like, this puts the nonlocality into the initial conditions instead of into the dynamics. Again, yet to be seen whether Physicists will like this kind of model.

    My web-site, http://pantheon.yale.edu/~pwm22", has links to a number of journal published papers on this (and also to ArXiv-accessible versions).

    A further counterpoint to the conventional point of view, which roughly follows Arthur Fine's 1982 approach to the violation of Bell inequalities (which essentially points the finger at measurement incompatibility and contextuality rather than at locality), can be found at
    I hope this paper is relatively accessible.
    Last edited by a moderator: May 3, 2017
  21. Dec 12, 2007 #20
    So here's a layman's view of the case.

    At some time somebody in a lab presses a big red button and two entangled electrons are created moving in parallel but opposite direction.
    <--A B-->
    Now no matter which model you chose the particles have opposite spin. The difference is "when" they have opposite spin.
    QT says they only have spin when you measure one of them. Until you measure the spin, both electrons have a superposition of spin up and spin down.
    HV says they are born with a certain spin, so Ex A is created with spin up, and B with spin down.

    Measuring the spin of both particles along an arbitrary axis, there will be a correlation between the measurements. Ordering the measurements in a certain way, you get a value S, which in a HV model can at highest be 50%, but can be up to 75% in QT.
    Bell Theorem is the theory that states this mathematically. A sort of "if you can get more then 50%, you prove that QT is right and HV wrong"-statement.

    The problem as I see it with the measurements made on the experiment is that measurement errors increase the S-value. So it's more like saying if you get S higher then 50% you either proved hidden variable wrong, or your setup isn't precise enough. And since it's easy then to just take a crappy setup and say "I proved quantum mechanics!" you don't get hard prof that HV is wrong. Besides this, the famous measurements made actually removed a large amount of measurement.

    More educated people are welcome to correct me on this, I haven't look much into the matter.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook