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

Coexistence of QT and Relativity

  1. Jul 13, 2008 #1
    Since QT violates SR is it safe to say one of the former will be disregarded / revised in the future? See EPR paradox.
  2. jcsd
  3. Jul 13, 2008 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    EPR doesn't violate SR since the wavefunction isn't an observable.
  4. Jul 13, 2008 #3
    Huh? I never said EPR violated SR. EPR isn't a theory its a paradox.....
  5. Jul 13, 2008 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Huh? Then what in quantum mechanics do you claim is violating SR? And what were you trying to say by mentioning EPR immediately after saying that "QT violates SR"?
  6. Jul 13, 2008 #5


    User Avatar

    Staff: Mentor

    Quantum electrodynamics (QED) is fully (special-)relativistic. Its predictions so far agree with experiment to very high precision.
  7. Jul 13, 2008 #6
    Then why did you mention EPR?

    Also, it is unclear what you mean when you say that QT violates special relativity. You need to be much more specific about what you have in mind.
  8. Jul 13, 2008 #7
    Well Bell would say that it isn't relativistic in a 'serious' sense, since standard QED is not a beable theory. Beable formulations of QED exist which are not Lorentz invariant on the level of individual beables, but only on the level of statistical ensembles in measurement interactions.
  9. Jul 13, 2008 #8
    Sorry I'll explain further. Take two particles A and B. The spin of A is unknown until we measure it. When we do measure the spin of A, B's spin wave function then instantaneously collapes to the opposite state. This transaction occurs faster than the speed of light, violating SR.
  10. Jul 13, 2008 #9
    You still haven't explained exactly what about that transaction violates SR. I assume you think it is the fact that the "collapse" occurs faster than light. But I'll just say that in the way you formulated the problem, there is no experimentally observable violation of SR since SR only prohibits superluminal matter and energy transport. The ethereal notion of "collapse of the wavefunction" as you described it does not fall into either category. That doesn't mean there isn't a violation of SR lurking in the EPRB correlations, but you first have to be more specific about what formulation of QM you are using to describe EPRB. That means you have to also state whether you regard the wavefunction as ontological or epistemological, and what proposed solution to the measurement problem you are using.
  11. Jul 13, 2008 #10
    So you're saying SR allows particles to travel faster and communicate at speeds that exceed c? Last time I checked this is a violation of SR so before answering your questions maybe you could elaborate and explain.
  12. Jul 13, 2008 #11
    In the way you described the problem, no particle is travelling faster than light because the wavefunction collapsing is not a particle, by definition. You still have not provided your interpretation of the wavefunction, i.e. whether it is ontological or epistemological, nor what formulation of QM you are using to analyze EPRB. If you analyze EPRB in de Broglie-Bohm pilot wave theory, or stochastic mechanics, or GRW spontaneous collapse, or two-time quantum mechanics, or textbook QM, they give different answers to your question. So please choose your framework first.
  13. Jul 13, 2008 #12
    Using textbook QM, Bell's inequality results in non local interactions taking place between particles. Do these non local interactions, which take place at speeds greater than c and in which distance does not effect interaction speed, have no consequence on what SR tells us?
  14. Jul 14, 2008 #13
    As textbook QM is a statistical formalism, there is no contradiction with Lorentz invariance in the treatment of EPRB on that level. There may however be a conflict with locality (the property that signals or causal influences cannot propagate faster than c in a vacuum between two spatially separated events). It depends on what physical assumptions of Bell's theorem one chooses to give up; locality or causality (the assumption that the two particles are physically independent of their future detector settings). If one gives up locality, and instead postulates nonlocal causal influences (and therefore correlations) between space-like separated particles, then indeed there is a conflict with special relativity only in the sense that superluminal causal influences are not a property of the local Minkowski spacetime structure in special relativity.

    But textbook QM, being only a statistical formalism, can only describe this nonlocality in terms of correlations, since it does not describe what the underlying physical causes those nonlocal entanglement correlations are. To look for explanations of the physical causes of the nonlocal entanglement correlations, and therefore a clearer and deeper understanding of the conflict between SR and nonlocality in QM, you have to analyze EPRB in terms of the beable formulations of quantum theory that I mentioned earlier. It should also emphasize that the textbook QM postulate of instantaneous "collapse of the wavefunction" is different from the nonlocality in the entanglement correlations, since the nonlocal entanglement correlations between the two particles can be expressed without any mention of the quantum mechanics of entangled wavefunctions. For the best expositions of this perspective, I highly recommend the following books:

    "Speakable and Unspeakable in Quantum Mechanics"
    J.S. Bell

    "Quantum Nonlocality and Relativity"
    Tim Maudlin

    If you choose to keep locality, then you must give up causality; but then you can save special relativity and even make its Minkowski spacetime structure fully compatible with the wavefunction formalism of quantum mechanics - in terms of textbook QM, this has been done with the two-time interpretation of QM developed by Aharonov, Bergemann, Lebowitz, and Goldhaber. In terms of a beable formulation of QM such as the pilot wave theory, this has also been done by Rod Sutherland in what is called a "Causally Symmetric Bohm Model". Both theories are fully consistent with SR and there is no need for irreducibly nonlocal causal influences or correlations.

    The Two-State Vector Formalism of Qauntum Mechanics: an Updated Review
    Authors: Yakir Aharonov, Lev Vaidman

    Causally Symmetric Bohm Model
    Authors: Rod Sutherland

    I also recommend the following book by Huw Price on this perspective:

    "Time's Arrow and Archimedes Point"
    Huw Price
    Last edited: Jul 14, 2008
  15. Jul 14, 2008 #14
    Thanks Maaneli. Your answer makes sense and I look forward to going through the attached suggested reading material you presented. One last question.

    So based on textbook QM (the others I admit I have to become more familiar with), because it cannot describe the underlying physical causes of these nonlocal entanglement correlations, we can safely say a conflict does exist between textbook QM (with its nonlocality) and SR (with its constant in all frames speed of c)?
  16. Jul 14, 2008 #15
    The fact that textbook QM cannot describe the underlying causes EPRB is not why it is in conflict with SR, since the conflict with SR becomes even more apparent when one constructs a beable description of EPRB (with the assumption of nonlocality held in tact).

    Rather, "[even though] it cannot describe the underlying physical causes of these nonlocal entanglement correlations, we can safely say a conflict does exist between textbook QM (with its nonlocality) and SR (with its constant in all frames speed of c)". Recall it is the fact that textbook QM gives up locality in favor of nonlocality in entanglement correlations. The conflict with SR is in the implication of superluminal causal influences, whereas SR only permits local causal influences in its Minkowski spacetime structure.
  17. Jul 14, 2008 #16
    Nothing about QM inherently violates SR. Depending on certain interpretations of the EPR paradox, the results are difficult to explain interpratively without resort to FTL influence, but FTL itself does not violate SR (tachyons for example can exist within SR's framework). FTL information exchange does violate SR but, as pointed out, is not possible in any interpretation of QM. Moreover, if realism is disregarded (see, e.g., MWI), there is no FTL problem at all.

    If an FTL influence were ever unambiguously discovered, it would be a breakthrough for physics, but both QM and SR would likely survive unscathed.

    It is different, but actually, and interestingly, I think, one of Einstein's very first challenges to Bohr in their famous exchanges was something along the lines of: "You say the wavefunction evolves and expands out to give a distinct probability of detection at every point on the detector. How does each point on the detector "know" instantaneously that the particle has finally been detected in only one space?

    The argument was initially dismissed as silly, but I believe Einstein inherently grasped the fundamental problem in that first question, which was how do we explain instantaneous action, whether it be a wavefunction collapsing like a bubble or entangled particles, without violating locality (the existence and reality of space and localized influence is fundamental to both SR and GR, even moreso the latter). Bohr's ultimate solution was essentially to say that it doesn't matter, because the wavefunction is just a tool for predicting experiments, and not necessarily a physical entity, and hence was born the Copenhagen Interpretation.
    Last edited: Jul 14, 2008
  18. Jul 14, 2008 #17
    Since you are raising this, I'll take the opportunity to ask whether somebody has a comment on :
    Can Bell's Prescription for Physical Reality Be Considered Complete?
    Joy Christian is generally considered to know what he is talking about when it comes to Bell's theorem. Last year, he published :
    Disproof of Bell's Theorem by Clifford Algebra Valued Local Variables
    Pretty much everything is in the title, he clearly shaw that his model is fully compatible with QM, the problem is with his concept of Clifford-values observables. It's very convincing and very unreasonable at the same time. I have not found anything serious against it.
  19. Jul 14, 2008 #18


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

  20. Jul 14, 2008 #19
    Wow. How does stuff like that get missed in peer review?

    While I was never involved in academic journals when I was in engineering school, I was on the law review in law school, and we meticuloulsy scouered every square inch of a paper to make sure everything was accurate. And we were students, not even peers.
  21. Jul 14, 2008 #20
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook