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

Why is superdeterminism not the universally accepted explanation of nonlocality?

  1. Feb 21, 2012 #1
    from my thinking nonlocality and entanglement are never a problem because in a totally determinstic universe, the information about what is going to be instantaneously tranferred from a to b is already known to the universe. we may not be in block time but the universe acts as if it were. this is the first thing ive come across that agrees with my resolution of instantaneous info transfer.

    even tho i personally believe that entanglement is basicly a zero sum static, and it is essentually noneffectual on the universe, just something we have to live with, but does not violate relativity because the information does not have any effect on anything anywhere. why is this not mainstream? do most people want to live in an undetermined future, thinking its closer to free will?
  2. jcsd
  3. Feb 21, 2012 #2
    It is true that Bell's theorem as such does not rule out the possibility local superdeterministic hidden variable theory. But there are solid reasons why people do not rush to embrace a superdeterministic explanation.

    First of all, such an explanation would be in some sense "conspiratorial" - it would mean that all the particles in the universe are secretly working together, each particle acting in just the right way as to make it seem as if local realism is false even though it is really true. So for instance, if you decide which way to orient your polarizer based on throwing dice (or some more sophisticated pseudorandom number generator), the motion of the dice in principle can be predicted from the initial conditions of the dice and all the air molecules. So all the air molecules were working together to turn the die in just the right way so that the right number would show up. And the air molecules could be affected by the time a person sneezed, which could be affected by something they ate. So the laws of the universe made the person choose just the right food so as to make him sneeze at the right time to make the air molecules turn the die in just the right way so that the polarizer will be turned to the exact orientation the universe wants. You can see that this sounds less like a scientific theory and more like a religious explanation - you have a God who is controlling all the little details of the universe in order to make the statistics of quantum entanglement experiments come out just right.

    Second of all, there's a reason why the Bohmian interpretation, which is a nonlocal realist interpretation, gets respect and superdeterminism does not (even though both are arguably "conspiratorial" to varying degrees). In the case of Bohmian mechanics, we actually have a fully developed mathematical theory which predicts all the phenomena of (nonrelativistic) quantum mechanics. In the case of superdeterminism, however, we just have a vague hope of the possibility of a scientific theory. There are some people like Gerard t'Hooft who have tried to create a "proof of concept" superdeterministic theory, just to show that it is possible for a local realist theory to exploit the superdeterminism loophole to Bell's theorem. But until there's an explanation that actually shows exactly how the particles of the universe are conducting this grand "conspiracy", there will probably be continued skepticism from the scientific community.
  4. Feb 22, 2012 #3
    Maybe a bit simple way of putting, but the short answer to your question is that not everyone thinks non-locality needs an explanation! On the contrary, it can rather be considered as an answer. If we ask how can our entangled results be explained, then non-locality is a part of one answer to this.

    As for your interpretational arguments, it's hard to say much since you seemed to have formed your view on it already. My personal view on interpreting it all, is that I try to keep my beliefs as close to measured reality as possible. For example, we currently measure both random and non-local phenomena, and even though there may exist explanations to these that are deterministic, we still need to be able to explain the random, non-local measurement results in the end. This means that the possible deterministic background contribute with nothing, and may therefore just as well be omitted from our theory.

    To put it shortly, when left with nothing else to go on, it's better to "believe" that the universe is actually as we see it (random and non-local) rather than to invent longer explanations.
  5. Feb 22, 2012 #4
    If you were a pure pragmatist, then wouldn't you assume that if you measured a particle to have a certain property, that it had that property before you measured it? Like if you see a dollar bill lying on the road, don't you assume that it was lying there before you saw it?

    Now that kind of naive noncontextual realism, although as a pragmatist you may be naturally led to it, runs into some difficulties if you apply it to the spin observable, because the Kochen-Specker theorem states roughly that any theory of spin that reproduces the predictions of quantum mechanics must be contextual, i.e. dependent on how you measured it. (Of course, if you're a superdeterminist that doesn't mean much to you, because you don't believe there is any free choice of how you choose to measure, so you would think a Kochen-Specker tests are a fraud for the same reason you would think Bell tests are a fraud.) I wonder, is there an analogous result like Kochen-Specker for the position observable? I suspect it would be problematic for Bohmian mechanics if there was, because part of the reason Bohmians think spin is "fake" is because of the Kochen-Specker theorem. Would they similarly think position is "fake", or is position too foundational for Bohmian mechanics?
  6. Feb 23, 2012 #5
    i have no idea what makes you think that superdeterminism is conspiratorial. i dont understand why superdeterminism is not the fundamental pillar that all science is built on. your dice analogy only seems to prove my point.

    but when you say your view of superdeterminism as "the laws of the universe made the person choose just the right food so as to make him sneeze at the right time to make the air molecules turn the die in just the right way so that the polarizer will be turned to the exact orientation the universe wants. You can see that this sounds less like a scientific theory and more like a religious explanation - you have a God who is controlling all the little details of the universe in order to make the statistics of quantum entanglement experiments come out just right.", you just seem to be incapable of accepting the possibility that the future is a 100 percent consequence of the events preceding it.

    do you believe in free will or something?
  7. Feb 23, 2012 #6


    User Avatar
    Science Advisor

    In principle, I don't think superdeterminism is less valid than standard interpretations. One has to realize, that the freedom to chose initial conditions ("free will") is an axiom in QM.

    The problem is that if you drop this axiom, I can't think of a way how to apply the scientific method in a meaningful way. Since the mainstream is doing science, I don't think he will ever embrace such an interpretation.

    So philosophically, I think, it stands on equal footing with standard interpretations. But for actually doing science, one has to assume at least some kind of "effective" freedom in choosing initial conditions.
    Last edited: Feb 23, 2012
  8. Feb 23, 2012 #7
    also, since entanglement must always originate locally, the info transfer between entangled particles is never going to be faster than c, therefore nonlocality is not an issue in superdeterminism and neither are hidden variables if im correct. instantaneous info transfer will not violate relativity because of this.
  9. Feb 23, 2012 #8
    or information theory for that matter
  10. Feb 23, 2012 #9


    User Avatar
    Science Advisor

    Jadrian, I think that you have misunderstood what is meant by "superdeterminism" in the context of avoiding nonlocality of QM. In this context, superdeterminism is NOT merely the idea that everything is deterministic, i.e., that future is completely determined by the past. If superdeterminism was only that, then it could not avoid nonlocality. Instead, superdeterminism is much more. It is the idea that
    1. Future is completely determined by the past.
    2. The past (i.e., initial conditions) is not arbitrary, but is fine tuned so that in the future we see correlations between distant object which never mutually interacted.

    Superdeterminism is not popular due to the property 2 (not 1). It is this second property (not the first) which makes it conspiratorial.
  11. Feb 23, 2012 #10
    It's important to make a distinction between a measurement result and what is "really" there. Just because a measured result is clearly contextual does not mean we can exclude realism!

    This is because your measurement apparatus changes the "real" state. To follow up on your dollar bill analogy, if you used a paper shredder as a measurement tool for detecting the bill and watch the output of it, then you could state that there must have been some form of dollar bill before but you don't necessarily see it as "really" was when it was lying on the road. But that does not mean it wasn't "really" there, it only means you have limited your range of answers to how you ask questions/what measurement you do. i.e. you measurements are contextual, but realism might still be valid.

    Similarly, in quantum mechanics, your output answers are often very limited. For example, measuring a superposition state in the computational basis (0 and 1), you can only get 0 or 1 as answer, and even measuring multiple times will not allow you to ascertain the "real" underlying state. Nevertheless, a superposition state can be said to be a "real" state, because it is even possible to draw a figure of the electron distribution of that particular superposition state! (or whatever system you used). That electron distribution will be different from both the 0 and the 1 state's distributions, and smarter measuring devices can allow you to determine it more precisely (don't shred you dollar).
  12. Feb 23, 2012 #11


    User Avatar
    Science Advisor

    There is one big difference between spin and position. Spin consists of 3 observables (corresponding to 3 directions in space) which do NOT commute with each other, while position consists of 3 observables which DO commute with each other. The Kochen-Specker theorem applies to any observables which do not commute with each other, so it doesn't apply to the position observable.
  13. Feb 23, 2012 #12
    I'm not talking about free will and I'm not rejecting determinism out of hand. All I'm saying is that in order for a local deterministic theory to exploit the superdeterminism loophole to Bell's theorem, it seems that it has to possess some properties that many would consider to be strange for a scientific theory to have. The conspiratorial nature is one of them: you have to assume that all the particles in the universe are doing just the right actions which will make local determinism seem false even though it is really true. And it seems pretty hard to come up with a theory that assigns to these particles exactly these "right actions", although as I mentioned earlier there has been some promising work in this direction by Nobel prize-winner Gerard t'Hooft and others.
  14. Feb 23, 2012 #13
    Free will is not an axiom of quantum mechanics; where did you get that idea? But it is true that the Copenhagen interpretation is compatible with the idea that the experimenter has free will (although it is also compatible with radical rejections of counterfactual definiteness where it becomes meaningless to even consider the possibility that the experimenter could have done something other than what they actually did).

    As for your second point, that "effective" freedom of initial conditions is necessary for science, yes, it's difficult to see how the scientific method can proceed if the initial conditions of all experiments have such large systematic biases as required by superdeterminism.
  15. Feb 23, 2012 #14
    I never said that nonlocality or violations of relativity are issues for superdeterminism.
  16. Feb 23, 2012 #15
    Demystifier, let's call the conclusion of the Kochen-Specker theorem "fakeness". If you have a set of noncommuting observables, does Kochen-Specker state that all of them are fake, or just that just at least one of them must be fake? The reason I'm asking is that position and momentum are non-commuting.

    On an unrelated note, doesn't Bohmian mechanics suffer from its own fine-tuning problems, namely that the universe got into just the right state that comports with the Born rule? I think this is a somewhat odd issue for BM, because decoherence can easily explain why the Born rule seems correct in practice. Why can't this explanation be carried over into BM, which anyway utilizes decoherence in its reduction of quantum uncertainty to classical uncertainty?
  17. Feb 23, 2012 #16
    why is 2 even necessary
    and initial conditions should have interacted at some point correct?
    i dont see any problem with nonlocality because it can never beat c to the finish
  18. Feb 23, 2012 #17
    could you give a simpler explanation of the superposition?
  19. Feb 23, 2012 #18


    User Avatar
    Science Advisor

    My bad, I intended to write "the Copenhagen interpretation".
    In the CI, measurements involve the freedom to choose what's being measured. If the measurements themselfs are to be explained by (quantum mechanical) laws, I think we let go of the CI.
  20. Feb 23, 2012 #19
    how about nonlocal superdeterminism? is that falsifiable?
    i really dont see why part 2 is necessary at all
  21. Feb 23, 2012 #20
    imo we have the illusion that we are choosing, i dont know how anybody could embrace ci. do they think they could have done measurements any differently than they did them? lol
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook