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

B Why is EM not non-local?

  1. Apr 24, 2017 #1
    Quantum mechanics is non-local
    EM or electromagnetism obeys QM.
    But why is EM not non-local?
    How to make EM non-local?
     
  2. jcsd
  3. Apr 24, 2017 #2

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    QM is not non local - that's a misconception but requires its own thread - start one if you like - note standard QM being based on the Galilean Transformations is explicitly not local but the explanation of whats going on, while not hard, is a bit wordy so really another thread is required.

    EM is not non local because it in fact is derivable from SR, assuming simply Coulomb's law, which is from its foundations local:
    http://www.cse.secs.oakland.edu/haskell/Special Relativity and Maxwells Equations.pdf

    QED, being a quantum field theory (QFD), obeys whats called the cluster decomposition property which is the version of locality used in QFT, and in fact since standard QM is explicitly not local is really what locality is in QM:
    https://www.physicsforums.com/threads/cluster-decomposition-in-qft.547574/

    Once you understand the above the issue of locality is seen as nothing like what popularization's make it out to be - a big deal - it isn't. But again that's for the separate thread where all these issues can be discussed. Basically though the question comes about because of Bells theorem, but that involves correlation which is specifically not part of cluster decomposition so even the whole thing is rather moot.

    Thanks
    Bill
     
    Last edited: Apr 24, 2017
  4. Apr 25, 2017 #3
    Is it the case that theories like QFT, which are based on cluster decomposition, are silent about the outcome of correlation type measurements?
     
  5. Apr 25, 2017 #4

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    Of course not.

    Its simply in the definition of locality used in QM correlations are specifically excluded so the question of locality in Bell type experiments depends on what you mean by locality. If its just in the Cluster Decomposition sense then its not even a question that comes up. Correlated systems have correlated results - big deal. But if you ask about the nature of those correlations then its different to classical correlations - that's Bells theorem. Want it to be the same - you can have that - but you have to extend your view of locality beyond The Cluster Decomposition property ie what QM demands. You don't have to do that if you don't want to - you can just say the question of locality doesn't apply to correlated systems so forget about the issue - its not relevant. Its what I do. You may do it because certain interpretations are specifically non local eg BM - or you have some other reason - but if you don't want to go down that path you don't have to - its entirely up to you.

    Thanks
    Bill
     
  6. Apr 25, 2017 #5
    So non-locality can be done away if one says locality doesn't even exist. So it's "non-blank" since there is no locality.. or since it's double none.. then its nothing. I went to that Cluster Decomposition thread. In a nutshell, is Cluster Decompoistion about correlations that don't involve locality?
     
  7. Apr 25, 2017 #6

    Demystifier

    User Avatar
    Science Advisor

    There are different types of locality (or non-locality), like signal locality or Bell locality. QM obeys signal locality but Bell non-locality. Quantum electromagnetism, just like QM, also obeys signal locality and Bell non-locality.
     
  8. Apr 25, 2017 #7
    please briefly define

    signal locality
    Bell non-locality

    thanks.
     
  9. Apr 25, 2017 #8

    bhobba

    User Avatar
    Science Advisor
    Gold Member

    Signal locality does not include correlations because they cant be used to signal anything. Signal locality is whats required for SR - you need to sync clocks which needs a signal to pass between them - the same for QFT which is built on SR. Bell non-locality is one way to escape the non classical correlations QM has - you can have classical correlations if non locality is broken between correlated objects.

    But one has to ask - why bother? There is no correct answer of course - I chose to not bother - things IMHO are much easier that way - but opinions vary.

    Thanks
    Bill
     
  10. Apr 25, 2017 #9

    Demystifier

    User Avatar
    Science Advisor

  11. Apr 25, 2017 #10

    martinbn

    User Avatar
    Science Advisor

    If Bell had used a better term instead of local/non-local, for example a term that was not already in use, then most of the discussions about Bell's theorem would not exist. What's wrong with terms like non-separable or non-factorizable?
     
  12. Apr 25, 2017 #11
    QFT and QED obey signal locality.. so we send radio waves over the atmosphere but it can't pass thru the earth core because of the core and mantle density. Why is there no signal that still obey SR but doesn't use EM.. so we can make receiver in one side of the earth and another side and it can appear without travelling but obey SR.. like using some king of higgs field but with vectorial component to communicate.. why doesn't such exist? what symmetry or laws of nature is violated by this (yet still obey SR)?
     
  13. Apr 25, 2017 #12

    stevendaryl

    User Avatar
    Staff Emeritus
    Science Advisor

    I wouldn't say that Bell non-locality means violation of Bell's inequalities. Rather it's an implication: Violation of Bell's inequalities implies Bell non-locality.
     
  14. Apr 25, 2017 #13

    Demystifier

    User Avatar
    Science Advisor

    Conceptually, I perfectly agree. But then it's not so easy to give a precise short definition of Bell non-locality without dwelling into vague philosophy.
     
  15. Apr 25, 2017 #14

    Demystifier

    User Avatar
    Science Advisor

    There are several reasons for that. "Non-factorizable" sounds like a purely mathematical property without a direct physical content. "Non-separable" sounds mysterious, perhaps even more than "non-local". But probably the main reason was the fact that Bell wanted to justify his favored interpretation of QM, which was the non-local Bohmian interpretation.
     
  16. Apr 25, 2017 #15

    mfb

    User Avatar
    2016 Award

    Staff: Mentor

    Neutrinos? It is not practical, but in principle you can transmit information via neutrino beams. They can easily pass through the Earth.
     
  17. Apr 25, 2017 #16

    Demystifier

    User Avatar
    Science Advisor

    It must use one of the four fundamental forces. The so called strong and weak forces act only on small distances. Gravity is too weak. What remains is the EM force.
     
  18. Apr 25, 2017 #17

    vanhees71

    User Avatar
    Science Advisor
    2016 Award

    Einstein used them, but nobody followed him, because everybody thought, he'd just not understood QT. I think the opposite is true, although I don't follow his argument that you can conclude that QT is "incomplete" in some way. As it looks today (much more than in 1948 when Einstein clarified his view which he felt not having been adequately represented by the famous EPR article; it's in German however: A. Einstein, Quanten-Mechanik und Wirklichkeit, Dialectica 2, 320 (1948)) QT seems to be rather complete as far as it is formulated and applicable (the great exception is the absence of a consistent QT of gravitation) is accurately describing the behavior of observable nature, including the irreducible probabilistic elements it includes.
     
  19. Apr 25, 2017 #18

    why are there no non gauge forces? what exact stuff in physics is violated?
     
  20. Apr 25, 2017 #19

    Simon Phoenix

    User Avatar
    Science Advisor
    Gold Member

    Personally, I don't find these terms to be too helpful and in some sense I think they're confusing. Intuitively the local/non-local thing is about whether things we do in our lab (local) can affect the results of experiments performed in some other lab (non-local) - and I suppose we also ought to add in the qualification that any such effect (if it exists) must not occur faster than the time it takes light to travel between the labs.

    In this sense QM is a fully local theory - even in its standard (non-relativistic) formulation. Of course the fact that in QM things we do 'here' do not affect results 'there' implies that we can't send any information, so the signal locality is really a consequence in my view.

    I personally don't really see much virtue in defining something called "Bell non-locality" since Bell's inequality is an entirely classical inequality that (loosely) says ##if## our system is described by variables that are (a) local and (b) have some meaning independent of measurement ##then## the correlations are constrained by the Bell inequality. It doesn't really say anything about QM, as such.

    What we can say is that the predictions of QM cannot be fully reproduced by a theory constructed from variables that have the properties (a) and (b). So clearly QM is not this kind of theory - which doesn't really tell us what kind of theory QM actually ##is##, it just says what kind of theory QM cannot be. To then say that QM is therefore "Bell non-local" seems a bit arse-about-face to me, to use a quaint English expression. I don't really get what describing QM as Bell non-local brings to the table, so to speak - it just seems to muddy the waters.

    And I'll stop there before the mixed metaphor and idiom police arrest me.
     
  21. Apr 25, 2017 #20

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    But that is precisely why we need a term like Bell non-local: your statement is not necessarily true, and Bell tests highlight this. It is absolutely the case that Alice's choice of measurement *appears* to change the nature of Bob's reality somewhere else faster than any light signal. (Of course, the reverse is equally true for Bob, and can even *appear* to be backwards in time.) You cannot rule that out - so whatever that is, it is Bell non-local and that is a perfectly reasonably description in my eyes. Or you can call it quantum non-locality, a term I also see used. But it is clearly different than signal non-locality.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Why is EM not non-local?
  1. Non-local and non-linear (Replies: 12)

Loading...