Pages

Thursday, October 13, 2022

Nobel Prize was Not for Non-locality

Commentary about the 2022 Nobel Physics Prize breaks down into two camps: (1) the experiments confirmed quantum mechanics as it has been understood since 1932; or (2) revolutionary experiments show nature is not real or local.

I pointed out that the Nobel committee stuck to (1), and pointedly did not endorse (2).

Another person who noticed this was Physics Philosopher Tim Maudlin, who wrote:

Unfortunately, much of this history has been garbled in the public discussion of Bell’s work and its experimental tests. The Nobel prize committee itself gets it wrong in its press release,
"John Clauser developed John Bell’s ideas, leading to a practical experiment. When he took the measurements, they supported quantum mechanics by clearly violating a Bell inequality. This means that quantum mechanics cannot be replaced by a theory that uses hidden variables."
But that statement is flatly false. Indeed, it was a theory that uses hidden variables—Bohmian mechanics—that inspired Bell to find his inequalities, and that theory makes the correct prediction that the inequalities will be violated.
The Nobel statement is correct if "theory" means "local theory". For the most part, theories have to be local to be scientific. A nonlocal theory can have magical action-at-a-distiance.

Bohr, Einstein, Bohm, Bell, and everyone else did not believe in nonlocal theories. Bell and Bohm only wrote about Bohm's theory as a mathematical curiosity. Bell was parially motivated by trying to find a local version of Bohm's theory, not to accept a nonlocal theory.

Maudlin is exceptional in that he believes in nonlocal interpretations. I think Botmian pilot wave theory is his favorite, and he claims it can be turned into a full interpreation of quantum mechanics.

Much as poeple like to argue that QM is strange, Bohmian mechanics is far stranger. In it, an electron can be observed in one place, and its ghost can be causing weird effects elsewhere.

Maudlin concludes:

What Bell’s theoretical work and the subsequent experimental work of Clauser, Aspect and Zeilinger proved was non-locality, not no-hidden-variables. Ultimately, they proved Einstein wrong in his suspicions against spooky action-at-a-distance. And that, surely, deserves the highest honors one can bestow.
Yes, they would deserve the highest honors if they proved non-locality, and that spooky action-at-a-distance really does happen. But the Nobel Prize citation pointedly does not say any of those things. The prize was only for experimental work confirming quantum mechanics as it was understood in 1932.

In his later years, Bell adopted a view that a true theory of nature should be based on hidden variables, which he called by a term he coined, "beables". So hidden variables became a philosophical necessity. So when the Bell test experiments ruled out local hidden variables, he adopted nonlocal hidden variables. I think that Maudlin was persuaded by that argument.

But the mainstream textbooks are not persuaded, and neither is the Nobel committee.

5 comments:

  1. Dear Roger,

    An aside:

    You are into relativity and Einstein and all that, aren't you?

    So, may be, my latest blog post might be of interest to you.

    As I said in the post, I haven't gone through the papers (or the books I mention / informally review on a pro bono basis), but I guess I already have removed the points, via a good philosophical thought, which the authors should in future find as being the stumbling blocks in taking further their research interests.

    Best,
    --Ajit

    ReplyDelete
  2. Since lambda can be psi, you could call QM non-local on its face. A big wave function just collapses over a space-like separation. Those who believe that the wave function is epistemic apparently don't realized they are saying there IS a deeper theory. Bell inequality violations occur with classical light, Brownian motion and even water waves (see the latest HQFT result). This can be explained with contextual non-Kolmogorov probability. Quantum mechanics is bunk and Bell understood that. He simply didn't realized that classical random fields or other background fields don't need to assume "superdeterminism" but just ordinary determinism. Probability is a subtle concept and took everyone for a ride.

    ReplyDelete
    Replies
    1. To avoid confusion, I think by QM people should stick to non-relativistic QM (NRQM for short), and explicitly use the term Relativistic QM (RQM) to identify that they mean the relativistic version (whether in Dirac's version or with the QFT (infinite series expansion) version). And, there is no need for people to be thinking of gravity in this context. It's too weak; that's why (and not because Quantum Gravity is an unsolved problem).

      NRQM is *completely* *non-local* --- on the face of it, and also at the core of it. Therefore, it's non-local also in terms of the predictions it makes for measurement events. All the discussions related EPR, Bell, and all, which I've seen proceed as if they referred only to NRQM, not to RQM. Show me a single reference that talks of EPR/Bell using a solution that has the 4-vector structure to it.

      As to the Relativistic QM (RQM), I am not as clear about how its maths comes together, as I would like to (I am still studying the topic). But if an informal comment can be made, from the viewpoint of sharing something about my research (of extending my iqWaves approach from NRQM to Dirac's QM), then the *feeling* which I have (as of now) is the following:

      As of now, I anticipate that (i.e., even though my theory eventually may *not* turn out to be this way, there are very strong conceptual points which suggest me that it is reasonable for me to keep the anticipation that):

      -- System evolution in my theory is going to be non-local (in the sense, interactions between two particles are always going to be simultaneously occurring at all points of space at every instant of time)
      -- However, *measurement* events are going to be, *practically* speaking, more or less describable as if they were local.

      Note that, as I anticipate the development, *in principle*, even the measurement events are going to be non-local. But it's just that if you take a careful look into how the detectors themselves work, I feel that my [yet to be done] theory should come to predict that a record of measurement events would look as if there were no instantaneous action at a distance to their generation.

      [BTW, the particle of primary concern to me is the electron, not the photon, and certainly not the positron.]

      As to what I mean by "practically" vs. "in principle": Here are two different examples: (i) The wavefunction of a single electron "in" your room would, in principle, be spread over all the universe, extending even beyond the known galaxies. Practically, its wavefunction is all confined to a small region of about a few hundred picometer in your room. (ii) The evolution of a non-linearly coupled system like the 3-body problem in Newtonian gravity, is in principle deterministic, but due to the sensitive dependence on initial conditions (SDIC), the time evolution of any practical 3-body system, in general, cannot be predicted at all. [There may be specific realizations which may be predictable, but the general case is, *practically* speaking, indistinguishable from the unpredictable.]

      In short: If my anticipation turns out to be correct, then discussing these issues in terms of local vs. non-local descriptions alone, as if all theoretical alternatives must get exhausted with this binary, would be shown, with good reason, as falling too short. [I anticipate that it should take a few months to a year for me to complete my RQM theory. I plan never to work on gravity + QM.]

      BTW, thanks for mentioning many points/results I didn't know about.

      Best,
      --Ajit

      Delete
  3. The Nobel committee did not endorse any of Bell's philosophical views.

    ReplyDelete
  4. NRQM is only nonlocal in the sense that if you replace it with a hidden variable theory, that theory must be nonlocal.

    ReplyDelete