"Violation of Bell's inequality in fluid mechanics"

[nonequilibrium]

I thought it might be interesting to point out this article:

Title: Violation of Bell's inequality in fluid mechanics
Authors: Robert Brady and Ross Anderson (Cambridge)
Abstract:

We show that a classical fluid mechanical system can violate Bell's inequality because the fluid motion is correlated over large distances

 

[atyy]

Are the equations they are studying relativistic? If not, then they are nonlocal.

I think they are studying solutions of Eq 1 of http://arxiv.org/abs/1301.7540 which looks like the non-relativistic Euler equation.

 

[DrChinese]

I thought it might be interesting to point out this article:

Title: Violation of Bell's inequality in fluid mechanics
Authors: Robert Brady and Ross Anderson (Cambridge)
Abstract:

This is a controversial result, and will not past muster across the board in the community. I seriously doubt it can be demonstrated in an experimental situation. A lot* of attempts have been made to demonstrate classical systems can display behavior similar to entanglement, none of which have been satisfactorily executed to date.

Physically viable local realistic systems do not violate Bell inequalities.


*At least 1 a month.

 

[jtbell]

Note that although this preprint is more than a year old, it has apparently not (yet?) passed peer review and been formally published in any journal.

 

[TrickyDicky]

Physically viable local realistic systems do not violate Bell inequalities.

The paper makes a point of distinguishing the local realistic system Bell proved incompatible with his inequalities and that is indeed shown to violate them in all quantum experiments, that posited point-like particles, from what is observed in fluids.
Collective excitations and pseudoparticles(of non-local nature) in a fluid seem like cannot be included in this concept of local realism the theorem of Bell refers to. Or am I missing the point of this Cambridge group?

 

[atyy]

I'm reporting this post, [STRIKE]since this thread is about a crackpot paper claiming that Bell's theorem is wrong[/STRIKE]. But let me ask some questions anyway, since the thread is not closed yet.

[STRIKE]We know the conclusion is wrong, but why is it wrong?[/STRIKE] In TrickyDicky's thread https://www.physicsforums.com/showthread.php?t=758324, the responses (#22-#24) from stevendaryl, DrChinese and bhobba are in line with my intuition that Newtonian gravity is nonlocal in the sense of Bell's theorem. If that is correct, Bell's theorem does not apply to Newtonian gravity (ie. it doesn't mean that Newtonian gravity can reproduce the predictions of quantum mechanics, but it does mean that Bell's theorem does not rule out gravity as a realistic theory that is able to reproduce the predictions of quantum mechanics). So [STRIKE]is the error in this paper that[/STRIKE] does Bell's theorem not apply to this paper because it starts from the non-relativistic Euler equations? Are the non-relativistic Euler equations non-local, just as Newtonian gravity is nonlocal?

Edit: I did report the post, but having read the paper more carefully, I see that they do agree that Bell's theorem does rule out local hidden variables. What they are claiming is that Bell's theorem does not rule out "local interactions", which they seem to consider different from local hidden variables. I would still like to know whether the non-relativistic nature of Euler's equations makes them nonlocal in the sense of the Bell theorem.

 

[atyy]

Another thought: Can one even set up a Bell test in Newtonian physics? In a Bell test, the pair of measurements have to be performed at spacelike separation. Thus, a Bell test seems to assume special relativity. But there is no concept of spacelike separation in Newtonian physics, so how can a Bell test be set up?

There can be emergent relativity from a non-relativistic system, and an effective "upper speed limit" from a Lieb-Robinson bound. But in which case, the Bell test must be carried out only in the low energy "emergent relativity" regime.

 

[DrChinese]

What they are claiming is that Bell's theorem does not rule out "local interactions", which they seem to consider different from local hidden variables. I would still like to know whether the non-relativistic nature of Euler's equations makes them nonlocal in the sense of the Bell theorem.

One of the usual flaws in proposed "anti-Bell" experiments of this general type is that the there is some communication channel between sides which is not explicit. That could occur here, not really sure.

Also, as you say, if any theoretical portion of the setup is based on non-local interaction, then the same applies. That too could occur here.