What do violations of Bell's inequalities tell us about nature?

bohm2

Please vote and if possible state the reasons for holding your belief. As a review here are the two major views with quotes by leading physicists in quantum foundations:

1. Observed violations of Bell's inequalities implies that nature is non-local:
2. Observed violations of Bell's inequalities implies anti-realism (e.g. quantum measurement results do not pre-exist)

LastOneStanding

This is a bizarre question. Violations of Bell's inequalities just tell us that at least one of (1) and (2) must be true. It doesn't prefer one or the other, nor does it rule out both of them being true (as is the case in the Copenhagen interpretation). Various people may well have preference for either anti-realism or non-locality but that preference can't possibly come from Bell's theorem alone. It's complete nonsense to say, "Observed violations of Bell's inequalities implies that nature is non-local," or, "Observed violations of Bell's inequalities implies anti-realism." Observed violations of Bell's inequalities imply neither.

Either you're misunderstanding Bell's theorem, or you did an extremely poor job of phrasing your question.

LastOneStanding

Also, your "other" category seems very confused. Alternative interpretations of QM are not exempt from having to deny either locality or counterfactual definiteness. Many worlds, for instance, does the latter.

bohm2

The exact same question was posed to leading experts in quantum foundations in this book here (see chapter 8). I'm interested in how people on this forum would respond. Some of those quotes come from that book chapter:

Elegance and Enigma: The Quantum Interviews
http://www.amazon.com/Elegance-Enigm.../dp/3642208797

Maui

There are different interpretations, but generally violations of Bell's inequalities imply what's already known - that classical mechanics(strict materialism) is just one aspect of reality and so no longer an adequate explanation of observations. As Heisenberg once put it/quoted by Nick Herbert in Quantum Reality/:

"The ontology of materialism rested upon the illusion that the kind of existence, the direct 'actuality' of the world around us, can be extrapolated into the atomic range. This extrapolation, however, is impossible... atoms are not things."

The way to keep the strict materialism intact is by accepting a small conspiracy - superdeterminsim or hidden variables(or to deny interest into the inner workings of reality).

bohm2

I don't think anybody has ever given a good definition of "materialism". Do you have one? And why do you think that a non-local, "realistic" model would still be considered "materialistic"?

DennisN

I voted "anti-realism". My reasons/opinions are:

• "Nature is non-local"; I wouldn't accept this without an underlying mechanism which describes it.
  
• "Other: Superdeterminism, backward causation, many worlds, etc"; I can't see how any of these interpretations would be falsifiable, and this makes me doubt their scientific value.

Therefore I lean towards "anti-realism". I am however pretty agnostic, and my views could change depending on future science and experiments. I would have preferred to vote on a fourth "softer" option; (observed violation of Bell's inequality tell us) there are parts of QM we can't yet fully comprehend/explain.

nanosiborg

I would vote that violations of Bell inequalities tell us nothing about nature if your poll had that as an option.

Bell's theorem proves that there's no function, ρ(λ), for which this correlation coefficient,
C(a,b) = ∫ ρ(λ) A(a,λ) B(b,λ) dλ , matches Malus' Law (cos²θ) .

The results of Bell tests involving photons entangled in polarization support the generalization of results from classical and quantum wave optics involving crossed polarizers in that the QM treatments of optical Bell test setups are evaluated using Malus' Law.

The results of Bell tests don't reveal anything new regarding fundamental empirically based tenets of wave optics. They certainly don't imply that nature is nonlocal ... though it's tempting to assume that nature is nonlocal by virtue of the fact that nonlocal hidden variable models of quantum entanglement are viable. They also don't imply the "other" option, which, as DennisN pointed out, are all untestable assumptions. For me they're just either meaningless (backward causation, many worlds) or superfluous (superdeterminism) as well. As for anti-realism, it isn't clear to me what is meant by "quantum measurement results do not pre-exist". The measurement results in Bell tests are either detection or nondetection within a coincidence interval. Obviously, these results don't "pre-exist". If it's simply meant that realism (ie., hidden variable accounts, or the existence of hidden variables) is ruled out, then we know that that's false. Realism isn't ruled out.

So, what are we left with? Just that there are hidden parameters operating to produce quantum entanglement stats that remain hidden (ie., unknown) -- and from that it still isn't known whether there is some sort of nonlocality in nature or if nature is evolving exclusively according to the principle of local action. But we do know that formulating models of Bell tests in terms of Bell locality is ruled out. Which means that models of quantum entanglement can't take the form that Bell's locality condition requires them to take.

Nugatory

I wish you had given us a fourth choice: "abstain, until such time as someone can propose an experiment that could distinguish (a) from (b)". That way my abstention could be recorded

bohm2

That's option 3: Other

Nugatory

I don't think that's a completely fair criticism (and I say this despite having already complained about the lack of an "abstain" option).

Both locality and realism are so natural and so deeply ingrained in our thinking that once we know we can't have both, it's interesting to ask "if you had to give one up, which would it be?"... And I doubt that many people would join Bohr and answer "lose 'em both!", although that answer certainly is not excluded by Bell experiments or anything else we know.

Nugatory

No, no, no... I will not cast a vote that might be counted with "superdeterminism, backwards causation, many worlds, etc.". I DEMAND a respectable abstention that allows me to shut up and calculate without committing myself to any position

danR

I vote for 1. I not only see no reason why quantum behaviour cannot be non-local, I could conjecture that some property/variable of the original universe did not expand with 4-space, which we might call quantum-field, and is a property that particles near the original size of the universe share.

bohm2

That was my reason also. It just seems that some "remnant" or "property" of the non-spatial-temporal stuff that gave "birth" to the big bang should still be with us.

DrChinese

I love it. Nugatory is not to be denied...

danR

I can't share the "seems...should" part, however. I just offer it as a conjecture: untestable, unfalsifiable.

Having said that, I would metaphysically ask why every single property of the primordial dimensionless point should necessarily be bound to a macroscopic, relativistically-governed spatio-temporal address.

Indeed, isn't the extraordinary part about the universe in that any property of it should have expanded at all? Why didn't it just all stay there in one a/non -local 'place' in the first place?

I asked one of my profs once what was the objection to non-locality was (i.e. "what really upsets you guys about it?"), and with me being an arts major he may have geared his answer to my understanding, and I may have misunderstood it, but it was something along the lines that it just made too many connections between distant objects.

In other words, they don't like non-locality because it sucks.

Well, that's just to bad. In our lectures and assignments and exams (this was a different prof, the first was teaching a more classical topic, though his specialty was quantum gravity) we were required to express confusion, puzzlement and great explanatory power in dealing with, say, two emitted photons; the spin of the one measured in Paris, and the spin of the other measured in Japan.

The wording is perpetually prejudiced toward the idea that two different spins, or spin-attributes, are being measured, instead of just one shared property. Perhaps I'm missing some deeper aspect to the issue that makes non-locality a problem nevertheless.

bohm2

Einstein felt the same way:
Others like Gisin question this preference of non-realism to non-locality, however:
Is realism compatible with true randomness?
http://arxiv.org/pdf/1012.2536v1.pdf