I Confused by nonlocal models and relativity

[A. Neumaier]

While contributions to quantum gravity by quantum foundations can't be ruled out a priori, I don't see it in recent literature. Is there some recent review that explores potential connections?

For example,

• Rovelli, C. (1996). Relational quantum mechanics. International Journal of Theoretical Physics, 35, 1637-1678.

The paper is not very recent, but has 269 citations in Google Scholar since 2022.

I suggest that the common unease with taking quantum mechanics as a fundamental description of nature (the measurement problem) could derive from the use of an incorrect notion, as the unease with the Lorentz transformations before Einstein derived from the notion of observer-independent time. I suggest that this incorrect notion that generates the unease with quantum mechanics is the notion of observer-independent state of a system, or observer-independent values of physical quantities.

Two recent papers are

• Carroll, S. M. (2022). Addressing the quantum measurement problem. Physics Today, 75(7), 62-63.
• Cavalcanti, E. G., Chaves, R., Giacomini, F., & Liang, Y. C. (2023). Fresh perspectives on the foundations of quantum physics. Nature Reviews Physics, 1-3.

See also my comments from 2017 here (Point 12).

You have to bump up against the nomological character of a physical theory at some point. A theory that completely determines future outcome from an initial state + dynamics would not explain why those dynamics are correct, as opposed to some alternative scheme.

The unification of relativistic QFT and general relativity is already so constrained that any successful unification would count as the most correct one, since it explains the most experiments! The question of assessing alternatives arises only if there are several competing successful unifications.

 

[vanhees71]

Correctly but not completely. It is not shown why the unitary quantum dynamics gives rise to unique macroscopic outcomes. This is an open problem that you simply ''solve'' by making the metaphysical assumption of intrinsic quantum randomness. But this is hiding the problem under the carpet, not solving it.

But there's not the slightest hint that this assumption is not correct. You can speculate a lot, but without empirical guidance unlikely to find more comprehensive theories. It's even not clear, whether there should be such a more comprehensive theory. All the Bell tests speak against it.

 

[Fra]

Few researchers search in the region backed by experiment.

Which specific region would you suggest this is?

I ask because there extrapolation from experimentally accessible domain to the hypothetical domain is rarely unique, it is often dependent on interpretation or in what form you seek the answers.

/Fredrik

 

[A. Neumaier]

But there's not the slightest hint that this assumption is not correct.

Neither is there the slightest hint that your assumption is correct. Both are metaphysical assumptions until someone finds a positive solution.

 

[A. Neumaier]

Which specific region would you suggest this is?

https://www.physicsforums.com/threads/confused-by-nonlocal-models-and-relativity.973876/post-6968312

I answered this already. See also the last paragraph here.

 

[vanhees71]

Neither is there the slightest hint that your assumption is correct. Both are metaphysical assumptions until someone finds a positive solution.

Of course, both are metaphysical assumptions, but there's huge empirical evidence for the correctness of the assumption that Nature is inherently random and none to the opposite.

 

[A. Neumaier]

Of course, both are metaphysical assumptions, but there's huge empirical evidence for the correctness of the assumption that Nature is inherently random and none to the opposite.

No. There is huge empirical evidence for randomness, but not the slightest evidence for inherent randomness.

 

[vanhees71]

So, how do you interpret all the confirmations of QT against "local realistic HV theories"?

 

[Morbert]

You have to bump up against the nomological character of a physical theory at some point. A theory that completely determines future outcome from an initial state + dynamics would not explain why those dynamics are correct, as opposed to some alternative scheme.

The unification of relativistic QFT and general relativity is already so constrained that any successful unification would count as the most correct one, since it explains the most experiments! The question of assessing alternatives arises only if there are several competing successful unifications.

I should have clarified: My statement was about the position that outcomes following probabilistically from their antecedents needs explanation while outcomes following deterministically from their antecedents can be accepted without controversy.

 

[A. Neumaier]

So, how do you interpret all the confirmations of QT against "local realistic HV theories"?

They are confirmations of the minimal interpretation of quantum mechanics, but not of your metaphysical extension that claims irreducible randomness.
They confirm equallly well the thermal interpretation, in which no irreducible randomness exists.

 

[vanhees71]

So you claim that the confirmation of QT against local realism is compatible with the assumption of determinism? If so, then you'd need to provide a (then necessarily non-local) deterministic HV theory compatible with the violation of Bell's inequalities that is consistent with Einstein causality, as is relativistic microcausal QFT.

 

[A. Neumaier]

My statement was about the position that outcomes following probabilistically from their antecedents needs explanation while outcomes following deterministically from their antecedents can be accepted without controversy.

Well, if something is determined, why ask for a reason for its value? The determinstic law is the reason! But if something is not determined but can take several values, curiosity arises as long as we can point to no law for the precise value.

 

[A. Neumaier]

So you claim that the confirmation of QT against local realism is compatible with the assumption of determinism? If so, then you'd need to provide a (then necessarily non-local) deterministic HV theory compatible with the violation of Bell's inequalities that is consistent with Einstein causality, as is relativistic microcausal QFT.

The N-point functions of a relativistic QFT satisfy the deterministic but nonlocal Schwinger-Dyson equations and describe everything observable about QFT.

The Schwinger–Dyson equations (SDEs) or Dyson–Schwinger equations, named after Julian Schwinger and Freeman Dyson, are general relations between correlation functions in quantum field theories (QFTs). They are also referred to as the Euler–Lagrange equations of quantum field theories, since they are the equations of motion corresponding to the Green's function.

 

[vanhees71]

The Wightman functions of local observables do not imply determinism for the observables but only for the probabilities of their outcome when measured or rather you can calculate these probabilities using them.

Also, I don't know, what you mean by "non-local" here. If it comes to observables, these must be represented by local field operators, obeying the microcausality condition.

 

[A. Neumaier]

The Wightman functions of local observables do not imply determinism for the observables

The observable macrovariables are all given by locally smeared 1-point functions, without thinking about probabilities. They are fully determined by the Wightman functions.

but only for the probabilities of their outcome when measured or rather you can calculate these probabilities using them.

Deterministic equations and hidden variables are always for beables!

I thought we had agreed that the beables in quantum theory are the probability distributions and not the individual outcomes.

Also, I don't know, what you mean by "non-local" here. If it comes to observables, these must be represented by local field operators, obeying the microcausality condition.

In the thermal interpretation, the observables are the N-point functions and what can be computed from them. Every quantum phycisists known how to observe these observables for sufficiently stationary objects, at least for N=1 and N=2.

A 2-point function is already nonlocal since the two arguments can be arbitrarily far apart. Such observables are not allowed in Bell's local hidden variable analysis.

 

[vanhees71]

The observable macrovariables are all given by locally smeared 1-point functions, without thinking about probabilities. They are fully determined by the Wightman functions.

These are the expectation values not the outcomes of measurements. We discussed this already some time ago: When you measure the spin-$z$ component of an electron prepared to have the spin-$x$ component determined to be, say, +1/2, you don't get the expectation value $\langle s_z=0 \rangle$ when measuring $s_z$ accurately but $s_z=\pm 1/2$ with probability 1/2 for each outcome.

Deterministic equations and hidden variables are always for beables!

Can you say, what you mean? We just have discussed that "beables" is an undefined word play by Bell!

I thought we had agreed that the beables in quantum theory are the probability distributions and not the individual outcomes.

I don't know, what "beables" are. I talk about measurement outcomes in a lab. The probability distribution can of course be measured by preparing "statistical samples" as proxies of the corresponding "ensembles". The probability distributions are of course determined but not the outcome of measurments on observables (except those, which are determined in the given state of course).

In the thermal interpretation, the observables are the N-point functions and what can be computed from them. Every quantum phycisists known how to observe these observables for sufficiently stationary objects, at least for N=1 and N=2.

The $N$-point functions do not predict with certainty the outcome of measurements but the probabilities/expectation values of the measured observables given the state you use to evaluate the corresponding expectation values.

A 2-point function is already nonlocal since the two arguments can be arbitrarily far apart. Such observables are not allowed in Bell's local hidden variable analysis.

Of course, measurement devices can be placed arbitrarily far away from each other. E.g., the two-point function of the electric field describes two-photon detection (see, e.g., Garrison&Chiao Sect. 6.6.2), used in the Bell tests (coincidence measurements on two entangled photons). That doesn't make the QFT non-local. Also in Bell's LHV analysis he considers of course the same, i.e., the measurement of parts of a system at far-distant places.

 

[A. Neumaier]

When you measure the spin-$z$ component of an electron

This is not a macrovariable. The corresponding macrovariable actually measured is the amount of silver on the plate, and this is a smeared 1-point function of the silver density field.

Can you say, what you mean? We just have discussed that "beables" is an undefined word play by Bell!

No. We discussed that it is a defined word in a paper about beables. All his foundational work is about the consequences of having beables. His hidden variables are the parameters characterizing his beables, and he proves that local beables must satisfy his inequalities.

I don't know, what "beables" are.

For the present discussion (which at the moment is about the thermal interpretation), the beables are the N-point functions, and since you know what the latter are, you know what I mean by beables in the thermal interpretation!

I talk about measurement outcomes in a lab.

Bell's beables are theoretical generalizations of macroscopic things - pointer positions, colors, image patterns - the things that one can read off with certainty from the equipment in the lab, before one translates them into properties of invisible objects like electrons. He said that in Maxwell's theory, the electromagnetic fields are beables (in contrast to the vector potential); their smeared values are measurable with certainty.

The $N$-point functions do not predict with certainty the outcome of measurements

The 1-point function of an electromagnetic current predicts with certainty the current read from a current meter. Thus it is a beable in bells sense.

Of course, measurement devices can be placed arbitrarily far away from each other. [...] That doesn't make the QFT non-local.

Not the QFT but 2-point function are nonlocal. Since the Schwinger-Dyson equations couples the N-point functions for all N, their deterministic dynamics is Bell nonlocal. Thus there is no contradiction with Bell's theorem, since the latter assumes Bell locality.

This Bell nonlocality has nothing at all to do with the fact that relativistic QFT is local in the sense used by field theorists. It only has to do with the form of the deterministic equations!

 

[Fra]

Well, if something is determined, why ask for a reason for its value? The determinstic law is the reason!

I disagree and this is think again a symptom of that we seek answers of different forms.

My view towards this is summarized well be this philosophical quote from Charles Sanders Pierce, that Lee Smolin quoted in one of his papers of evolution of law.

"...To suppose universal laws of nature capable of being apprehended by the mind and yet having no reason for their special forms, but standing inexplicable and irrational, is hardly a justifiable position. Uniformities are precisely the sort of facts that need to be accounted for. Law is par excellence the thing that wants a reason. Now the only possible way of accounting for the laws of nature, and for uniformity in general, is to suppose them results of evolution..."
-- https://arxiv.org/abs/1201.2632

/Fredrik

 

[A. Neumaier]

this philosophical quote from Charles Sanders Pierce, that Lee Smolin quoted in one of his papers of evolution of law.

"...To suppose universal laws of nature capable of being apprehended by the mind and yet having no reason for their special forms, but standing inexplicable and irrational, is hardly a justifiable position. Uniformities are precisely the sort of facts that need to be accounted for. Law is par excellence the thing that wants a reason. Now the only possible way of accounting for the laws of nature, and for uniformity in general, is to suppose them results of evolution..."
-- https://arxiv.org/abs/1201.2632

I reject this view. The fundamental physical laws are eternal, since they are supposed to hold in the whole universe at all times. Thus they cannot evolve. Only our understanding of them evolves.

 

[Fra]

I reject this view. The fundamental physical laws are eternal, since they are supposed to hold in the whole universe at all times. Thus they cannot evolve. Only our understanding of them evolves.

I figured you would. This is also why I see why it follows that you view some of my views as circular.

One of the reasons why I tend to consider physical interrelations in material nature beyond human knowledge, is that its the only conceptual way I found to explain for example entanglement correlation and therefore its for personally me a heavy argument. We know from Bells inequality that human ignroance can not explain this (unless one adds even MORE strange things that instant actions etc, that i dont see adds any explanatory value at all) there seems to be an ignorance built into the physical interactions themselves. And its these, that I still in principle that is evolving, and the fixed laws we de facto are aware of in the limited domain of validity are of natures self-organisation to mutual understanding. But all this is "interpretations" of course.

/Fredrik

 

[A. Neumaier]

I tend to consider physical interrelations in material nature beyond human knowledge,

But the physics community is human and knows a huge amount about the physical interrelations in material nature - much more than you do. So I don't think your view carries any weight.

 

[Fra]

But the physics community is human and knows a huge amount bout the physical interrelations in material nature

But apparently not enough to have found a unified theory of interactions.

Alot of intelligent people has spent alots of time on these problems for many decades and still many hard unsolved problems. Perhaps new different thinking is required.

/Fredrik

 

[PeterDonis]

But apparently not enough to have found a unified theory of interactions.

We do have one for all the interactions except gravity. It took only a couple of decades for that unified theory to be developed and tested.

Alot of intelligent people has spent alots of time on these problems for many decades and still many hard unsolved problems.

This has always been true and will always be true.

Perhaps new different thinking is required.

And if you think this, you should go do it. But PF is not the place for it, since PF is not for discussion of personal research. Get your new different thinking published in a peer reviewed paper and then we can talk. But unless and until you do that, there is no point in just continuing to wave your hands and assert that you're not satisfied with what we have now. Yes, we know that. But it offers no basis for useful discussion here.

 

[Fra]

We do have one for all the interactions except gravity. It took only a couple of decades for that unified theory to be developed and tested.

There is no distinguished unification of QCD and electroweak theory that would reduce the number of free parameters.

/Fredrik

 

[PeterDonis]

There is no distinguished unification of QCD and electroweak theory that would reduce the number of free parameters.

You're quibbling. We have a unified theory of all the interactions except gravity. It might not meet your idiosyncratic requirements, but why should I care?

In any case, as I said, if you think our current theories are incomplete, go do research to help complete them. Complaining about it in a PF thread is pointless and adds no value. So please stop doing it.

 

[bhobba]

The problem is which region? Is there any empirical evidence of a quantum-gravitational effect?

Would Hawking radiation count? My understanding is it can be explained by the Effective Field Theory (EFT) of gravity, which is valid to about the Plank scale.

I agree that QM is incomplete because we only have a quantum theory of gravity as an EFT. But then again, I thought the modern view was all our field theories are EFTs. So I think it may go beyond just gravity.

Thanks
Bill

 

[bhobba]

In any case, as I said, if you think our current theories are incomplete,

Ditto.

Thanks
Bill

 

[PeterDonis]

Would Hawking radiation count?

Not as empirical evidence since it has not been observed and there is no prospect of it being observed any time soon.

 

[PeterDonis]

I thought the modern view was all our field theories are EFTs.

Yes, that is correct. So our modern field theories offer no evidence for QM being complete.

 

[Fra]

In any case, as I said, if you think our current theories are incomplete, go do research to help complete them. Complaining about it in a PF thread is pointless and adds no value. So please stop doing it.

Surely never meant to "complain".

I thought discussing in a constructive but critical way the current state of the art theories and in partucular their foundations and was what we did here - but without adding personal speculation, which I try carefully to not do even if everyones way of reasoning is naturally slighly coloured by personal,bias.

If adding perspectives adds no value to anyone then fine with me. I personally enjoy understanding other perspectives even if I don't share them.

/Fredrik