I QM: Interesting View - Get the Inside Scoop

[gentzen]

Isn't this the same issue in any probabilistic prediction? We all know the frequentist interpretation has foundational issues:
https://math.ucr.edu/home/baez/bayes.html

But applied mathematicians use it all the time, like taking dx etc., as tiny changes in x.

The Bayesian interpretation (both objective and subjective) too has certain issue. Luckily those are different issues than plague the frequentist interpretation. One interesting issue mentioned in your link is that despite your prior being subjective, it can still be wrong:

Now there's a whole sub-thread stemming from this plaint by Daryl McCullough:

I don't want to be told how my probabilistic guesstimates are supposed to change with time.

This has already been answered well:- it's the natural probabilistic way they ought to change, if you have any probability models at all. But I can see how you might still feel grumpy about this. Why shouldn't you go back and change your prior if it looks like the subsequent data a making it look *really* stupid!
This is tough to answer. For one thing, if your prior was so silly as to have zero probabilities in it, (or zero-density intervals, in the continuous case), then you may *have* to. F'rinstance, if you declared that there was *zero* prior chance of a six turning up on a dice - but then a six *did* turn up; well, you're completely stuffed! You just have to go back and start again without the silly zeros. And it'd be much the same if you had the prior not quite zero but about 10^(-35). It'd still take billions of sixes turning up before you'd posteriorly admit there was a reasonable chance of getting some sixes. Clearly that was a silly prior. (Not *wrong*, note, just silly; even by your own standards.)

Note however that the attempt to solve this issue by replacing the word *wrong* by the word *silly* is ridiculous. It makes the Bayesian interpretation "not even wrong" for no good reason. Even worse, it makes you blind to actual problems of the Bayesian interpretation, namely that there are situations were you should not assign definitive probabilities with too high precision, despite the apparent catch-all fallback that they are just subjective personal judgments.
This blindness also plagues QBism, when it just stops after a statement like:

That probability-1 assignments are personal judgments, like any other probability assignments, is essential to the coherence of QBism.

Staying silent is a suboptimal way of dealing with the question what to do when you were wrong. Just because Bruno de Finetti didn't address this question is no good excuse for ignoring it forever.

The work of A. Neumaier contains important ideas and elaborations how to overcome critical issues of the frequentist interpretation. One of those critical issues is that it doesn't apply to single systems, but only to ensembles. And part of the solution is to be realistic about the precision of magnitudes (including probabilities) appropriate for the concrete situation you want to talk about.

Another issue mentioned in your link are improper priors:

concept of an "improper prior" (by which I guess you must mean an infinite measure, ...) ... You can get into big trouble with them (e.g., the paradoxes where you "pick a random real number"), but there still seem to be cases (like the above) where one wants to think of them as a kind of prior.

A. Neumaier discussed this sort of trouble in older material (his still unpublished book "Classical and quantum mechanics via Lie algebras" or now "Algebraic quantum physics"). In the end, this is the place where I see the non-intuitiveness emerge again in his interpretation. In a certain sense, I believe that he knows this, but he has polished it away in newer material for the moment. I am really curious whether his unpublished book will ever appear, and whether that material will stay or not. But in a certain sense, it would really be unfair if this would be held against his interpretation, because the issue was there all the time for the Bayesian interpretation, and they coped with it by simply staying silent. And it is a difficult to understand technical issue. To get some feeling for how technical, see for example:
Consistency and strong inconsistency of group-invariant predictive inferences (1999) by Morris L. Eaton and William D. Sudderth
Dutch book against some `objective' priors (2004) by Morris L. Eaton and David A. Freedman

 

[EPR]

Yes, and the problem with this is that it's not religious but just observing what physicists concerned with physics and not with religion doing with quantum theory as a physical theory.

I think with all the Bell tests confirming quantum theory and with all the physical substantial knowledge used in engineering now all the apparent quibbles of interpretation are obsolete. Physicists asked Nature a question about her behavior by doing precise experiments and got an answer. The really puzzling aspect of all this is that some physicists don't accept the answer and think there are still quibbles because it's counterintuitive to their world view. However, the very purpose of the natural sciences and the empirical basis of it is not to confirm anybodies worldview or religious believes but to learn how Nature behaves. With a problem settled one should accept the result and go on to the real scientific problems. There are enough within the known physics and if you like really fundamental questions there's the still unsolved problem of how to describe the gravitational interaction consistently within QT.

It may well be that QT is indeed incomplete and one needs a completely new concept to achieve this goal, but I don't think that it's incomplete, because it's indeterministic as Einstein believed. I think the lesson we've learned is that nature is at the most fundamental level not deterministic.

This is a self-defeating argument because if true, nature cannot be deterministic at any level. This is the crux of the issue and it's not religious in nature. I believe it's a problem for biology and neuroscience rather than for philosophers.

 

[vanhees71]

What do you mean? I think a determinstic world would be a big problem for biology. Without randomness I guess there'd not have developed all the various higher lifeforms to begin with.

 

[EPR]

I mean that a true, uncaused randomness of the micro world calls into question the macroscopic causality and with it the foundations of science. Our current scientific framework rests heavily on the notion that objects and events do not somehow emerge but are the result of a very long chain of causes. If there is zero determinism in the micro world, this notion cannot be supported any longer as it lacks the elements of reality needed for its existence.
You seem to have abandoned any attempt to dig deeper than the formalism. Esp. about how the quantum world relates to 'classical' reality. It may seem like a pointless excercise at this point but it will change.

 

[Lord Jestocost]

The really puzzling aspect of all this is that some physicists don't accept the answer and think there are still quibbles because it's counterintuitive to their world view. However, the very purpose of the natural sciences and the empirical basis of it is not to confirm anybodies worldview or religious believes but to learn how Nature behaves.

Nevertheless, even physicists are human beings having foundational questions. As Richard Conn Henry remarks in “Review of ‘Quantum Enigma’ by Bruce Rosenblum and Fred Kuttner” (Journal of Scientific Exploration, 21, 185, 2007):

“No, the mystery is not quantum mechanics. The mystery is our own existence. …..

….. If you are not simply to be like a squirrel or a rabbit, you must choose some quantum mechanics interpretation (as it is called - it is not really ‘an interpretation,’ of course; it is your theory of yourself and of your experience of observations).“

 

[vanhees71]

I mean that a true, uncaused randomness of the micro world calls into question the macroscopic causality and with it the foundations of science. Our current scientific framework rests heavily on the notion that objects and events do not somehow emerge but are the result of a very long chain of causes. If there is zero determinism in the micro world, this notion cannot be supported any longer as it lacks the elements of reality needed for its existence.
You seem to have abandoned any attempt to dig deeper than the formalism. Esp. about how the quantum world relates to 'classical' reality. It may seem like a pointless excercise at this point but it will change.

Quantum theory is also not completely acausal. The probabilities develop with time by the known dynamical laws. The determinism of macroscopic observables is of course indeed emergent. It's valid in situations where you are satisfied with pretty coarse grained descriptions and the fluctuations of the corresponding collective observables are negligible within the accuracy you need to know them for a satisfactory description.

 

[bhobba]

It may well be that QT is indeed incomplete and one needs a completely new concept to achieve this goal, but I don't think that it's incomplete, because it's indeterministic as Einstein believed. I think the lesson we've learned is that nature is at the most fundamental level not deterministic.

Don't get me wrong. If Einstein believed it was incomplete because it is probabilistic (I am not sure that was his fundamental reason - but will leave that to historians of science), that is not my reason. We lack a fully quantum theory of gravity below the Plank scale and the Wilsonian view that QFT can emerge from pretty much anything at very high energies (again thought to be about the Plank scale). Physicists would like all theories to be more than just effective; I know I would. This in no way, of course, detracts from the magnificent achievement of the Standard Model or of Wilsons insights.

Thanks
Bill

 

[Lord Jestocost]

Murray Gell-Mann on Einstein's objections to quantum mechanics regarding "completeness" (in “The Quark and the Jaguar”, chapter: “Quantum Mechanics and flapdoodle”):

“What Einstein required was roughly the following. If, by means of a certain measurement, the value of a particular quantity $Q$ could be predicted with certainty, and if by an alternative, quite different measurement, the value of another quantity $R$ could be predicted with certainty, then, according to the notion of completeness, one should be able to assign exact values simultaneously to both of the quantities $Q$ and $R$.”

Einstein was on the wrong "classical" path, nothing more.

 

[RUTA]

And in spite of his paper, the sequence of conferences has continued. Clearly his stance didn't solve the problem!

He's advocating a principle approach to QM a la SR as a possible way to stop the endless "constructive efforts" (to quote Einstein). We don't have a consensus principle account of QM either, so the conferences continue!

 

[EPR]

The determinism of macroscopic observables is of course indeed emergent.

You are saying this in passing as if it was something trivial.
I don't understand how this fact doesn't seem to concern or bother you, since it's the groundwork of everything that we ever claim to comprehend about the reality. It appears we have build a castle in the sky - all knowledge about anything turns out to be provisional, conditional and with no foundation. You have to shrug off all acquired knowledge so far as emergent, since determinism itself is emergent.
This philosophy ends in the now popular statements describing reality merely as "observable facts" and science as "statements about the observable facts". You say you don't like Bohr but this as Bohr as it gets imo.
You have not resolved the basic paradox, and judging from other similar threads, you seem to be of the opinion that the paradox is exactly as it should be and that it's the natural state of affairs. Hence, no further explanation needed

 

[gentzen]

I don't think gravity needs to be quantized at all. Mostly because of my inability to conceive of a coherent superposition of two different spacetime geometries. Gravity is just too different an animal.

People like Freeman Dyson and Sean Carroll managed to convince me too that it might not be necessary to quantize gravity. I was less convinced by Penrose. But when I later understood the explanation by Dieter Zeh why gravity must be quantized, my reaction was like yours that "my inability to conceive of a coherent superposition of two different spacetime geometries" led me to still not believe it. But I still didn't find Penrose convincing, and had no better proposal/idea either. Later Freeman Dyson basically convinced my with your argument that "Gravity is just too different an animal."
Did you came to your conclusion all by yourself, or did you also get convinced by the opinions or arguments of other people?

 

[PeterDonis]

it might not be necessary to quantize gravity.

It might not, but "might" is not the same as "will". The plain fact is that nobody knows for sure at this point whether or not gravity is quantized. There are plausible arguments on both sides, which just means there's a lot we don't yet understand.

In such a situation, one should not come to any conclusion. One should wait for more information.

 

[dextercioby]

It might not, but "might" is not the same as "will". The plain fact is that nobody knows for sure at this point whether or not gravity is quantized. There are plausible arguments on both sides, which just means there's a lot we don't yet understand.

In such a situation, one should not come to any conclusion. One should wait for more information.

If one comes up with a plausible and mathematically meaningful quantization of gravity (plain GR, or Einstein-Cartan generalization), but this theory makes predictions outside of any human (for the hundreds or thousands of years we will inhabit this planet or any other) experimental check possibility, how would you assert "gravity is quantized" or the opposite?

 

[PeterDonis]

If one comes up with a plausible and mathematically meaningful quantization of gravity (plain GR, or Einstein-Cartan generalization), but this theory makes predictions outside of any human (for the hundreds or thousands of years we will inhabit this planet or any other) experimental check possibility

Isn't that basically the situation we're in now with string theory?

how would you assert "gravity is quantized" or the opposite?

You couldn't. You'd be in the same position I described in my previous post: waiting for more information.

 

[dextercioby]

That's right with string theory, yet nobody said/says that "string theory quantized gravity in a satisfactory manner". So one is not necessarily (hopefully still) searching for a theory of quantum gravity, but perhaps for an illusive ToE (a novel theory which "radiates" to QFT/SM and GR in a mathematically sound manner).

 

[PeterDonis]

nobody said/says that "string theory quantized gravity in a satisfactory manner"

Really? I thought the fact that a massless spin-2 field automatically pops out as one of the low energy string modes was one of the selling points.

 

[dextercioby]

Then I am misinformed, which means that "gravity is quantized through (exact description of) string theory", so the subject is closed. However, is everybody in agreement? Carlo Rovelli, for example.

(Sorry for partially derailing the subject of discussion)

 

[PeterDonis]

is everybody in agreement?

I don't think so. But even if they were, it would still only be about a theoretical speculation, which would not be testable by humans any time in the foreseeable future.

 

[PeterDonis]

one is not necessarily (hopefully still) searching for a theory of quantum gravity, but perhaps for an illusive ToE (a novel theory which "radiates" to QFT/SM and GR in a mathematically sound manner)

As far as I know, all of the quantum gravity alternatives that are currently being investigated are both of these: a theory that quantizes gravity, and a theory that has our current QFT/SM and our current GR as approximations in appropriate limits.

 

[gentzen]

As far as I know, all of the quantum gravity alternatives that are currently being investigated are both of these: a theory that quantizes gravity, and a theory that has our current QFT/SM and our current GR as approximations in appropriate limits.

I doubt that it was shown for any of the alternatives that are currently being investigated that it has SM as approximation in appropriate limits. QFT yes, but SM no. Even for GR as we currently know it (which might not be fully correct), I would not be so sure whether it really occurs. A massless spin-2 field yes, but not the concrete GR with small positive cosmological constant.

 

[PeterDonis]

I doubt that it was shown for any of the alternatives that are currently being investigated that it has SM as approximation in appropriate limits.

It hasn't been. But the proponents of each alternative are optimistic that it eventually will be.

Even for GR as we currently know it (which might not be fully correct), I would not be so sure whether it really occurs. A massless spin-2 field yes, but not the concrete GR with small positive cosmological constant.

The massless spin-2 field part is known to have the Einstein Field Equation as its classical limit.

The small positive cosmological constant is currently an issue, yes, since no one understands why its value should be nonzero but so small, and AFAIK none of the quantum gravity alternatives have any explanation for that.

 

[vanhees71]

That's similar to the old question, how the (now around 25 or so) free parameters of the Standard Model can be explained. That's another question unsolved today. All we know is that for the energies available for observation (and with the so far applied detectors!) the Standard Model is a very good description, but we don't know how to get the 25 parameters from some theoretical principle (maybe another hitherto unknown symmetry?). We just have to fit them to experimental data. That looks indeed like a typical "effective theory".

The same holds for the cosmological constant if interpreted as belonging to the right-hand side of the EFE, i.e., as part of the energy-momentum tensor of "matter and radiation". Within the Standard Model we need an amazing fine tuning to adjust it to the observed (?) value. Many if not most physicists are not very satisfied with this.

The trouble, of course, is that we don't have very clear signals for "physics beyond the Standard Model". Those tend to pop up at any of the big summer HEP conference and then go away with more statistics or more accurate measurements at the next conference. Also none of the speculative theories beyond the Standard Model have been successfully tested yet, including SUSY, but maybe there is some hope by just checking with a different kind of detectors.

The worst is of course gravity, because it's so weak. We can only test it with very massive macroscopic (astronomical) bodies, and there it's pretty hopeless to resolve some quantum effects. As discussed above, it might also be a question, whether there's a need for quantizing gravity at all. On the other hand there are the unavoidable singularities, and singularities are somehow always a hint that something hasn't been completetely understood yet. The big hope then is that, as in QED, the quantization of the theory somehow resolves the notorious "point-particle issue" aka singularities in the solutions of the classical field theory.

 

[A. Neumaier]

Quantum theory is also not completely acausal. The probabilities develop with time by the known dynamical laws.

No, only the probability amplitudes!

 

[vanhees71]

If the probability amplitudes develop causally then also any quantity calculated from them develop causally.

 

[A. Neumaier]

If the probability amplitudes develop causally then also any quantity calculated from them develop causally.

They can be computed from something that develops causally. This is different from themselves developing causally.

 

[vanhees71]

That doesn't make sense to me. We are lost in semantics. I think we should pause the discussion a little to cool it down.

 

[A. Neumaier]

That doesn't make sense to me. We are lost in semantics.

Then state your notion of causality. My notion of causality is:
x develops causally if x(t) is causally determined by x(s) for s<t. Not if x(t) is determined from y(t) for y<t.

 

[vanhees71]

If $x(t)$ is determined by $x(s)$ for $s<t$ then also $f[x(t)]$ is determined for any $t>s$.

 

[A. Neumaier]

If $x(t)$ is determined by $x(s)$ for $s<t$ then also $f[x(t)]$ is determined for any $t>s$.

But it is determined by x(s), not by f[x(s)]).

With your notion of determined, everything is determined. For any x(t) is determined by z(s):=x(s+1) for s=t-1<t.

 

[Lynch101]

It describes the observable properties of an individual quantum system, particularly the probabilities of observables that don't take determined value by the preparation. You may question that this is a complete description, but there's no hint that it's not complete.

It may be that it is in principle impossible to describe the quantum properties of an individual quantum system, or even meanigless to talk about them. So, this need not be a deficiency of this interpretation.

There is a distinction that can be drawn between the following questions:
1) Is QM a complete theory?
2) Is QM the most complete theory possible?
3) Does QM give us a complete description of physical reality?

It's possible to answer 'Yes' to questions 1 & 2 and 'No' to question 3.

EPR were asking question 3 and they set out the following as a necessary requirement for a theory to be considered 'complete':
Every element of the physical reality must have a counterpart in the physical theory.

If the mathematics of QM only makes predictions about the outcomes of ensembles of experiments (and possibly even individual experiments) or it only describes the properties of the system upon interaction with a measurement device, then, by definition, it does not give a complete description of reality, since it doesn't describe the system prior to measurement.

 

[Interested_observer]

There is a distinction that can be drawn between the following questions:
1) Is QM a complete theory?
2) Is QM the most complete theory possible?
3) Does QM give us a complete description of physical reality?

It's possible to answer 'Yes' to questions 1 & 2 and 'No' to question 3.

EPR were asking question 3 and they set out the following as a necessary requirement for a theory to be considered 'complete':
Every element of the physical reality must have a counterpart in the physical theory.

If the mathematics of QM only makes predictions about the outcomes of ensembles of experiments (and possibly even individual experiments) or it only describes the properties of the system upon interaction with a measurement device, then, by definition, it does not give a complete description of reality, since it doesn't describe the system prior to measurement.

I suggest that until we have a theory that predicts exactly where an individual photon will hit the screen in a 2-slit experiment, and the path of an individual gas atom in a confined space, and why one atom of uranium decays and not the one next to it, etc., we will not have a complete physical theory of reality. My guess is that this will not happen. But we should keep looking, for we will undoubtedly learn much from the search.

 

[AlexCaledin]

But why any "reality" has to be "physical"? It might well only use some physics to be better ordered to a certain extent and no more.

 

[vanhees71]

We are back at the problem that the word "reality" is useless, because nobody knows the concise scientific meaning of it anymore. You have to define, what you mean by the word "reality" applied in the context of physics.

By chance, I just stumble over a nice AJP paper by Fuchs. If expressed in this way, I'd even could soon become a qubist ;-)):

https://arxiv.org/abs/1311.5253
https://doi.org/10.1119/1.4874855

 

[A. Neumaier]

But why any "reality" has to be "physical"? It might well only use some physics to be better ordered to a certain extent and no more.

Because physical reality is what we actually observe, with or without measurements.

 

[AlexCaledin]

physical reality is what we actually observe

- then physical = actual? why then use the two words with identical meaning?

 

[gentzen]

- then physical = actual? why then use the two words with identical meaning?

He also wrote "observe". However, I liked your "original" comment, but others interpreted "physical" different from how I read it. I read it in the sense of our physical theories with their natural laws.

 

[A. Neumaier]

- then physical = actual? why then use the two words with identical meaning?

They are not identical. Because physical refers to substance, i.e., what actually exists, while actual is the contrast to imagined.

 

[Interested_observer]

We are back at the problem that the word "reality" is useless, because nobody knows the concise scientific meaning of it anymore. You have to define, what you mean by the word "reality" applied in the context of physics.

By chance, I just stumble over a nice AJP paper by Fuchs. If expressed in this way, I'd even could soon become a qubist ;-)):

https://arxiv.org/abs/1311.5253
https://doi.org/10.1119/1.4874855

From the 1st paper cited above:

"Why, then, do many people wrongly claim that quantum mechanics is nonlocal? They do so by denying at least one of three fundamental precepts of QBism: (1) A measurement outcome does not preexist the measurement. An outcome is created for the agent who takes the measurement action only when it enters the experience of that agent. The outcome of the measurement is that experience. Experiences do not exist prior to being experienced."

This "fundamental precept" seems to me to be more "hand-waving" (as A. Neumaier fondly puts it) in order to sweep the "measurement problem" under the rug.

 

[vanhees71]

Well, but I think it's the essence to cure the confusion on locality (in the sense of relativistic QFT, where it means microcausality and excludes faster-than-light causal effects/interactions by construction) and what's often called "nonlocality" but what's really meant is "inseparability" in Einstein's sense, and this doesn't give rise to violations of causality, because it describes strong correlations between far-distant observations on entangled parts of a quantum system. The correlations are, when interpreted in the sense of Fuchs's "qbism", due to the preparation in the entangled state.

I've another quibble with Fuchs's presentation of qbism: My problem is the claim that the probabilities and the quantum state are purely subjective. That's also not true, because operationally a quantum state is not defined by some agent's knowledge but by a "preparation procedure". If you have a complete preparation procedure, i.e., if you determine a complete set of compatible observables the state is uniquely determined to be the corresponding pure state.

It's of course true that there is a general problem how to choose the probabilities of a probabilistic theory given incomplete information. There are different attempts. The one that seems most successful in physics is the principle of maximum entropy with von Neumann-Shannon-Jaynes entropy in quantum theory.

 

[Lynch101]

I suggest that until we have a theory that predicts exactly where an individual photon will hit the screen in a 2-slit experiment, and the path of an individual gas atom in a confined space, and why one atom of uranium decays and not the one next to it, etc., we will not have a complete physical theory of reality. My guess is that this will not happen. But we should keep looking, for we will undoubtedly learn much from the search.

It is possible that we cannot ever have a complete description of physical reality. Some of the no-go theorems of QM actually point to this, don't they?

 

[Lynch101]

We are back at the problem that the word "reality" is useless, because nobody knows the concise scientific meaning of it anymore. You have to define, what you mean by the word "reality" applied in the context of physics.

By chance, I just stumble over a nice AJP paper by Fuchs. If expressed in this way, I'd even could soon become a qubist ;-)):

https://arxiv.org/abs/1311.5253
https://doi.org/10.1119/1.4874855

If we're talking about a complete description of [physical] reality we are essentially just talking about a complete description of the universe, or the parts of the universe under consideration.

Some interpretations might propose complete descriptions of the universe but which cannot be verified by observation as there may be a limit to how far we can probe nature. However, some interpretations rule themselves out of being complete, by definition.

We can make certain deductions beyond the measurement results by applying if this-then that reasoning.

 

[Lynch101]

Because physical reality is what we actually observe, with or without measurements.

Does the term 'observe', here, imply measurement?

 

[A. Neumaier]

Does the term 'observe', here, imply measurement?

No, but it includes it. We observed long before we were able to measure.

 

[Lynch101]

No, but it includes it. We observed long before we were able to measure.

Does it imply then that physical reality is limited to what we can observe and/or measure?

 

[A. Neumaier]

Does it imply then that physical reality is limited to what we can observe and/or measure?

No. It must include what we can observe and/or measure, but to count as scientific it must also be defined in an objective, observer-independent way.

 

[Lynch101]

No. It must include what we can observe and/or measure, but to count as scientific it must also be defined in an objective, observer-independent way.

Thanks for the clarification.

 

[A. Neumaier]

By chance, I just stumble over a nice AJP paper by Fuchs. If expressed in this way, I'd even could soon become a qubist ;-)):

https://arxiv.org/abs/1311.5253
https://doi.org/10.1119/1.4874855

? What is nice about this paper?
It is essentially devoid of physics, replacing it by a set of should's for the personal beliefs of individual agents, without telling how the agents come to a mutual, objective understanding of the physical world.

 

[facenian]

Well, but I think it's the essence to cure the confusion on locality (in the sense of relativistic QFT, where it means microcausality and excludes faster-than-light causal effects/interactions by construction) and what's often called "nonlocality" but what's really meant is "inseparability" in Einstein's sense, and this doesn't give rise to violations of causality, because it describes strong correlations between far-distant observations on entangled parts of a quantum system. The correlations are, when interpreted in the sense of Fuchs's "qbism", due to the preparation in the entangled state.

I am very suspicious that a theory based on quantum mechanics could be local by construction. It seems more a wishful declaration of how we want the theory to be. On the other hand, inseparability would be another word for nonlocality.
However, I believe there is an unambiguous way of considering the theory local by a more careful definition of locality.

 

[A. Neumaier]

there is an unambiguous way of considering the theory local by a more careful definition of locality.

See the book 'Local quantum physics' by Rudolf Haag. This is the most far reaching notion of locality, and holds without known exception.

Bell nonlocality is not a statement that the dynamics of microsystems is nonlocal. It is only a statement that if modeled by a classical hidden variable theory, quantum mechanics would be nonlocal. Indeed, Bell's assumption are completely independent of quantum mechanics!

 

[vanhees71]

I am very suspicious that a theory based on quantum mechanics could be local by construction. It seems more a wishful declaration of how we want the theory to be. On the other hand, inseparability would be another word for nonlocality.
However, I believe there is an unambiguous way of considering the theory local by a more careful definition of locality.

In relativistic QFT a very important defining ingredient is locality aka microcausality of local observables, i.e., the Hamiltonian density must commute with all local observables at spacelike separation of the arguments of the corresponding operators. That's the only solid definition of locality I know of, and it's sufficient to exclude spooky actions at a distance.