What exactly are interpretations of quantum mechanics?

[member 529879]

What exactly are interpretations of quantum mechanics? Is only one of them correct?

 

[atyy]

In quantum mechanics, a classical observer and a quantum system are needed. It doesn't seem possible to describe the entire universe as a quantum system without an observer. How can this be, since the observer himself can be described as a quantum system observed by another observer?

Should there be a theory in which one doesn't need an observer who stands apart from the system? What sort of theories can reproduce the successful predictions of quantum mechanics and get rid of the observer who stands apart? The various possible answers to these questions are the different interpretations of quantum mechanics.

Some jargon: these questions are generally known as the "measurement problem".

 

[StevieTNZ]

I would say there would be only one correct interpretation of QM -- which one, out of the many available now, is very hard to say.

I would wait for macroscopic superposition to be tested, as well as the Leggett-Garg inequality (as well as experiments similar to the thought experiment in 'Sneaking a Look at God's Cards' pgs 373-376), to see what is left on the table. Also if we can find testable predictions that differ between Bohmian Mechanics and standard Quantum Mechanics, that would be a bonus.

 

[atyy]

I would say there would be only one correct interpretation of QM -- which one, out of the many available now, is very hard to say.

I would wait for macroscopic superposition to be tested, as well as the Leggett-Garg inequality (as well as experiments similar to the thought experiment in 'Sneaking a Look at God's Cards' pgs 373-376), to see what is left on the table. Also if we can find testable predictions that differ between Bohmian Mechanics and standard Quantum Mechanics, that would be a bonus.

Surely, there have to be at least 2! Unless we count Copenhagen as not an interpretation?

Or could it be that Copenhagen, and Copenhagen alone is correct - no hidden variables, no retrocausation, no many-worlds?

 

[member 529879]

What I don't understand is how some physicists say that the many worlds interpreatation is their favorite. Is there any evidence to support it? Why does it seem like they are just picking which one they like rather than which one is most likely to be correct?

 

[Nugatory]

What I don't understand is how some physicists say that the many worlds interpreatation is their favorite. Is there any evidence to support it? Why does it seem like they are just picking which one they like rather than which one is most likely to be correct?

All interpretations predict the same results, so there's no way of running an experiment to decide which one is right. Thus, we are free to choose whichever interpretation we find most tasteful - and people are always going to disagree on matters of taste. More than once I've said "De gustibus not dusputandum est", and more than once the mentors have closed a thread because it has degenerated into an argument about whose interpretation is better-looking.

 

[StevieTNZ]

Surely, there have to be at least 2! Unless we count Copenhagen as not an interpretation?

Or could it be that Copenhagen, and Copenhagen alone is correct - no hidden variables, no retrocausation, no many-worlds?

Copenhagen, as I understand, splits the universe into two: one where quantum rules hold, and the macroscopic world where classical mechanics holds. But if we combine a macroscopic apparatus, which is evolving deterministically, with a microscopic system, then by virtue of that the microscopic system being measured has had its measurement result already determined (because the result is shown on the classical system which has what is going to show already determined at the beginning of the universe).

Please correct me if I am wrong in saying Copenhagen splits the world into two realms.

 

[member 529879]

All interpretations predict the same results, so there's no way of running an experiment to decide which one is right. Thus, we are free to choose whichever interpretation we find most tasteful - and people are always going to disagree on matters of taste. More than once I've said "De gustibus not dusputandum est", and more than once the mentors have closed a thread because it has degenerated into an argument about whose interpretation is better-looking.

But how is the idea that there are multiple universes a matter of taste? That sounds like an idea that is objectively right or wrong.

 

[Nugatory]

But how is the idea that there are multiple universes a matter of taste? That sounds like an idea that is objectively right or wrong.

Here are two hypotheses:
1) There is one universe.
2) There are multiple universes, but it is impossible for anything that happens in one universe to ever affect anything that happens or can be observed in any other universe; thus, it is impossible, even in principle, to construct an experiment whose outcome will be different from the outcome predicted by hypothesis #1. Nothing that we ever do will make the other universes show themselves to even the tiniest degree.

How would we ever determine which of these hypotheses is objectively right or wrong? Everything is completely the same whether #1 is right or #2 is right. The only reason for preferring hypothesis #2 is that it avoids an annoying wart in the mathematical formalism... and as a matter of personal taste, some people find the wart less unsightly than the assumption of multiple universes, while other people find the idea of multiple universes more appealing than this ugly wart.

 

[bhobba]

But how is the idea that there are multiple universes a matter of taste? That sounds like an idea that is objectively right or wrong.

The point with all these interpretations is no one has figured out how to experimentally distinguish between them - and that includes many worlds.

That being the case the one you choose is purely a matter of taste.

The value of studying interpretations is it deepens your understanding of what the formalism without an actual interpretation says. This is quite interesting because at first brush you might, for example, think QM isn't deterministic. Yet we have a number of interpretations where it is. QM is actually silent on many things people like to think it says.

Yet another interesting thing is discussion about interpretations of QM is to a large extent a discussion about the meaning of probability:
http://math.ucr.edu/home/baez/bayes.html

Thanks
Bill

 

[atyy]

All interpretations predict the same results, so there's no way of running an experiment to decide which one is right. Thus, we are free to choose whichever interpretation we find most tasteful - and people are always going to disagree on matters of taste. More than once I've said "De gustibus not dusputandum est", and more than once the mentors have closed a thread because it has degenerated into an argument about whose interpretation is better-looking.

But how is the idea that there are multiple universes a matter of taste? That sounds like an idea that is objectively right or wrong.

I don't think they are necessarily matters of taste. Some would say that Many-Worlds is falsifiable, because it predicts no deviations from quantum mechanics, eg. Sean Caroll: "There are other silly objections to EQM, of course. The most popular is probably the complaint that it’s not falsifiable. That truly makes no sense. It’s trivial to falsify EQM — just do an experiment that violates the Schrödinger equation or the principle of superposition, which are the only things the theory assumes. Witness a dynamical collapse, or find a hidden variable." http://www.preposterousuniverse.com...ion-of-quantum-mechanics-is-probably-correct/

Similarly, Bohmian Mechanics predicts deviations from quantum mechanics, in the same way that we expect there to be non-equilibrium deviations from classical statistical mechanics. It's basically as falsifiable as string theory :) Just as there's minute hope for stringy effects like large extra dimensions at the LHC, there are proposals for rather large Bohmian effects: http://arxiv.org/abs/1306.1579.

There are also discussions as to how one might test theories in which collapse is real: http://arxiv.org/abs/1410.0270.

So interpretations is really beyond the standard model, and the Bell theorem is analogous to the Weinberg-Witten theorem.

 

[vanhees71]

Surely, there have to be at least 2! Unless we count Copenhagen as not an interpretation?

Or could it be that Copenhagen, and Copenhagen alone is correct - no hidden variables, no retrocausation, no many-worlds?

If you take the right flavor of Copenhagen (no collapse, i.e., the minimal statistical flavor!) I also think it's the only correct one we have at the moment. Whether or not QT is complete, I don't dare to decide. I tend to say it's incomplete, because if you follow the minimal statistical interpretation, it doesn't make sense to apply the notion of a quantum state to the entire universe, because you can never validate a probabilistic statement on a system which by definition can be realized only once, which for sure is the case for the entire universe.

 

[Ilja]

What exactly are interpretations of quantum mechanics? Is only one of them correct?

Of course. And it is quite uncertain (and even not very probable) that we have already found it.

Which is the correct one, remains an open question, which will be probably answered only when a more fundamental, subquantum theory is found. This subquantum theory would have a quantum limit or quantum approximation, and the interpretation which is the closest one to this limit will be the correct one.

 

[Ilja]

The point with all these interpretations is no one has figured out how to experimentally distinguish between them - and that includes many worlds.

Here I would disagree. Of course, interpretations cannot be experimentally distinguished, but some can be rejected as nonsensical. This is the case of many worlds. SCNR

 

[ephen wilb]

why can't bohmian mechanics pilot wave be guided by true random quantum potential? why does it have to be deterministic? how rules would be broken if you combine BM with true randomness?

 

[bhobba]

why can't bohmian mechanics pilot wave be guided by true random quantum potential? why does it have to be deterministic? how rules would be broken if you combine BM with true randomness?

You could probably develop such a theory. Want to work out the details and post it here?

If you find that hard then maybe that is the answer to your query.

Thanks
Bill

 

[ephen wilb]

You could probably develop such a theory. Want to work out the details and post it here?

If you find that hard then maybe that is the answer to your query.

Thanks
Bill

Why not just consider it random a priori? Quantum mechanic like Demystifier treat it as like classical nut and bolt.. that's what car mechanics do.. this is why he believes determinism rule the universe and we could predict every occurrence for the million of years like the name of his 1000th grandchildren.. but can't we just go beyond that and just accept randomness dictates the pilot wave or quantum potential?

 

[bhobba]

Why not just consider it random a priori?

Because that will not work with BM - the particle is guided deterministically by it. You need a detailed theory with it in from the start and associated equation that gives QM. BTW I have no idea how to do it.

Thanks
Bill

 

[Nugatory]

This would be a good time to remind everyone that original question in this thread was "what are interpretations and is there consensus about which one is correct?". Is the OP still around, and is he sorry he asked?

 

[Ilja]

You could probably develop such a theory. Want to work out the details and post it here?

There is already something similar - Nelsonian stochastics. And, given the recent interest in epistemic models for the wave function, an interesting variant of it developed by A. Caticha -- Entropic Dynamics, Time and Quantum Theory, J. Phys. A44:225303, 2011, arxiv:1005.2357

 

[Ilja]

Thus, we are free to choose whichever interpretation we find most tasteful - and people are always going to disagree on matters of taste. More than once I've said "De gustibus not dusputandum est", and more than once the mentors have closed a thread because it has degenerated into an argument about whose interpretation is better-looking.

We should not forget that there was a time when the atomic hypothesis was only an interpretation. The lessons that it appeared that this "interpretation" was right and the alternatives wrong have not even been forgotten, they were never taken. With Mach's positivism appearing a little earlier, it could have been that the progress of atomic theory would have been stopped, and we would today think about the question if atoms really exist or not - and be banned from physics forums for such unscientific speculations.

The point is that different interpretations are different starting point for future research. Once it is forbidden to discuss interpretations, there simply will be no progress toward a more fundamental theories - or at least this becomes quite probable, given that the main road toward this is closed.

The interpretation which is winning in a situation where discussing interpretations is forbidden is, of course, the most inert one - the one which accepts the math of the theory unquestioned, and does not even think about the possibility that it may be only an approximation. So we play around with sci-fi-mathematics of wormholes and causal loops, but nobody cares about possibilities to reject this because the theory is only an approximation of some different theory, and wormholes and causal loops only applications of the mathematics far beyond the limits of applicability of this approximation.

 

[Nugatory]

The point is that different interpretations are different starting point for future research. Once it is forbidden to discuss interpretations, there simply will be no progress toward a more fundamental theories - or at least this becomes quite probable, given that the main road toward this is closed.

That's a good point, and to some extent it speaks to to the purpose and role of Physics Forums. On the one hand, PF is not a platform for publishing new research in a field, so we can suppress discussion of particular topics without shutting down the progress of science or denying good new ideas a fair chance to be heard.

On the other hand, we want PF to be a good place to come to if you want to understand what are generally recognized to be good ideas in science. Inconclusive and interminable threads on interpretational issues are an uneasy fit with the latter goal; these threads are disproportionately difficult to moderate effectively and we get a fair amount of complaints from members about them - "The usual suspects are just repeating the arguments they made last year and will still be making next year, and it will never be settled".

I'm more tolerant of interpretational threads than some of the other mentors, mainly because I find them interesting (moderating them, not so much - that's a chore!) but after seeing the feedback from members, I am convinced that the "matter of taste" answer is the best starting point for most of the PF membership most of the time... at least given the current state of understanding, which is what PF is about.

I can imagine this situation changing. If PF had been around in 1935 when EPR had been published, we probably would have spent the next three decades trying to damp down discussions of the "incompleteness" of QM on the grounds that it couldn't be settled by experiment, the discussion wasn't going anywhere, and it annoyed a lot of people. But once Bell's theorem landed in the peer-reviewed literature and there was time to digest the implications, the discussion would have moved back in scope for PF. I wouldn't have considered that to be either inconsistent or an acknowledgment of a past error, but rather as an appropriate response to an important new development.

 

[Ilja]

I can imagine this situation changing. If PF had been around in 1935 when EPR had been published, we probably would have spent the next three decades trying to damp down discussions of the "incompleteness" of QM on the grounds that it couldn't be settled by experiment, the discussion wasn't going anywhere, and it annoyed a lot of people. But once Bell's theorem landed in the peer-reviewed literature and there was time to digest the implications, the discussion would have moved back in scope for PF. I wouldn't have considered that to be either inconsistent or an acknowledgment of a past error, but rather as an appropriate response to an important new development.

But Bell's theorem - which can be proven only based on the spacetime interpretation of SR, not in the Lorentz ether interpretation which would not forbid hidden FTL in a preferred ether frame - was obviously not sufficient to remove the ban for discussions about the Lorentz ether, not? And to discuss ether papers which have landed in the peer-reviewed literature remains forbidden too?

 

[Nugatory]

But Bell's theorem - which can be proven only based on the spacetime interpretation of SR, not in the Lorentz ether interpretation which would not forbid hidden FTL in a preferred ether frame - was obviously not sufficient to remove the ban for discussions about the Lorentz ether, not? And to discuss ether papers which have landed in the peer-reviewed literature remains forbidden too?

Bell's theorem was a new insight that was (not in principle, but in practice) subject to experimental test. If anyone ever comes up with a no-go theorem that allows the experimentalists to distinguish Lorentz ether theory from SR and peer-reviewed experimental results are published... I expect that we would consider dropping the ban on ether theory discussions. Of course that's just a hypothetical at the moment.

 

[Ilja]

Bell's theorem was a new insight that was (not in principle, but in practice) subject to experimental test. If anyone ever comes up with a no-go theorem that allows the experimentalists to distinguish Lorentz ether theory from SR and peer-reviewed experimental results are published... I expect that we would consider dropping the ban on ether theory discussions. Of course that's just a hypothetical at the moment.

No, it is not at all hypothetical, Bell's paper was this paper. Because the simple fact is that you cannot prove Bell's theorem in the Lorentz ether - the Lorentz ether has a preferred frame and classical causality as defined by its true time. There is no way to derive Einstein causality in the Lorentz ether, thus, no way to prove Bell's theorem, which requires Einstein causality. But in SR, in the space-time interpretation, it has been proven by Bell.

Given that this difference is quite trivial and IMHO well-known, one cannot publish it, except possibly as a side remark in a paper published for other novelties, like "The violation of Bell’s inequality proves that any realistic interpretation of quantum theory needs a preferred frame" in the published paper http://arxiv.org/abs/gr-qc/0205035
But the simple fact remains - the Lorentz ether is not sufficient to prove Bell's theorem, SR is, Bell's theorem leads to observable consequences and these consequences have been observed.

Of course, to prove it in SR, we need some minimal additional assumptions - or "realism" or simply a variant of causality which includes Reichenbach's principle of common cause. But the difference remains - adding the same additional assumptions to the Lorentz ether does not allow to prove Bell's theorem.

Your rejection of Lorentz ether discussions is also unfair if one compares this with the fact that discussing string theory, supersymmetry, and other things which are purely speculative and not supported by any empirical evidence can be discussed. String theory has not even made predictions to make experiments about this, but for allowing discussions of ether theories you require not only that ether theories are published which make predictions, which are, in principle, open to experimental tests, but you require even that these experiments have been already done and published. Sorry, but this is the moment where it is already too late to discuss them.

 

[Point Conception]

That's a good point, and to some extent it speaks to to the purpose and role of Physics Forums. On the one hand, PF is not a platform for publishing new research in a field, so we can suppress discussion of particular topics without shutting down the progress of science or denying good new ideas a fair chance to be heard.

On the other hand, we want PF to be a good place to come to if you want to understand what are generally recognized to be good ideas in science. Inconclusive and interminable threads on interpretational issues are an uneasy fit with the latter goal; these threads are disproportionately difficult to moderate effectively and we get a fair amount of complaints from members about them - "The usual suspects are just repeating the arguments they made last year and will still be making next year, and it will never be settled".

Isnt the above the best case for favoring the Ensemble Interpretation ( Statistical ) ( Minimalist ) Since these interpretations do not have any Alice in Wonderland features attached like non locality ,many worlds and retro causation.

 

[bhobba]

Isnt the above the best case for favoring the Ensemble Interpretation ( Statistical ) ( Minimalist ) Since these interpretations do not have any Alice in Wonderland features attached like non locality ,many worlds and retro causation.

No.

Science is about correspondence with experiment - not a particular view of the 'weirdness' of a theory.

Here I depart from Ilja - I don't hold to MW, but having investigated it, it is logically consistent IMHO, and hence valid, same with time symmetric interpretations.

The true value of interpretations IMHO is it allows a better understanding of the QM formalism - many things people think it says it really is silent on.

I hold to the ensemble ignorance interpretation personally (it's a slight variation of the ensemble), but that means nothing - except perhaps as a comment on my views of physics.

Thanks
Bill

 

[Ilja]

Isnt the above the best case for favoring the Ensemble Interpretation ( Statistical ) ( Minimalist ) Since these interpretations do not have any Alice in Wonderland features attached like non locality ,many worlds and retro causation.

No, you should not mingle a simple refusal to make any statements about reality and causality with the absence of problems about reality and causality. "Shut up and calculate" does not solve problems, it ignores them - the ostrich or head-in-the-sand way of solving them. BTW, in some interpretation of SR we are forbidden to speak about there would be no such problems of compatibility with relativity.

 

[Ilja]

I don't hold to MW, but having investigated it, it is logically consistent IMHO, and hence valid, same with time symmetric interpretations.

I don't consider MW as logically consistent, in its derivation of the Born rule it presupposes common sense, but I see no way to justify the applicability of common sense in MW.

 

[atyy]

Isnt the above the best case for favoring the Ensemble Interpretation ( Statistical ) ( Minimalist ) Since these interpretations do not have any Alice in Wonderland features attached like non locality ,many worlds and retro causation.

Let's just say that as great a believer in the minimal interpretation as Dirac did not agree. One can use it while acknowledging that it has problems. Dirac presumed that quantum mechanics is incomplete because of the problems of the minimal interpretation.

His comments are in his essay "The Evolution of the Physicist's Picture of Nature".
http://blogs.scientificamerican.com/guest-blog/the-evolution-of-the-physicists-picture-of-nature/