Quantum Interpretations history

[peter0302]

Vanesch, thanks to your explanations, this view seems to make perfect sense in MWI. :)

I just still can't see how it's consistent with any sensible single-world interpretation. You have a choice between assuming the photons are magically avoiding the grid, or assuming the grid is magically re-directing / re-emitting photons that it otherwise should be blocking. (Or just giving up a la CI). Of these, the former is more palletable, but I don't really love either of them.

You might be making an MWI believer out of me...

Anyway thanks again. Much appreciated.

 

[vanesch]

Vanesch, thanks to your explanations, this view seems to make perfect sense in MWI. :)

I just still can't see how it's consistent with any sensible single-world interpretation. You have a choice between assuming the photons are magically avoiding the grid, or assuming the grid is magically re-directing / re-emitting photons that it otherwise should be blocking. (Or just giving up a la CI). Of these, the former is more palletable, but I don't really love either of them.

You might be making an MWI believer out of me...

I'm not really an MWI *believer* you know. I think we don't have all the pieces of the puzzle yet to raise it to the ultimate goal of "the meaning of life, the universe and everything". But I like the MWI view on quantum mechanics: I find that it gives the "explanation" that is most in sinc with the formalism, and is most of the time, once understood, rather crystal-clear (no ambiguity left in the interpretation of a setup). The only problem is of course that it is quite crazy! However, its crazyness can be dealt with if you delve into some philosophy. That is to say, one can have a logically sound reasoning which helps you overcome a *logical* objection to the crazy world view of MWI. That doesn't avoid that intuitively you say: "bollocks!".

However, I have to say that in many quantum-mechanical situations where one delves into "metaphysical" considerations about certain aspects, sticking to an MWI view on it "for the sake of argument" helps me to get a quite clear view on it. Especially EPR experiments and delayed choice quantum eraser experiments become "totally logical" in this view.

The intuitively most acceptable view to me is BM. However, I find it formally a monstruousity, because of the explicit impossibility of formulating its machinery in a lorentz-invariant way (which is NOT the case for the pure wavefunction part). That said, when you look closer at BM, its intuitiveness suffers a bit. I would say that if we never had relativity, and we were stuck with NR QM, then BM is really very nice. But I hate its clash with the spirit of relativity. Also, many situations are intuitively difficult to analyse in BM - at least for me. I have first to switch to an MWI kind of picture, understand it from that PoV, and THEN I'm sometimes able to understand the BM picture. But getting an intuition for BM as such is not easy!

Anyway thanks again. Much appreciated.

thanks

 

[colorSpace]

The wave functions are of course common among all interpretations. The difference arises when there is a change from superposition and/or interference to a specific outcome relative to an observer or measurement device, meaning: at which point will the parts of the wave function stop to interfere. It seems the discussion of this experiment has avoided this crucial question by looking only at the wave function.

But it doesn't take MWI to look only at the wave function, since, again, it is common to all interpretations. Or?

 

[colorSpace]

[Continued]

Also, one shouldn't overlook the potential of interpretations such as BM to show possibilities of explaining specific outcomes in a more precise way than "randomly", at least for specific situations. Perhaps an analysis of particle dynamics will show that it is possible to create situations in which a specific outcome is predictable. CI and MWI have in common that they exclude such a possibility in regards to a specific observer, since from a subjective point of view randomness is built-in, and seemingly unavoidable. So they will discourage research into looking for models which might be more precise (at least in some situations) than the wave function.

 

[vanesch]

But it doesn't take MWI to look only at the wave function, since, again, it is common to all interpretations. Or?

Eh, MWI *does* only look at the wavefunction. It declares (coarse grained) quantum states of a body as "subjectively experienced" (like classical brain configurations are to be "experienced") and the probability to be experienced by you is given by the amplitude squared of the product state in which this body state appears.
The difference with other interpretations is that it doesn't ADD anything to the wavefunction (such as BM, or TI), and that it doesn't fiddle with it or declares it non applicable (CI) to certain systems. Wavefunction all the way!

So if, in the wavefunction, two terms devellop (though normal unitary evolution), one which contains a body state which is close to a classical body state with a brain that sees a red light flashing, and another term with a body state which is close to a classical body state with a brain that sees a green light flashing, then in MWI you say that these two terms are two branches, one with an observer seeing a red light, and the other with an observer seeing a green light flashing, and that the probability for YOU to be/experience that first or that second observer state is given by the amplitude squared of each of the two terms.

Now, you can think it to need a kind of miracle to develop such classically-looking terms, but in fact, decoherence shows us that this happens in fact quite fast and that these terms are then stable (that is, they become classically tracable throughout the further evolution under U).

 

[colorSpace]

Eh, MWI *does* only look at the wavefunction.

What I said is that you don't need MWI in order to look at the wave function.

 

[vanesch]

What I said is that you don't need MWI in order to look at the wave function.

Sure ! The only thing you need to avoid is to do OTHER things than to look at the wavefunction, such as blah blah about through which slit the photon came, and whether or not it "behaves as a particle or a wave" and so on!

The analysis of the Afshar experiment I presented here is - as I tried to point out - independent of interpretational issues, but is probably easier to follow in an MWI mindset where you are not *perturbed* by other, often misled considerations, such as "did it hit the wires or not".

The nice thing about this experiment is that it helps you to understand better the intimate workings of each interpretation, which is in any case always based on the wavefunction. It might chase erroneous views on certain interpretations. For MWI, this is very easy, as once we have the wavefunction, things are over! That's why I like that view, btw.

 

[colorSpace]

Sure ! The only thing you need to avoid is to do OTHER things than to look at the wavefunction, such as blah blah about through which slit the photon came, and whether or not it "behaves as a particle or a wave" and so on!

My understanding is that "behaving as a wave" means there is interference seen by a specific observer; "behaving as a particle" means there isn't (at least not seen by a specific observer). That difference must also exist in MWI.

 

[vanesch]

My understanding is that "behaving as a wave" means there is interference seen by a specific observer; "behaving as a particle" means there isn't (at least not seen by a specific observer). That difference must also exist in MWI.

But these statements "behaving as a wave" or "behaving as a particle" are relics of a time when one didn't understand much about quantum dynamics, and one wanted to attach it to classical notions. The *actual* difference is: when does a superposition become equivalent to a statistical ensemble ? That is:

when can we consider something like u |a> + v|b> as:
|u|^2 chance to have |a> and |v|^2 chance to have |b> ?

Answer: when we do an irreversible measurement. Period.

Now, if the measurement is in the position representation, then this means:
when does a "superposition of position states" (a wave) can be seen as a statistical mixture of different positions (possible "particle" positions) ?
Answer: when we do a position measurement !

Now, and this is where much confusion rules, especially in "which way" experiments and so on, SOMETIMES we get the right results out, even if we PREMATURELY consider a superposition of position states (a wave) as "already a statistical mixture". So, though this is IN PRINCIPLE wrong, we might get out nevertheless the right result this way. If this happens, we say that we don't see any interference. The point is, we would have obtained the right result also of course if we would have followed the *correct* procedure: namely: keep superpositions as such, until we do an irreversible measurement.

Why do we sometimes switch "too early" to the statistical mixture ? First of all, of course, because it is closer to our classical intuition. But second, because especially the CI introduces an ambiguity, by this "switching to classical when a measurement is performed" clause. Indeed, this should read: an IRREVERSIBLE measurement (for instance: with a result printed on paper and so on). So some people extend that concept and call everything that interacts with the system under study "a measurement". Like the grid which performs "a measurement" on the "positions" of the photons - although no result is printed out! As I said, sometimes one can get away with this: the results come out the same. But sometimes, not. In any case, it is an error of principle to do the superposition -> ensemble transition BEFORE an irreversible measurement is done.

MWI has the advantage of clearly not introducing a different concept of interaction for a "measurement" or for an "interaction", and so one is not tempted to make too early a transition to a statistical ensemble. But even in CI, if interpreted correctly, one shouldn't have done that!

 

[colorSpace]

Answer: when we do an irreversible measurement. Period.

MWI has the advantage of clearly not introducing a different concept of interaction for a "measurement" or for an "interaction", and so one is not tempted to make too early a transition to a statistical ensemble. But even in CI, if interpreted correctly, one shouldn't have done that!

If "irreversible measurement" means 'a measurement which leaves a trace of information' (roughly) then I guess this puts MWI and "modern" CI back on the same page in this specific regard.

 

[vanesch]

If "irreversible measurement" means 'a measurement which leaves a trace of information' (roughly) then I guess this puts MWI and "modern" CI back on the same page in this specific regard.

Yes! That's why it is kind of silly to try to distinguish between them experimentally!
Caveat: at least, that is how *I* understand CI, because you never find two identical accounts of what is exactly CI...

 

[Hans de Vries]

With all the different quantum interpretations out there, which is your favourite " ?

You asked me for my favorite interpretation in a PM, Now, since I use
my own interpretation I'll post the elementary ideas here.


Just to give it a name: Sea-of-particle interpretation


Q: What is a particle?
A: A single surplus/absent particle in the see-of-particles


Q: Why can a particle turn up everywhere in the wave-function?
A: First this:

00_000000000000
000000000000_00

How long does it take for the 'gap' to go from the left to the right?
In principle not more as it takes the ten '0' take to go one step to
the left. There is in principle no SR limitation for the 'gap' to be either
at the left or the right.


Q: What is the wave function?
A: A single absent/surplus particle will distort the whole 'grid'
Instead of a single gap of say '40' there will be many small gaps
like for instance:

1-2-4-6-7-7-6-4-2-1

This is the wave function. The bigger the gap, the higher the chance
of detection. No single gap corresponds to the single deficit particle.
No single 'sea-of-particles' particle corresponds to the one surplus
particle.


Q: What is the detection of a particle?
A: The removal of the single surplus / deficit particle from the sea-of-
gates. Fixed and localized for instance to an atom.


Q: What happens with the wave function?
A: Without the single surplus/absent particle there is nothing anymore
to 'distort' the grid of the sea of particles. The wider wave-function will
die out. The wave-function will become fixed /localized to an atom for
instance.


Q: Shouldn't the wave-function disappear instantaneously? A remaining
gap could be detected as a particle which would violate unitarity.

A: This describes particle-pair creation, it leaves an extra surplus particle.
This happens all the time, temporary, because of energy conservation.
Single particle unitarity does not exist.


Q: What is the physical interaction?
A: The physical interaction is generally with the whole wave-function.

example: In Afshar's experiment there are more than 10^23 interactions
(~Avogadro's number) between the single photon's wave-function and
the dielectric molecules of the lens. These are real physical interactions.
Atoms are displaced back and forward. They detect the wave-function
but they do not -detect- the photon in the sense that they remove a
single surplus/absent particle from the see-of-particles.



For so far, In the hope to give you something which is more acceptable... :^)

Regards, Hans

 

[akhmeteli]

You asked me for my favorite interpretation in a PM, Now, since I use
my own interpretation I'll post the elementary ideas here.


Just to give it a name: Sea-of-particle interpretation


Q: What is a particle?
A: A single surplus/absent particle in the see-of-particles

Interesting. Actually, I discussed a similar interpretation in http://arxiv.org/abs/quant-ph/0509044 .
Could you give references to your (or other people's) publications on this approach?

 

[vanesch]

You asked me for my favorite interpretation in a PM, Now, since I use
my own interpretation I'll post the elementary ideas here.


Just to give it a name: Sea-of-particle interpretation

It is difficult to imagine how this is in 1-1 correspondence with the quantum formalism. After all, what you describe is almost litterally the dynamics of crystal defects in materials (Frenkel defects and all that). For instance, how does one get simply, say, the helium spectrum out of such a picture ?

 

[Hans de Vries]

For instance, how does one get simply, say, the helium spectrum out of such a picture ?

That shouldn't be so difficult.

The atomic spectra arise when the Laplacian is combined with a central potential
and the Laplacian occurs almost everywhere where you have particle assembles
since it's part of the classical wave equation.


Regards, Hans

 

[Hans de Vries]

Could you give references to your (or other people's) publications on this approach?

I don't have anything serious written up in Latex yet. it's a developing
idea and I would like to fill in more of the details.



Regards, Hans

 

[Count Iblis]

Many worlds interpretation is my favorite, but I'm not 100% conviced.
If you take unitary time evolution serious, then some of the intuitive ideas about quantum randomness etc. are wrong. E.g., unitary time evolution implies that there exists an observable for observing the position of a particle some time into the future. If you could somehow measure that observable, you could tell exactly where the particle will be in the future.

And, of course, you can make this as crazy as you like. There exists an observable for measuring the contents of your newspaper in the year 2100. The universe will then effective collapse into a specially prepared state which will evolve to some state consistent with the contents of the newspaper.

 

[vanesch]

That shouldn't be so difficult.

The atomic spectra arise when the Laplacian is combined with a central potential
and the Laplacian occurs almost everywhere where you have particle assembles
since it's part of the classical wave equation.

Let us not forget that (hence my *helium* and not *hydrogen*) a significant part of the spectrum of helium is determined by configuration interaction, which is due to the entanglement of the two orbitals (that is, is due to the deviation from the true ground state from a Slater determinant). Very many "quantum interpretations" which are in fact naive hidden variable field theories are quite OK as long as one looks at single-particle states, or product states (or Slater determinants for Fermions), but break down when coming to "real quantum mechanics" where actual entanglement takes place (that is, when the wavefunctions are essentially not factorisable). One thinks of course of EPR situations, but there are many more "entanglements" in quantum phenomena, and the one that comes most obviously to mind is the Helium spectrum.

 

[vanesch]

If you take unitary time evolution serious, then some of the intuitive ideas about quantum randomness etc. are wrong. E.g., unitary time evolution implies that there exists an observable for observing the position of a particle some time into the future. If you could somehow measure that observable, you could tell exactly where the particle will be in the future.

This is the quantum analogon of the classical canonical variable change in Hamilton-Jacobi theory, where the "canonical variables" simply become the "initial conditions" which don't evolve in time anymore (up to the angle variable, which has a linear time variation).

However, it is just as irrealistic in MWI as it is in classical physics (except for simple toy systems), or even more so, because in order to be able to construct that observable, you have to know unitary evolution well enough, INCLUDING all "observer" and "environment" interactions.

 

[peter0302]

Since this thread is _still_ alive :) I thought I'd ask another related question, and this is generally directed to Vanesch but anyone who also has an opinion is welcome to jump in.

When distinguishable fermi particles are scattered, the probability of finding one or the other in a particular state is simply what one would expect with classical physics. However, when _indistinguishable_ bose particles are scattered, it is more likely that a bose particle will end up in a paritcular state if the other particles did.

Does this not intuitively support a many-worlds theory? The rationale would be that because the particles are completely indistinguishable, the universes simply do not "split" (because they're indistinguishable!) and consequently the probability amplitude of a paritcular branch of the wavefunction is higher (the universe is "thicker") than it would be if the particles were distinguishable. This is reflected in seeing a higher probability of viewing the bose particles in particular states.

Can a pilot wave or other hidden variable interpretation account for this?

 

[akhmeteli]

It is difficult to imagine how this is in 1-1 correspondence with the quantum formalism. After all, what you describe is almost litterally the dynamics of crystal defects in materials (Frenkel defects and all that). For instance, how does one get simply, say, the helium spectrum out of such a picture ?

Let us not forget that (hence my *helium* and not *hydrogen*) a significant part of the spectrum of helium is determined by configuration interaction, which is due to the entanglement of the two orbitals (that is, is due to the deviation from the true ground state from a Slater determinant). Very many "quantum interpretations" which are in fact naive hidden variable field theories are quite OK as long as one looks at single-particle states, or product states (or Slater determinants for Fermions), but break down when coming to "real quantum mechanics" where actual entanglement takes place (that is, when the wavefunctions are essentially not factorisable). One thinks of course of EPR situations, but there are many more "entanglements" in quantum phenomena, and the one that comes most obviously to mind is the Helium spectrum.

It may well be that no 1-1 correspondence with the quantum formalism can be achieved. However, that would not be the end of the story. Strictly speaking, what we need is not 1-1 correspondence with the formalism, but agreement with the existing body of experimental data. That is, we need agreement with the formalism up to the experimentally achievable accuracy. However, that accuracy can be extremely high, so maybe I am hair-splitting. The reason I emphasize this distinction is that, for example, the results of Barut's self-field electrodynamics seem to emulate those of QED with high accuracy. To the best of my knowledge, it is not clear if Barut's theory is in disagreement with the body of experimental data. Thus, the Dirac sea interpretation (or you may call it "plasma interpretation", or "vacuum polarization interpretation") may be just added to Barut's theory or to some version of that theory.

 

[reilly]

See any E&M text dealing with E&M radiation, Landau and Lifschitz, for example. Diffraction is diffraction is scattering is interference no matter what blocks or impedes the radiation.

As long as the detector is any distance from the slit, there's a probability that the photon detected "came" from the other slit. So, who knows?

Certainty is hard to come by.

Regards,
Reilly Atkinson

But the slits don't cause diffraction like a lens or prism does. The slits merely pick out photons in such a way as to make it ambiguous which slit they went through. But I still say they only went through one slit or another, as evidenced by the fact that if you put detectors immediately beyond the slits before any interference pattern can show up, you only see one detector or the other go off.

 

[reilly]

Think for a moment about a volt meter. To be sure, we know how such a voltmeter works. Even with AC, we can assert that a voltage measurement will pick out one and only one value at any time whenever it is used -- up to the limits of measurement errors. Why just one value? Who knows? But, not many seem to worry about this basic property of Nature. Among other things, this "one per measurement", when appropriate, is fundamental to all empirically based science -- along with the idea that the physics of the immediate future is as it is now. No one worries much about this idea either. So, there's really a lot we choose not to question or understand. For some of us QM fits right in there with other stuff we don't know much about -- like, what's an electric charge?.

Humans need interpretations and meanings. This certainly true in physics, and the field is somewhat hung-up on QM. Having a strong sense of how QM came to be, and a long background in all manner of statistics and probability, I see many solid, practical reasons for a QM interpretation that's as simple as possible, Occamized if you will. I should note that my interest is in the application of QM to physical problems. So, what is the minimum set of ideas and stipulations that will allow us to deal with the demands of day-to-day physics?

A short version is: Key Experiments, Schrodinger, Born, Dirac.

Why do you need more if you do? It's can't be in search of a better description/theory of Nature, because the most fundamental questions of science are never asked, nor answered. You want better? Don't forget electric charge, don't forget that we perceive in 3D, why? QM is just the tip of the iceberg.
Regards,
Reilly Atkinson

(Key experiments -- black body radiation, atomic spectra, photoelectric, Davisson Germer, etc.)

 

[peter0302]

I think the main reason QM holds a special place in the hearts of those looking for physical interpretations is because it is so drastically different than anything humanity had seen before. Until the HUP, physics was really taking a slow but steady journey in one direction, namely, that there was a physical reality and no fundamental limits in our ability to describe and predict it. People truly believed that it was possible, in theory, to know _everything_ about the universe, including the future. Even Relativity did not contradict this; in fact, in some ways it helped reinforce this, because it placed a fundamental limiti on action, i.e., locality.

QM threw all this out the window. It really was a huge step _backwards_ in the formulation of a theory of everything. Where we thought we were "getting there", now we don't even know where "there" is.

So I think the quasi-obsession that is seen in finding intuitive interpretations of QM stems from a desire to recapture that feeling that we were actually on the right track in terms of completing our knowledge of the universe. Obviously nobody believes that we can understand everything there is to understand anytime soon, or ever, but we want to be going in the right direction!

I think, personally, that interpreting QM is very important because beneath the philosophical babble, it encourages scientists to search for ways to challenge the rpedictions of QM, and every challenge, even the ones that fail (which is so far ALL of them!) is valuable. Rarely, in the search of interpretation, someone finds a theory that agrees with every experimental test of QM performed thus far but yet still disagree somewhere else. Eventually, some needle in a haystack experiment may come along that does disprove some aspect of QM. That will be an historic day.

Until then, I think we really need to recognize that physics did, in fact, have to take a step backwards before it could move forward, which it did, dramatically, after the discovery of QM.

We're in a state now not too unlike where we were in 1905. Let's not forget that Einstein originally thought his "corpuscles" of light were just an "Interpetation", a mathematical trick that happened to fit the data. It took Bohr and others to realize that the mathematical "trick" had profound implications. I believe someone "interpreting" QM will eventually make a similarly profound realization and build from that an even deeper understanding than we have now.

 

[reilly]

Bohm's work has not produced any new physics in over 50 years; MWI similarly for more like 40. These approaches, however, are excellent producers of controversy. In the meantime, QM has totally transformed our world.
Regards,
Reilly Atkinson

Interpretations are more than "just" interpretations: they have consequences for future research. I've read from proponents of both Bohm's interpretation as well as of MWI, that a future development of their theories will lead to additional predictions which will be verifiable.

 

[akhmeteli]

Bohm's work has not produced any new physics in over 50 years.

In https://www.physicsforums.com/showpost.php?p=1565868&postcount=8 I asked you the following:

"As for new physics, do the Bell's inequalities qualify as "new physics"? As far as I know, Bohm's interpretation was the inspiration for Bell."

I did not get an answer.

Furthermore, does the Aharonov-Bohm effect qualify as new physics? I know that A-B were not the first to discover it theoretically, but the effect became famous after their work.

I value your thoughtful posts, and I would appreciate your reply.

 

[vanesch]

Can a pilot wave or other hidden variable interpretation account for this?

I would think so, but I have to admit not to have worked out any of this in detail, nor having it seen worked out. If it is true that Bohmian mechanics reproduces ALL of quantum mechanical statistical predictions, then this must be part of it. But I admit never having given it much thought of how indistinguishable particles are dealt with in BM.

 

[vanesch]

It may well be that no 1-1 correspondence with the quantum formalism can be achieved. However, that would not be the end of the story. Strictly speaking, what we need is not 1-1 correspondence with the formalism, but agreement with the existing body of experimental data. That is, we need agreement with the formalism up to the experimentally achievable accuracy. However, that accuracy can be extremely high, so maybe I am hair-splitting. The reason I emphasize this distinction is that, for example, the results of Barut's self-field electrodynamics seem to emulate those of QED with high accuracy. To the best of my knowledge, it is not clear if Barut's theory is in disagreement with the body of experimental data. Thus, the Dirac sea interpretation (or you may call it "plasma interpretation", or "vacuum polarization interpretation") may be just added to Barut's theory or to some version of that theory.

There are two different questions here. The first question is: in a toy world where quantum mechanics is supposed to hold strictly, how to give some sense or some intuitive understanding of the workings of that world ? That's the question one asks when dealing with an interpretation of the quantum formalism.

The other question is: in what measure is our actual world described, or not, by quantum theory ? That's a scientific question of the validity of quantum theory.

In as much as it is of course useful to probe eventual limits of the applicability of quantum theory to the real world, one shouldn't mix both. It is not because one has interpretational problems (that one is intellectually unsatisfied with the picture that quantum theory offers you) that you should bet on the other horse. That's a bit like "hoping the problem will go away".

Of course, doing so might suggest challenges to quantum theory where both differ, in order to find potential ways to find ultimate limits to the applicability of quantum theory. So as a "generator of challenges", this might be a useful exercise.

Personally, I'm affraid that Barut's theory is a kind of "semiclassical" approach to QFT. We know many instances of semiclassical approximations in quantum theory which give very good results, and in some cases, exactly the same results as the full-blown quantum machinery. In fact, this is the case each time when at no point, quantum interference is an essential component in the setup, and one is allowed to interchange statistical mixtures and superpositions. As such, I'm not a priori impressed by a semi-classical calculation that is in full agreement with a quantum result. It is always nice to know, of course, but the existence of the semi-classical explanation of a result in full agreement with the quantum prediction doesn't necessarily mean that this will be the case in all generality. There are too many results known where there IS a genuine difference between a semi-classical approach and a full-blown quantum calculation. The challenge is more on that side. There's no real difficulty in reproducing semi-classically the hydrogen spectrum. But once there are more electrons, things become harder. For instance, the helium spectrum. Configuration interaction in quantum chemistry. Things like that.

 

[vanesch]

I think the main reason QM holds a special place in the hearts of those looking for physical interpretations is because it is so drastically different than anything humanity had seen before. Until the HUP, physics was really taking a slow but steady journey in one direction, namely, that there was a physical reality and no fundamental limits in our ability to describe and predict it. People truly believed that it was possible, in theory, to know _everything_ about the universe, including the future. Even Relativity did not contradict this; in fact, in some ways it helped reinforce this, because it placed a fundamental limiti on action, i.e., locality.

QM threw all this out the window. It really was a huge step _backwards_ in the formulation of a theory of everything. Where we thought we were "getting there", now we don't even know where "there" is.

So I think the quasi-obsession that is seen in finding intuitive interpretations of QM stems from a desire to recapture that feeling that we were actually on the right track in terms of completing our knowledge of the universe. Obviously nobody believes that we can understand everything there is to understand anytime soon, or ever, but we want to be going in the right direction!

I think, personally, that interpreting QM is very important because beneath the philosophical babble, it encourages scientists to search for ways to challenge the rpedictions of QM, and every challenge, even the ones that fail (which is so far ALL of them!) is valuable. Rarely, in the search of interpretation, someone finds a theory that agrees with every experimental test of QM performed thus far but yet still disagree somewhere else. Eventually, some needle in a haystack experiment may come along that does disprove some aspect of QM. That will be an historic day.

Until then, I think we really need to recognize that physics did, in fact, have to take a step backwards before it could move forward, which it did, dramatically, after the discovery of QM.

We're in a state now not too unlike where we were in 1905. Let's not forget that Einstein originally thought his "corpuscles" of light were just an "Interpetation", a mathematical trick that happened to fit the data. It took Bohr and others to realize that the mathematical "trick" had profound implications. I believe someone "interpreting" QM will eventually make a similarly profound realization and build from that an even deeper understanding than we have now.

I'm very much in tune with what you wrote here...

 

[akhmeteli]

Personally, I'm affraid that Barut's theory is a kind of "semiclassical" approach to QFT. ... There are too many results known where there IS a genuine difference between a semi-classical approach and a full-blown quantum calculation. ... There's no real difficulty in reproducing semi-classically the hydrogen spectrum. But once there are more electrons, things become harder. For instance, the helium spectrum. Configuration interaction in quantum chemistry. Things like that.

I am not sure Barut's is a semiclassical theory. The reason is as follows. He eliminates the electromagnetic field from his theory using a causal Green function. I believe this is equivalent to quantization of electromagnetic field (please correct me if I am wrong). You may say that this procedure used by Barut stinks to heaven, and I may have a hard time looking for objections, but this does not look like your typical semiclassics.

As for identical particles, Barut uses another trick, introducing an antisymmetrized expression for the current into the action. As a result, it looks like he should not have problems, say, with helium. Again, you may question this procedure, but technically it may actually work.

Again, if you say that there is no fully satisfactory solution so far, I'll have to agree, but I still don't think the Dirac sea interpretation is clearly indefensible.