This is wrong. Scientific theories provide predictive power, yes, but they should also provide explanatory power.
Which is to say, they should describe a model of the world because this is how we devise experiments, and experiments are clearly critical for science.
Two theories with equivalent predictive power but unequal explanatory power are not equally good theories.
Can you tell me how to calculate whether a system of particles will cause another system of particles to collapse or not?
Can you tell me under what circumstances a system of particles will evolve unitarily or not?
Can you shade the region of the spacetime diagram of EPR where the wavefunction is collapsed? How about in a delayed choice quantum eraser experiment?
If you tell me "you're not allowed to ask those questions" (or "hm, I never thought of that!"), then you're directly illustrating the complaint here about physics!
The common narrative is that Copenhagen, many worlds, and the other interpretations of QM are all equivalent, but they are not. Copenhagen adds an extra physical event-- collapse-- where the wavefunction is suddenly nonunitary. The burden is on Copenhagen to tell me how and when this happens, and in fact to prove that it happens at all. Many worlds, on the other hand, predicts the same phenomenon-- the apparent collapse of a wavefunction to an eigenstate-- without adding unitarity violation, or any other phenomena at all beyond the normal, extremely well-verified mechanics of multi-particle system scattering; it merely treats the environment is a multi-particle quantum system.
Many worlds is the null hypothesis (no, the extra worlds are not extra suppositions, they are predictions of known physical laws), and the burden is on Copenhagen to show that unitarity violation exists, and the burden is extremely high (possibly insurmountable?) for EPR and eraser experiments.
When Copenhagen was devised, most believed that there was some "underlying state" in QM, and that measurement told you something about what the underlying state was. Bell's theorem should have sent a shockwave through the community which forced everyone to reevaluate fundamental assumptions and self-correct wrong ideas. But by accident of history, John Bell was too shy to publish in a major journal, and no one even read his result for four years. The implications of Bell's theorem were slow to diffuse through the community, and so the disruptive moment of reckoning that should have happened never came. By the time it was well-accepted, the narrative in physics had become "you're not allowed to ask about what quantum mechanics means, just shut up and calculate", and so the cognitive dissonance was cast aside. Annealing this attitude and this misstep away is taking excruciating decades.
EDIT: I also don't mean to imply that there aren't deep and beautiful things in modern physics. There are! But physics is concertedly avoiding asking deep questions in areas where (I strongly believe) it is most important. The claim that "the meaning of QM is outside the scope of science" is exemplary, I think, of that attitude.
You appear to be somewhat mistaken. Copenhagen interpretation does not postulate any specific explanation or mechanism for wave-function collapse, merely that "upon measurement, the wave-function collapses into an eigenstate of the observable being measured". This, of course, is a physically verified phenomenon. Now Copenhagen purposefully leaves the precise meaning of "measurement" undefined, since in my view there is no convincing empirical evidence that supports a specific mechanism for this phenomenon. Other interpretations posit mechanisms (decoherence (doesn't fully account), von Neumann "consiousness" (not empirical), etc.) for this collapse.
My biggest complaint about many-worlds interpretation is how it is in its essence a non-scientific theory, as it makes assertions about an unobservable reality. It postulates other parallel realities that by definition do not communicate. Again, this makes it intrinsically a non-scientific theory.
>Can you tell me under what circumstances a system of particles will evolve unitarily or not?
Everywhere except on measurement, in which the state collapses.
>How about in a delayed choice quantum eraser experiment?
I'm not familiar with this experiment.
>If you tell me "you're not allowed to ask those questions"
By all means, ask as many questions as you want. That's after all the essence of scientific endeavour. But it's not very nice to misrepresent other positions, nor is it to claim everyone else is deluded (without strong evidence on your side at least).
>Bell's theorem should have sent a shockwave through the community which forced everyone to reevaluate fundamental assumptions and self-correct wrong ideas.
Bell's inequalities force no re-evaluation. They simply prove that the search for a local hidden-variable theory is impossible. It certainly raised important ideas for research, but it does not do anything to discredit established QM.
>The claim that "the meaning of QM is outside the scope of science" is exemplary, I think, of that attitude.
If a physical argument can be made about this problem, then it's in the scope of physics. Otherwise, it is not. As simple as that.
No it's not. There are interpretations in which collapse does not exist, so these experiments are not measuring what you think they're measuring unless you already assume the conclusion.
Also, your view on other interpretations of QM are outdated. Arguably one of the more famous results on quantum mechanics, Bell's theorem, would have never existed if not for Bohmian mechanics.
Don't discount the value of explanatory power. The fact that other interpretations provide far more explanatory power than Copenhagen makes them far more valuable. Many important results in quantum foundations would have never happened if everyone were a Copenhagenist.
The major interpretations of QM are all similar in the fact that they are not testable and don’t really affect the resulta. It’s metaphysics.
> the implications on conditional probabilities hold for other measurements throughout the entire spacetime, present and past. [emphasis theirs]
And the math in the paper which supports that statement works by keeping all the superpositions around and allowing us to project to them at any time. This is the picture of many worlds! Copenhagen says the opposite: When you make a measurement, the unmeasured superpositions go away. The paper confirms, that's nonsense! If you shade any region of the spacetime diagram where the superposition is gone ("collapsed"), you'll be wrong!
Also: "It is important to note that arriving at our conclusions did not require introducing new physics. We only relied on elementary quantum mechanics: not on novel ‘backwards time’ concepts, nor on any particular interpretation: we only used the Born rule ‘as is’. [...] With the remarks and intuition presented here, there really is no mystery whatsoever in any of the discussed experiments."
This experiment is not more mysterious than the rest of QM, but of course if you find the whole theory unsatisfactory you won't be satistified with this explanation either.
Also, this is a fantastic example of the "don't ask questions" attitude I think is so shameful. If they had bothered to take the small step of asking, they'd have come to a clear, evidence-based refutation of Copenhagen! There are regions of spacetime after the measurement, where the wavefunction is not collapsed, which the authors explicitly point to. That's in direct contradiction to the premise of Copenhagen.
Edit: solving the problem in the "natural" order would give the same probabilities but is much more difficult to handle. You need to get a probabilistic outcome for the position of each single photon on the screen, which "collapses" the state and determines the wave function of the other photon. At this point, the first photon has been detected somewhere but it can only be labeled as "interference" or "not interference" later after the detection of the second photon. The probability of being labeled as "interference" or "not interference" does depend on the position (because the collapsed wave function depends on what the outcome of the previous measurement was). When everything is said and done, looking at the subset of events labeled "interference" there is an interference pattern and looking at the subset of events labeled "not interference" there is not an interference pattern. There is no mystery.
Before measurement, the state of the particle system is Σ α_i|i>. Copenhagen says, "after you measure the system, all but one of the α_i go to zero". The authors don't do that, and in fact say that you can't. Unless I am misunderstanding something, they keep all the α_i around at all times and project the measurement for each particle separately, regardless of whether it has been (or will be) measured somewhere else. There is nothing philosophically or physically wrong with this, as they point out, but it is different than what Copenhagen says you should do when you measure something.
And if you look at what it means (which they refuse to do), you'll see that after Alice makes her measurement, Bob's quantum state is still explainable in terms of a superposition. When is the measurement (joint or individual) "finished"? Answer, per the authors: Never. The superposition permeates spacetime. That's how we escape the need for a causal connection between Alice and Bob's measurements, and, naturally, that's how many worlds does it.
(And if we take it further and ask, when Bob makes his measurement, "What is the state of Alice's particles?" we'll see that she is in a superposition of being entangled with each of the superpositions of the particle, which remains in superposition before, throughout, and after our ~measurement of~ entanglement with it).
What he shows in that paper is that the order of the measurments is irrelevant [1]. So he does solve the problem in the reverse order where it can be done easily writing a few quantum states [2].
Note that QM is not about causal connections, is about correlation. Once a pair is correlated, the correlation may appear when measurements are done. But it's not that observing one outcome here and now "causes" a particular outcome there and later (or before). One doesn't need to keep superpositions to think it "correlation" terms (instead of "causal" terms).
[1] Using the projection postulate in the derivation: "Say we have indeed measured on B and got OB = bJ . The state then collapses onto ..." He concludes: "None of this looks very surprising, but we want to stress that the total probability to find OA = aI and OB = bJ does not depend on the place or time at which the measurements occur."
[2] He also uses the projection postulate here in the usual way: "So the experimental outcome (encoded in the combined measurement outcomes) is bound to be the same even if we would measure the idler photon earlier, i.e. before the signal photon by shortening the optical path length of the downwards configuration. Then, if the idler is detected at D4 for example, the above state ‘collapses’ onto ..."
And yes, you don't need to ask "when does the wavefunction collapse" to manipulate joint probabilities of measurements that happen at disparate locations and times. In fact, that's my objection: If you do ask, you find that there is no consistent answer! (And I suggest it's because wavefunction collapse is not a thing the universe does).
Re: "interference tagging"-- do you have a link to some material? (I'd love to understand something specific before commenting).
EDIT: Also, I didn't see Appendix B at first-- The authors do understand and even advocate the Everettian view! Though I still don't quite understand/agree with their earlier timidity about finding and interpreting conflicts with Copenhagen.
I'm not sure what's the problem. Why do have to "ask" if the answer doesn't really matter? What answer more consistent than "it doesn't really matter" would you like? Anyway (standard) QM is a non-relativistic theory, QFT may be more satisfactory from that point of view.
Re: "interference tagging" - what I mean is that first you detect the photons and later check if they "did happen" to go through two slits (interference appears) or one slit (no interference). But the interference pattern is not visible for a single photon and at that point the individual events are still a superposition of both possibilities (so for the events at a certain position part of them will be in the end identified as coming trough one slit and some of them from both). Only after the second measurement is done you know how to group the previoulsy recorded events to see the interference. It's not that the later measurement causes interference to appear. Or at least it doesn't affect at all where the photons were detected, it just lets you know how to group the existing events to make it apparent (selecting only those where, once the full mesurement on the pair has been done, the path taken remains uncertain).
Edit: maybe this picture helps https://upload.wikimedia.org/wikipedia/commons/thumb/c/c8/De...
If all the events are taken together there is no interference pattern. But when they are grouped according to where the second photon is detected in two cases there is still no interference but in the other cases complementary interference patterns appear.
You tell me; Copenhagen is the one that says collapse exists. It sounds like maybe we are on the same page that collapse isn't necessary to explain quantum mechanical observations? In that case, we are both Everettians :).
In some cases it is a safe approximation to ignore those extra states for the remainder of our experiment/calculations, but with a small change to the experiment we can make that a bad approximation.
a) You do the measurement first on the "screen" side (and project the quantum state of the pair of photons according to the measurement, the "extra universes" disappear). You do then the measurements on the "idler" side (and project again the quantum state according to standard QM).
b) You modify slighly the setup to reverse the order of the measurements. You do the measurement first on the "idler" side (and project the quantum state of the pair of photons according to the measurement, the "extra universes" disappear). You do then the measurements on the "screen" side (and project again the quantum state according to standard QM).
QM predicts that the outcomes in the original experiment (a) and the "reversed" experiment (b) are the same. And those predictions are verified empirically.
Why do you say that his 1964 paper (I assume) was not read by anyone for four years?
Copenhagen interpretation showed up when they were faced with a choice between preserving locality or determinancy.
They chose wrong — we know now QM is non-local, and that the underlying justification for the Copenhagen model is an extraneous philosophical proposition.
But much like an extra dependency in a software project, no one wants to remove it now that it’s used everywhere, and there’s a lot of “good enough” stuff using it.
That said, it seems to be one of the major impediments to a unified theory: by dropping the extraneous assumption, we have fewer things to reconcile with GR, and can start looking for GR geometries that have quantized non-local behaviors.
ie, dropping Copenhagen and giving geons another look is probably worth it. (And is basically what loop quantum gravity people are doing, as far as I can tell.)
What underlying justification are you talking about and what do you understand by "the Copenhagen model" precisely? At least in the Einstein vs Bohr debates the one denying that QM could be a complete theory because of its non-locality was Einstein, I think.
> Einstein's refusal to accept the revolution as complete reflected his desire to see developed a model for the underlying causes from which these apparent random statistical methods resulted. He did not reject the idea that positions in space-time could never be completely known but did not want to allow the uncertainty principle to necessitate a seemingly random, non-deterministic mechanism by which the laws of physics operated.
The underlying assumption for Copenhagen was to try to preserve locality by assuming non-determinism. However, there’s no saving locality — and non-locality is enough to leave determinism — so there’s no reason for the non-deterministic axiom.
I think I also meant “definite” instead of “deterministic”, but it works out the same.
I found this quote from David Lindley in a review of Adam Becker's book referenced in this thead (http://www.math.columbia.edu/~woit/wordpress/?p=10147):
"The problem with Copenhagen is that it leaves measurement unexplained; how does a measurement select one outcome from many? Everett’s proposal keeps all outcomes alive, but this simply substitutes one problem for another: how does a measurement split apart parallel outcomes that were previously in intimate contact? In neither case is the physical mechanism of measurement accounted for; both employ sleight of hand at the crucial moment."
Physics of today would be unrecognisable to a scientist at the start of the 20th century. Indeed the physics of 1935 would be unrecognisable to a scientist at the start of the 20th century. Our understanding progressed enormously AND it did so because people put forward radical theories in complete rupture with the established, 300-year old classical mechanics. There was no "traditionalist resistance" to it after it became apparent that classical mechanics failed to explain phenomena that quantum theory did explain.
>am astonished at how aggressively physicists resisted and suppressed "unconventional" interpretations of quantum mechanics (like the many worlds interpretation) in favor of the obviously-wrong Copenhagen interpretation [emphasis mine]
Ahahaha, you clearly have no idea what you're talking about. There is nothing "obviously wrong" about the standard Copenhagen interpretation (unless you have some new insight you would like to share), nor was there any suppression of ideas. Many debates have been waged in the past ~100 years, and many alternative interpretations have been put forward, like Bohmian theory, superdeterminism, or "shut-up-and-calculate".
>decided that it was meaningless to ask any deep questions about what QM actually meant
Physics, indeed all science, studies observable reality. Any "deep" questions about why or about things not measurable, quantifiable, or empirically observable, are by definition outside the scope of science. It is therefore as unreasonable as complaining about why don't geologists study epistemology. The answer is the same: it's outside the scope of their study.
I also don't like that your phrase seems to imply that what physicists study is not "deep" as opposed to your philosophical questions. There are many deep and beautiful ideas in physics.