I Is the collapse indispensable?

[atyy]

vanhees71 is wrong for the following reasons.

1. Unitary evolution and the "filtering" that he imagines will allow the projection to be derived cannot do it, because the unitary evolution and partial trace caused by the "filtering" only produce an improper mixture. To get the definite outcome, one must further assume that the improper mixture is converted to a proper mixture, which is the same as assuming collapse. Ballentine and Peres are probably missing this assumption in their erroneous books.

2. The "locality" of QFT that is enforced by the "local" interactions has the meaning of "no superluminal transmission of classical information" (and a little more). It does not mean local interactions and local causality. vanhees71 consistently confuses multiple meanings of "local".

3. Collapse is consistent with the "locality" of quantum field theory. It is not consistent with relativistic causality, but neither is quantum field theory.

 

[A. Neumaier]

Ballentine and Peres are probably missing this assumption in their erroneous books.

I think you should moderate your language.

The books by Ballentine and Peres are highly respectable books that provide all the information one ever needs to understand the basics of quantum mechanics in theory and practice. Calling them erroneous based on your own subjective view of the interpretation issues is inappropriate.

Your arguments are not that impeccable that you would be justified to call their treatment erroneous. (Write your own book and you'll see that it will most likely contain even more glaring problems.)

 

[atyy]

I think you should moderate your language.

The books by Ballentine and Peres are highly respectable books that provide all the information one ever needs to understand the basics of quantum mechanics in theory and practice. Calling them erroneous based on your own subjective view of the interpretation issues is inappropriate.

Your arguments are not that impeccable that you would be justified to call their treatment erroneous. (Write your own book and you'll see that it will most likely contain even more glaring problems.)

I'm quite sure I am right, and the standard interpretation is done that way for good reasons. Ballentine and Peres are wrong.

 

[A. Neumaier]

I'm quite sure I am right, and the standard interpretation is done that way for good reasons. Ballentine and Peres are wrong.

Ballentine and Peres were also sure of what they wrote (this is visible from how they defend it elsewhere in their publications). Moreover, the way they wrote it was done for good reasons.

So it is view against view. In such a case the credentials count, and you as an outsider should be temperate about your bold assertions.

By the way, there is no standard interpretation. You probably mean your favorite interpretation, or your favorite version of the Copenhagen interpretation.

 

[ddd123]

Unitary evolution and the "filtering" that he imagines will allow the projection to be derived cannot do it, because the unitary evolution and partial trace caused by the "filtering" only produce an improper mixture. To get the definite outcome, one must further assume that the improper mixture is converted to a proper mixture, which is the same as assuming collapse. Ballentine and Peres are probably missing this assumption in their erroneous books.

Wouldn't it be strange if they missed it? A book cannot contain every retort to every attack, maybe they've elaborated further on this, elsewhere.

 

[atyy]

Ballentine and Peres were also sure of what they wrote (this is visible from how they defend it elsewhere in their publications). Moreover, the way they wrote it was done for good reasons.

So it is view against view. In such a case the credentials count, and you as an outsider should be temperate about your bold assertions.

By the way, there is no standard interpretation. You probably mean your favorite interpretation, or your favorite version of the Copenhagen interpretation.

I am not an outsider. I am stating that the standard texts are right.

Ballentine and Peres are the outsiders.

 

[atyy]

Wouldn't it be strange if they missed it? A book cannot contain every retort to every attack, maybe they've elaborated further on this, elsewhere.

Wouldn't it be strange if standard quantum mechanics were wrong? Wouldn't it be strange if even Nielsen and Chuang were wrong?

 

[ddd123]

Wouldn't it be strange if standard quantum mechanics were wrong? Wouldn't it be strange if even Nielsen and Chuang were wrong?

I have no idea :D
I guess we should just stick to the pure arguments (or rather, you, I'm too low level).

 

[A. Neumaier]

I am not an outsider. I am stating that the standard texts are right.

Ballentine and Peres are the outsiders.

Which standard texts are you referring to? Who but you decided that they are the standard?

These two books, together with the commented reprints in Wheeler and Zurek, are the modern standard! (There is also decoherence theory, which is newer than these; but this is silent on collapse.) They devote considerable space to the foundations, whereas typical textbooks on quantum mechanics only have short sections where they parrot what they glean from elsewhere, often from the long past.

 

[ddd123]

I'm afraid this thread is at risk of getting closed. Would be a pity if nobody actually answered or provided a source with an answer to atyy's point I quoted in post #104.

 

[A. Neumaier]

To get the definite outcome, one must further assume that the improper mixture is converted to a proper mixture, which is the same as assuming collapse.

Could you please (possibly in a new thread) explain these terms and how you think the conclusion follows, according to your understanding, so that your statement can be critically discussed?

 

[atyy]

Which standard texts are you referring to? Who but you decided that they are the standard?

These two books, together with the commented reprints in Wheeler and Zurek, are the modern standard! They devote considerable space to the foundations, whereas typical textbooks on quantum mechanics only have short sections where they parrot what they glean from elsewhere, often from the long past,

That is not correct. Wheeler and Zurek are research papers. Incidentally, Zurek states standard QM with collapse. His attempt at Quantum Darwinism is a research attempt to remove collapse. Removing collapse from QM is not standard, but a matter of research till this day, and is BTSM.

Standard texts include Landau & Lifshitz; Cohen-Tannoudji, Diu & Laloe; Nielsen & Chuang; Sakurai; Weinberg; Holevo.

 

[kith]

So you do accept the collapse as necessary, just not its physicality!

I wouldn't say that the use of projection operators necessarily needs a justification. If we use the quantum description for the full filtering system, the final state is a superposition which contains a term where the particle leaves the apparatus and a term where the particle is trapped inside the apparatus. The second term has zero overlap with states localized outside the apparatus. So the Born rule gives a probability of zero for all outcomes of all future measurements for the second term. So the predictions are the same whether we use both terms or only the first. In this case, the use of a projection operator to get rid of the second term is purely a matter of convenience.

I thought that you and I had already reached agreement on something similar in an older thread. The conclusion I remember is that we need more sophisticated situations like Bell tests to analyze whether collapse is necessary or not.

 

[atyy]

I wouldn't say that the use of projection operators necessarily needs a justification. If we use the quantum description for the full filtering system, the final state is a superposition which contains a term where the particle leaves the apparatus and a term where the particle is trapped inside the apparatus. The second term has zero overlap with states localized outside the apparatus. So the Born rule gives a probability of zero for all outcomes of all future measurements for the second term. So the predictions are the same whether we use both terms or only the first. In this case, the use of a projection operator to get rid of the second term is purely a matter of convenience.

I thought that you and I had already reached agreement on something similar in an older thread. The conclusion I remember is that we need more sophisticated situations like Bell tests to analyze whether collapse is necessary or not.

I thought we had agreed that collapse was not necessary provided that successive measurements were not made, eg. using something like the https://en.wikipedia.org/wiki/Deferred_Measurement_Principle.

 

[A. Neumaier]

Standard texts include Landau & Lifshitz; Cohen-Tannoudji, Diu & Laloe; Nielsen & Chuang; Sakurai; Weinberg; Holevo.

Why are they today's standard regarding the foundations, while Ballentine and Peres are not? What is the criterion that makes them standard?

None of these except perhaps Holevo (would have to check) specializes on the foundations but treats it in a very short way, that disqualifies it as a standard. For example Nielsen & Chuang devote just 16 pages (Section 2.2) to the topic, out of a total of over 600 pages. And even in these pages they cover a lot of ground, not just the postulates and their discussion.

The collapse is a frequently used textbook device simply because it is a convenient starting point, allowing one to bridge the abyss of quantum foundations in a few words, in agreement with the early history but without having to spend time on getting it fully correct.

You never find it discussed in a quantum field theory book, which is the true foundation of modern theoretical physics. Here everything is in terms of (in principle measurable) correlation functions, which is enough for all uses of quantum mechanics and quantum field theory in the applications.

 

[A. Neumaier]

Wheeler and Zurek are research papers.

Wheeler and Zurek is a commented reprint collection of research papers displaying the spectrum of serious alternatives, and their historical origin. It is unique in this respect (superseding an older treatise by Jammer).

This comprehensiveness and uniqueness makes it a standard. It displays the disagreement on basic issues, versions of which were already then treated as definite statements in many textbooks, among them some you cited - proving that the textbooks selected for convenience rather than representing an agreement (suggested by calling it a ''standard'').

 

[atyy]

Why are they today's standard regarding the foundations, while Ballentine and Peres are not? What is the criterion that makes them standard?

None of these except perhaps Holevo (would have to check) specializes on the foundations but treats it in a very short way, that disqualifies it as a standard. For example Nielsen & Chuang devote just 16 pages (Section 2.2) to the topic, out of a total of over 600 pages. And even in these pages they cover a lot of ground, not just the postulates and their discussion.

The collapse is a frequently used textbook device simply because it is a convenient starting point, allowing one to bridge the abyss of quantum foundations in a few words, without having to spend time on getting it correct.

You never find it discussed in a quantum field theory book, which is the true foundation of modern theoretical physics. Here everything is in terms of (in principle measurable) correlation functions, which is enough for all uses of quantum mechanics and quantum field theory in the applications.

Collapse is found in 2 QFT books: Weinberg and Dimock.

In addition to Holevo, Paul Busch's books on foundations contain collapse as a postulate.

If Ballentine and Peres were right, the measurement problem would be solved. You can see from the Laloe's http://arxiv.org/abs/quant-ph/0209123 and Wallace's http://arxiv.org/abs/0712.0149 that there is no consensus as to whether any interpretation can solve the measurement problem.

 

[A. Neumaier]

Collapse is found in 2 QFT books: Weinberg and Dimock.

Can you please give page numbers?

 

[A. Neumaier]

there is no consensus as to whether any interpretation can solve the measurement problem.

But no consensus also means no standard. In this case there is no obvious right or wrong, while you took sides and declared your favorite to be right (''the standard'') and the others wrong.

 

[atyy]

But no consensus also means no standard. In this case there is no obvious right or wrong, while you took sides and declared your favorite to be right (''the standard'') and the others wrong.

No, what it means is that standard quantum mechanics works. It makes successful predictions consistent with all observations to date. That is why quantum mechanics with collapse is standard - and yes it is obviously right in the sense of making successful predictions.

Now the question is whether the others can do just as well. Since there is no consensus as to whether they can, standard quantum mechanics remains the standard and consensus.

 

[atyy]

But no consensus also means no standard. In this case there is no obvious right or wrong, while you took sides and declared your favorite to be right (''the standard'') and the others wrong.

Furthermore, no consensus does mean there is a standard - the standard version with collapse and with the measurement problem.

 

[ddd123]

I went to read where atyy's speaks of in Weinberg and was a little disappointed. Weinberg recaps the standard collapse postulate in a paragraph about basic quantum theory which he takes as his own. It is then never mentioned again (it's a QFT book).

 

[stevendaryl]

Maybe he means something like this? http://arxiv.org/abs/hep-th/0205105

That paper is interesting, but it's hard for me to believe that it is correct.

 

[vanhees71]

vanhees71 is wrong for the following reasons.

1. Unitary evolution and the "filtering" that he imagines will allow the projection to be derived cannot do it, because the unitary evolution and partial trace caused by the "filtering" only produce an improper mixture. To get the definite outcome, one must further assume that the improper mixture is converted to a proper mixture, which is the same as assuming collapse. Ballentine and Peres are probably missing this assumption in their erroneous books.

The projectors are an effective description of the filtering process. As I said, it's hard to imagine to be able to describe the filtering process in all microscopic detail. I Ballentine and Peres are erroneous, then quantum theory itself is erroneous. There's no single empirical hint for that.

2. The "locality" of QFT that is enforced by the "local" interactions has the meaning of "no superluminal transmission of classical information" (and a little more). It does not mean local interactions and local causality. vanhees71 consistently confuses multiple meanings of "local".

3. Collapse is consistent with the "locality" of quantum field theory. It is not consistent with relativistic causality, but neither is quantum field theory.

Local QFT IS by construction consistent with relativistic causality, an instanteneous collapse obviously not! So there is a contradiction in the foundations, if you assume the collapse to be a physical process. I prefer to abandon the collapse and stick to minimally interpreted relativistic local QFT.

 

[vanhees71]

I'm afraid this thread is at risk of getting closed. Would be a pity if nobody actually answered or provided a source with an answer to atyy's point I quoted in post #104.

Well, I guess it's high time to close it. We are now at a point where the very foundations, common to all metaphysics ("interpretations") added on top, are called erroneous, which is simply ridiculous, because it means we don't even have a consensus on the physical part of the theory. How then can we expect to solve the more complicated metaphysical issues?

 

[kith]

I thought we had agreed that collapse was not necessary provided that successive measurements were not made, eg. using something like the https://en.wikipedia.org/wiki/Deferred_Measurement_Principle.

Yes. But a crucial point of the discussion was that using this terminology, filtering experiments like the one I described above shouldn't be considered measurements and therefore need no collapse. So I don't get the discussion between vanhees71 and you. To me, it looks like you two are talking at cross purposes.

 

[zonde]

That paper is interesting, but it's hard for me to believe that it is correct.

The paper says:
"Incidentally, since it is impossible to prepare a state with a definite number of photons, and since such an uncertainty for any given process cannot be made arbitrarily small, we may even argue that it is not possible to give a physical reality (not even locally) to any observable, except to the charges and masses (that are the invariants of the theory)."
It seems that the reasoning relies on detection loophole (photons can't be paired up to arbitrary high level).
I would say that this view is falsified by experiment. Say loophole free quantum steering experiment: http://arxiv.org/abs/1111.0760
And electron based experimental violations of Bell inequalities are outside the scope of the paper.

 

[ddd123]

Local QFT IS by construction consistent with relativistic causality, an instanteneous collapse obviously not! So there is a contradiction in the foundations, if you assume the collapse to be a physical process. I prefer to abandon the collapse and stick to minimally interpreted relativistic local QFT.

How does it explain Bell pairs phenomenology though? A paper I linked was deemed "probably wrong" by steveandaryl.

 

[atyy]

Yes. But a crucial point of the discussion was that using this terminology, filtering experiments like the one I described above shouldn't be considered measurements and therefore need no collapse. So I don't get the discussion between vanhees71 and you. To me, it looks like you two are talking at cross purposes.

That could well be. As far as I can tell, I don't have a technical disagreement with you. I still do have technical disagreements with vanhees71.

 

[zonde]

Well, I guess it's high time to close it. We are now at a point where the very foundations, common to all metaphysics ("interpretations") added on top, are called erroneous, which is simply ridiculous, because it means we don't even have a consensus on the physical part of the theory.

I don't want to take side in your discussion with atyy (I don't have clear viewpoint on collapse) but as I see atyy have valid reasons to doubt Ballentine approach.
Ballentine in his book says:
"There is no such difficulty with interpretation B [that is used by Ballentine], according to which the state vector is an abstract quantity that characterizes the probability distributions of the dynamical variables of an ensemble of similarly prepared systems."
So it seems like Ballentine is promoting sort of LHV. And we know now that it does not work.