What is the correct reply on Einstein`s thought experiment?

[Adel Makram]

In attempt to refute quantum mechanics, Einstein proposed a thought experiment where a box containing a particle is weighted before and after opening a small hole to release that particle. The precision of measuring the time and the energy of the particle can thus be known accurately which violates the uncertainty principle. Bohr answered by his famous reply that when the particle exists the box, the later moves to the opposite direction by an indeterminate distance to a new space-time value in the gravitational field that creates uncertainty in measuring the time.
But this answer is not convincing at all, simply because gravitational metric is not included in formulation of quantum mechanics. So what would be the best reply on that thought experiment?

 

[Strilanc]

But this answer is not convincing at all, simply because gravitational metric is not included in formulation of quantum mechanics.

If we're not including gravity, then how is the box being weighed?

This has been asked and answered well on the physics stackexchange.

 

[Adel Makram]

If we're not including gravity, then how is the box being weighed?

This has been asked and answered well on the physics stackexchange.

The problem is not in measuring the weight/energy of the particle, but it is in the way of explaining the indeterminacy in time that required by uncertainty principle. Here are two main reasons;
1) QM formulation does not require GR at all, which means it does not need additional effect from GR to explain its prediction.
2) Bohr treated that particle as classical particle which has a definite spacetime value and energy at any given time and this alone is against the basic foundation of QM.
It is funny that both scientists forgot their own theories in that debat, Einstein forgot to contribute GR and Bohr adopted a pre-quantum interpretation of treating the particle.
In conclusion, uncertainty principle will still hold even if Newtonian gravity is applied with absolute universal time. The link you attached went to prove that GR will still be in keeping with the uncertainty principle which is not a completely satisfactory answer. My question, how, within the domain of QM, Bohr would have replied?

 

[Nugatory]

Gravity and general relativity are not essential to this thought experiment. Although Einstein suspended the box from a scale in a gravitational field because that made the description simple, it would work just as well if the box were being subjected to acceleration by a constant force.

 

[Adel Makram]

Also, a particle inside Einstein box has descrite values of energy so when the hole opens for sometime, we are not sure which energy value the particle has at the time it exits the box. So the uncertainty still holds without any relevance to GR.

 

[Demystifier]

But this answer is not convincing at all, simply because gravitational metric is not included in formulation of quantum mechanics. So what would be the best reply on that thought experiment?

In my opinion, the best reply is this:
http://lanl.arxiv.org/abs/1203.1139 [Eur. J. Phys. 33 (2012) 1089-1097]

 

[bhobba]

1) QM formulation does not require GR at all, which means it does not need additional effect from GR to explain its prediction.

If I remember correctly full GR wasn't really invoked to resolve it - just the low gravitational limit and space-time curvature was neglected.

I am personally with Demystifier.

Historically this was the last time Einstein ever attacked the validity of QM - he accepted it as correct from that point on - he simply believed it incomplete. If Einstein had seen into it a bit deeper things may have been different - who knows.

Thanks
Bill

 

[N88]

In my opinion, the best reply is this:
http://lanl.arxiv.org/abs/1203.1139 [Eur. J. Phys. 33 (2012) 1089-1097]

From page 9 of your arxived article:

"As well understood today, the correct resolution of the EPR contradiction is not incompleteness, but nonlocality. Namely, the momentum measurement of the first particle affects also the state of the second particle. Thus, the momentum measurement leading to (29) implies that the state of the second particle is | − p⟩, which has an infinite position-uncertainty ∆x. Consequently, contrary to the EPR argument above, one cannot determine both momentum and position of the second particle. The contradiction obtained by EPR was an artefact of the wrong assumption that measurement on one particle cannot influence the properties of the other. (Emphasis added.)"​

I would welcome comments on the view that the contradiction obtained by EPR was an artefact of the wrong assumption that classicality could be introduced into QM via their elements of physical reality.

 

[bhobba]

I would welcome comments on the view that the contradiction obtained by EPR was an artefact of the wrong assumption that classicality could be introduced into QM via their elements of physical reality.

There is no contradiction in EPR. Its simply a non classical type of correlation.

But this is getting way of topic - start a new thread if interested.

Thanks
Bill

 

[N88]

There is no contradiction in EPR. Its simply a non classical type of correlation.

But this is getting way of topic - start a new thread if interested.

Thanks
Buill

OK. But Demystifier's essay (relating to the OP) points to a contradiction in EPR. In my view the EPR contradiction is based on classicality. Bohr seems to invoke a similar classicality in the Einstein experiment.

 

[bhobba]

OK. But Demystifier's essay (relating to the OP) points to a contradiction in EPR. In my view the EPR contradiction is based on classicality. Bohr seems to invoke a similar classicality in the Einstein experiment.

He is taking about an apparent contradiction. There is no actual contradiction. All that's going on is if you want quantum objects to have properties when not observed you need superluminal action at a distance.

Thanks
Bill

 

[Demystifier]

In my view the EPR contradiction is based on classicality.

No, it's based on locality. But then again, locality can be considered as one of properties of the classical world, so I guess it is not totally wrong to say that it is based on one specific property of classicality.

 

[Ilja]

The EPR criterion is: "If, without in any way disturbing a system, we can predict with certainty the value of a physical quantity, then there exists an element of physical reality which corresponds to this physical quantity."

I see no classicality here. Except one uses "classicality" as a bad word for realism. And the EPR criterion of reality is all we need. Together with "locality" (better Einstein causality), which tells us that the measurement of Alice does not disturb Bob's system.

 

[vanhees71]

Well, since the premise of the famous Einstein quote is not correct, of course the EPR criterion is correct but useless. According to QT you cannot prepare a system such that some given observable has a certain value without in any way disturbing it, and you cannot prepare it in such a way that all observables take certain values. So if you define reality in this (classical deterministic) way, it's not the reality we observe in nature.

In QFT one calls interactions local if the corresponding terms in the Hamiltonian are polynomials of the fields and their derivatives at one space-time point. Together with the microcausality condition (local observables at space-like distances are represented by commuting operators) this implies the linked-cluster principle.

Also, it is well known that Einstein was not very satisfied with the presentation of the point he wanted to make in this EPR paper. There's a paper by himself as a single authors, which is much clearer. What bothers Einstein is indeed the possibility of entangled states describing long-distant correlations between parts of a system, i.e., what he calls the "inseparability" described by such states. The paper is in German:

A. Einstein, Dialectica 2, 320 (1948)

His quibbles are completely resolved, if you don't invoke the collapse hypothesis when measuring an observable, i.e., you don't introduce the necessity of instantaneous actions of a measurement at one part of the entangled system on another far-distant part of this system. That's just the minimal interpretation: Although the measured observable on one part of the system (e.g., the single-photon polarization states of entangled photon pairs in a polarization-entangled state as used in many Bell-test experiments) are undetermined, the correlations are 100% due to the preparation of the system. Thus there's no cause-effect assumption of the one measurement on one part of the system on the other far distant part of the system necessary to explaine these 100% correlations. So Einstein's criticism is in fact about the collapse hypothesis within (some flavors of) the Copenhagen interpretation not agains QT per se in the minimal interpretation.

 

[Ilja]

Well, since the premise of the famous Einstein quote is not correct, of course the EPR criterion is correct but useless. According to QT you cannot prepare a system such that some given observable has a certain value without in any way disturbing it, and you cannot prepare it in such a way that all observables take certain values.

Sorry, I don't understand this point. According to QT, the preparation of the initial state, together with the measurement of spin in direction a by Alice, prepares the system by Bob in such a way that for the measurement in direction a by Bob we can predict the result, with certainty. Not?

If we assume Einstein causality, we can exclude that the measurement by Alice has in any way disturbed Bob's system. Not? Then the criterion tells us that the spin measurement in direction a by Bob has a certain value.

So, this makes sense only if you presuppose that QT violates Einstein causality, that the decision by Alice what to measure influences Bob's system.

That in QFT a definition of "causality" is used which is much weaker, and would better be named correlability or so, is quite irrelevant, it does not save Einstein causality.

That's just the minimal interpretation: Although the measured observable on one part of the system (e.g., the single-photon polarization states of entangled photon pairs in a polarization-entangled state as used in many Bell-test experiments) are undetermined, the correlations are 100% due to the preparation of the system. Thus there's no cause-effect assumption of the one measurement on one part of the system on the other far distant part of the system necessary to explaine these 100% correlations.

It is the very point of the EPR criterion of reality allows us to say that if the causation by the far away measurement is excluded, then the 100% correlation requires that the value corresponds on some real, physical thing. Which, then, leads to Bell's inequality, thus, is false.