Is QM Inherently Non-local in EPR and Bell Discussions?

[Sherlock]

If no viable theory can get along without a certain feature, then that feature is part of nature, right? That's what it *means* to say that the theories can't get along without it -- they can't agree with *experiment* without it, they can't match the *facts* without it.

The feature that quantum theory can't get along without, if it's going to provide a conceptual understanding, and accurate predictions, of quantum correlations in a local universe, is the conservation laws.


The feature that quantum theory would require in order to be explicitly local is the ability to experimentally track the trajectories of entangled particles (if you want to think in terms of particles), or, (if you want to think in terms of wave structures) the ability to experimentally track the evolutions of entangled disturbances. (I'll speak of particles for convenience.)


There's a theoretical constraint on the extent to which such tracking can be done. But quantum theory can still accurately predict the average results of experiments on entangled particles due to the incorporation of the conservation theorems into the theory -- which could be ported to quantum theory from classical theory, because the classical conservation laws don't require following the particles through continuous paths in space time.


From Bohm's text, Quantum Theory:

"Even in an impulsive collision in which we cannot follow the motion continuously, these laws apply for the collision as a whole. Such laws do have meaning even in discontinuous processes. It is an experimental fact that these laws can all be taken over directly into the quantum theory. ...
... Hence, not all classical deterministic laws must be abandoned, but only those requiring a description in terms of continuous processes."


Quantum theory isn't explicitly local because it can't be. It isn't explicitly non-local because it was developed in line with the assumption of locality, and the idea that quantum correlations can be conceptually understood (if not explicitly described) as emerging from relationships (caused by local interactions), due to conservation laws, between particles which have a common source, or which have interacted, or which have been altered by a common torque, etc.


So, there's no reason to scrap the assumption of locality in Nature just because OQM (or any theory of quantum processes if the principles of OQM are essentially correct) can't be formulated as an explicitly local theory.

 

[ttn]

And: loosening the standard for what we deem "local" (so that the OI condition is not required), oQM qualifies as such a local non-realistic theory.

And loosening the standard for what we deem "a peanut butter sandwich", Beethoven's 5th symphony qualifies as a peanut butter sandwich!

Sorry, I couldn't resist.

By the way, if you're interested in the relation of Bell Locality to Outcome Independence and Parameter Independence, you absolutely must read chapter 4 (esp. the section called "What does factorization signify?") of Tim Maudlin's spectacular book "Quantum Non-Locality and Relativity." He shows that the parsing of Bell Locality into "OI" and "PI" isn't unique, and that all the conclusions change if you do it a different way. Let me clarify that. Bell Locality is, as you noted, the conjunction of OI and PI. But it is also the conjunction of another pair of requirements, call them OI' and PI'. OQM violates OI but not PI. But on the other hand OQM violates PI' but not OI'. And vice versa for Bohmian Mechanics -- it violates PI but not OI, and OI' but not PI'. So which one is really local and which one isn't? It simply isn't clear... which is precisely why it's better to drop the whole "deflated elephant" (as one researcher once described it to me) of Outcome- and Parameter-Independence, and simply talk about Bell Locality.

(BTW, note also that Jarrett, who discovered the whole OI/PI thing using different terminology, initially asserted that a violation of PI means a violation of signal locality. And that is just plain wrong. So you have to be very very careful believing what people -- even "experts" -- say on these topics. Except Maudlin. He's right about everything.)

To be fair to your position, we could also say that oQM does not fall within the class of theories affected by Bell's Inequality. Now that we have agreed on this point, what possible connection does any of this have to Bohmian Mechanics? Why would Bell have felt compelled to mention BM IF he already saw oQM as a good non-local theory (as you think is obvious and always has been) ?
I personally think the answer is (and this is somewhat speculative): Bell was not sure if OI made sense as a requirement of locality. After all, he saw oQM as Lorentz invariant! On the other hand, mentioning BM had the advantage of at least indicating that there existed theories other that oQM which might meet the standards of Bell's Theorem.

I'm not sure what you're getting at here. Bohm's theory doesn't really have anything to do with Bell's Theorem, except that we know, in part from Bell's Theorem, that an empirically viable theory (like Bohm's) will have to violate Bell Locality (which Bohm's theory indeed does).

Were you wondering why anyone would believe in Bohm's theory when there's already a perfectly good (but non-Bell-Local) theory out there, namely OQM? Here there's an easy answer: because Bohm's theory solves the measurement problem.

 

[ttn]

1. We know that one cannot construct a viable theory which is both local and realistic, and we agree on this point. If it is possible to construct a theory which mimics the predictions of QM but is non-local and realistic - that would be BM - then... it might also be possible to construct a a theory which mimics the predictions of QM but is local and non-realistic. That is simple logic.

Well, I don't agree. The kinds of hidden variables Bell assumes in deriving the inequality turn out to be the minimally necessary ones to build a Bell Local explanation of the perfect correlations. And that means it is *not* going to be possible to build an empirically viable Bell Local theory. That is, it is not "possible to construct a theory which mimics the predictions of QM but is local and non-realistic."

But I'm sure that if I'm wrong about this, you will be forthcoming with a counterexample to prove me wrong.

 

[Careful]

Just a few notes:

(a) Experiment does not show at all that the predictions of OQM are correct; to be precise, every experimental outcome has a local realist explanation up till now.
(b) It would be better for the discussion to specify precisely which Bell type of inequalities you are referring to: the orginal one, the CHSH inequality or the CH74 no enhancement inequality..
(c) There are factors in the experiments which are not explained by the simple singlet state description of OQM, such as an assymetry between detectors readings or detection probabilities which are dependent upon the relative detector settings.

Moreover, as I seem to remember, there are even contrived local realist models which fully agree with the outcome of OQM (the assumptions for Bell's theorem are violated here in a more subtle way - particles in the cavity are supposed to figure out what the dectors are going to do in the future and act accordingly). All processes involved are stil causal, but not Bell local...

 

[DrChinese]

And loosening the standard for what we deem "a peanut butter sandwich", Beethoven's 5th symphony qualifies as a peanut butter sandwich!

Sorry, I couldn't resist.

You must realize, as Bell did, that the $64,000 question is whether relativistic theories are compatible with Bell's Theorem. So naturally, the Bell Locality condition does have some baggage associated with it since it is a slightly different definition. Is the light cone a fundamental border for the propagation of causes and effects? And how does collapse fit into the equation... is this a cause or effect or what? So asking if OI could be dropped is not an irrelevant issue by any means. And again, I believe that generally accepted scientific opinion is more along the lines that I have outlined.

 

[NateTG]

Well, I don't agree. The kinds of hidden variables Bell assumes in deriving the inequality turn out to be the minimally necessary ones to build a Bell Local explanation of the perfect correlations. And that means it is *not* going to be possible to build an empirically viable Bell Local theory. That is, it is not "possible to construct a theory which mimics the predictions of QM but is local and non-realistic."
But I'm sure that if I'm wrong about this, you will be forthcoming with a counterexample to prove me wrong.

Bell's theorem relies on the assumption that certain physically untestable probabilities are well-defined. Without those probabilities, Bell's inequality is invalid, and the theorem does not apply. And, since these probabilties are physically untestable, they are not a necessary aspect of a hidden variable theory (or any physical theory.)
Thus, the assertion that "the kinds of hidden variables Bell assumes in deriving the inequality turn out to be the minimally necessary..." is questionable.
In fact, it should be quite easy to prove that any realistic theory which allows for 'faster than light correlation' but not for 'faster than light communication' can be mimiced by a theroy that is local but (potentially) unrealistic.

 

[Careful]

Two comments NateTG :

(a) By *Bell's theorem relies on the assumption that certain physically untestable probabilities are well-defined*, you probably refer to the fact that you need more than one dector setting on one side to get the inequality. If you check the derivation of the Bell inequalities carefully, you will see this is not a problem at all. The correlations can be (and have to be) interpreted as coming from independent measurements; otherwhise the Bell inequalities would be just trivial arithmetic inequalities (and therefore trivially satisfied by any experiment).

(b) The common statement that quantum field theory does not allow faster than light signalling is false. You can use the correlations beyond the lightcone in an experiment with three observers to send info faster than the speed of light (see Sorkin: Impossible measurements on Quantum fields). Another point of view would be that the measurement theory in QFT is incomplete.

 

[DrChinese]

Just a few notes:
(a) Experiment does not show at all that the predictions of OQM are correct; to be precise, every experimental outcome has a local realist explanation up till now.

As you are new to this forum, you may not realize that the subject of violation of Bell Inequalties is often a source of contentious discussion. I would recommend that you move to an existing thread - or start a new one - if you want to discuss that subject - or the accuracy of the predictions of QM. This thread is not related to tests of QM or Bell's Theorem.

If you would care to provide an example of any experiment which does not support the predictions of QM, please, go ahead. QM is a very successful theory.

As to the second part of your comment, local reality has been ruled out by at least a dozen different experiments. Besides the Aspect experiments and its subsequent highly improved PDC variants, there are the more recent GHZ type which violate local reality in individual tests. The only folks still adhering to the local realistic program are a relatively small band of diehards (such as Santos, Thompson, and our own nightlight).

Again, further discussion on this should be carried on elsewhere.

 

[Careful]

I do not think I have to inform you why these *diehards* have a point. Indeed, this thread is *is QM inherently non-local?*; however most people, including yourself, have turned this question into *Is nature inherently non local* with of course an affirmative answer attached. Now, I have nowhere seen any interception from your part when this deviation occured. Moreover, it is completely nonsensical to have a discussion about the *exact* issue of this thread; the answer is clearly yes (as well in the Bell as the operational sense of non-locality). You did not say anthing my comments concerning the violation of causality in the operational sense at all, and this is clearly relevant for the discussion.

Concerning your second and third paragraphs, I have already given the answers on that in my previous posting, I would appreciate any accurate comment on that.

 

[NateTG]

Two comments NateTG :
By *Bell's theorem relies on the assumption that certain physically untestable probabilities are well-defined*, you probably refer to the fact that you need more than one dector setting on one side to get the inequality. If you check the derivation of the Bell inequalities carefully, you will see this is not a problem at all. The correlations can be (and have to be) interpreted as coming from independent measurements; otherwhise the Bell inequalities would be just trivial arithmetic inequalities (and therefore trivially satisfied by any experiment).

My post was in response to the assertion that Bell's theorem cannot be circumvented by local non-realistic theories.
Let's say that I have a neutral electron source and I have a single detector that will randomly measure the electron's spin orientation along one of three axes a,b or c at 0,120 and 240 degrees off of the vertical respectively. Then there are six possible results this device can produce: a^+,a^-,b^+,b^-,c^+, and c^-.
Now, if I take all of the electrons that gave a result of, say a^+ what is the probability that those electrons would have given b^- if they had been measured along the b axis instead, and, more importantly is it possible devise an experiment that will test the prediction?
AFAICT QM essentially dictates that we cannot know - so the 'probability' is meaningless in an experimental sense. From a theoretical perspective, one can either postulate that there is a probability -- which is AFAICT an example what the 'realistic' in QM cannot be local and realistic corresponds to, or one could say it's undefined - like, for example \frac{0}{0}.
In the latter case, Bell's theorem doesn't make sense anymore since it involves adding and subtracting undefined quantities.

 

[DrChinese]

I do not think I have to inform you why these *diehards* have a point. Indeed, this thread is *is QM inherently non-local?*; however most people, including yourself, have turned this question into *Is nature inherently non local* with of course an affirmative answer attached. ... You did not say anthing my comments concerning the violation of causality in the operational sense at all, and this is clearly relevant for the discussion.

My views are fairly common, I suspect: I don't think nature is "inherently" non-local (as ttn does). I don't believe there is any violation of signal locality within oQM. I don't have an explanation for the mechanism for the collapse of the wave function, and I don't know if it should be considered to be a physically non-local. But I am open to new information on the matter.

And I accept that nature is not local realistic, which is also assumed to be true for purposes of this thread. If you want to discuss that particular point, I would request you take it to another thread as it is off-topic here.

 

[DrChinese]

So you have to be very very careful believing what people -- even "experts" -- say on these topics. Except Maudlin. He's right about everything.

That's pretty funny...

 

[NateTG]

DrChinese, would you consider the notion that a physically unmeasurable quantity may be undefined (in the \frac{0}{0} sense) as realistic?

 

[DrChinese]

DrChinese, would you consider the notion that a physically unmeasurable quantity may be undefined (in the \frac{0}{0} sense) as realistic?

Do you mean "realistic" per Bell?

I usually think in these terms: oQM sees the physically unmeasurable characteristics as non-existent or undefined, exactly as you describe them; and that implies "non-realism" to me. There is non-realism to those observables which can be measured individually but not simultaneously. That maps to Bell's definition exactly (what might be called "Bell Reality") because oQM does not meet that definition, and indeed was not supposed to. And Bell Reality also was supposed to map in some way to EPR's "elements of reality"; and indeed they are very close, even if not identical.

Bell wanted oQM on one side of the fence, and EPR's vision of a local reality on the other. He succeeded grandly.

 

[Careful]

My post was in response to the assertion that Bell's theorem cannot be circumvented by local non-realistic theories.
Let's say that I have a neutral electron source and I have a single detector that will randomly measure the electron's spin orientation along one of three axes a,b or c at 0,120 and 240 degrees off of the vertical respectively. Then there are six possible results this device can produce: a^+,a^-,b^+,b^-,c^+, and c^-.
Now, if I take all of the electrons that gave a result of, say a^+ what is the probability that those electrons would have given b^- if they had been measured along the b axis instead, and, more importantly is it possible devise an experiment that will test the prediction?

Let me try to make a suggestion. In quantum mechanics the wave function serves to make statistical predictions about the outcomes of a series of identical experiments. If I shoot the silver atoms (with electrons, you would get a null result, Stern Gerlach experiments with electrons do not provide any direct clue for spin) one by one out of the source and *assume* that they are measured each time when the detector is set on the a direction, I could after sufficiently many times determine the pure state of the electron up to a phase with sufficient accuracy. Now, I can fix the phase by doing a similar series of experiments in the b direction. From that momen on, I can predict whatever I want with fair accuracy.

 

[Careful]

Doctor chinese, you are avoiding again the reference of Sorkin concerning *impossible measurements in QFT* I provided you with (you simply comment that you do not believe it); also you ignore my comments concerning some mismatches between experiment and standard QM predictions. It seems therefore a bit contradictory to you say that you are open to info on the matter while you keep on avoiding the issues I raise. Science is no religion ...

 

[DrChinese]

Doctor chinese, you are avoiding again the reference of Sorkin concerning *impossible measurements in QFT* I provided you with (you simply comment that you do not believe it); also you ignore my comments concerning some mismatches between experiment and standard QM predictions. It seems therefore a bit contradictory to you say that you are open to info on the matter while you keep on avoiding the issues I raise. Science is no religion ...

Huh? I saw a mention of Sorkin and have no idea what it is about or in reference to. I certainly made no comment about it since there is nothing to comment on. Hopefully it is somehow relevant to this thread, perhaps if that is so you will explain it to us (so we will understand how it relates). Disagreement between QM and experiment does NOT belong in this thread, please start a new one if you want to discuss that. This thread is about the non-local nature of QM, and it is poor etiquette to hijack threads for your own purposes. Your participation is very welcome and encouraged at PhysicsForums, but you will not find poor manners tolerated here for long.

As to science being a religion, again I have no idea what you are talking about. And I am pretty sure there are plenty who would get a kick out of the idea that I avoid issues like this.

 

[DrChinese]

Let me try to make a suggestion. In quantum mechanics the wave function serves to make statistical predictions about the outcomes of a series of identical experiments. If I shoot the silver atoms (with electrons, you would get a null result, Stern Gerlach experiments with electrons do not provide any direct clue for spin) one by one out of the source and *assume* that they are measured each time when the detector is set on the a direction, I could after sufficiently many times determine the pure state of the electron up to a phase with sufficient accuracy. Now, I can fix the phase by doing a similar series of experiments in the b direction. From that momen on, I can predict whatever I want with fair accuracy.

Careful:

This post does not relate to this thread. This thread is about non-locality. Please start a new thread to discuss your perception of quantum mechanical statistics. This would be welcome in the proper thread, and is unwelcome in the wrong thread.

I will not participate in a discussion deliberately going out of topic.

-DrC

 

[CJames]

I am joining this discussion as an undergrad with no mathematical knowledge of QM thus far, so keep that in mind. However, I am having trouble seeing why this issue is so difficult to resolve. First off, it seems readily apparent to me that signal-locality is true of QM. I don't think anybody except Careful has said that QM violates special relativity. I can concieve of no method by which a signal can be sent via quantum entanglement.

It also seems readily apparent to me that Bell locality is falsified by experiment. I did not think there was any controversy as to what happens when one member of an entangled pair is observed. You instantaneously know information about the other particle. This does not mean information has traveled outside the light cone, but it does mean you know information about an object outside of the lightcone. In this way the universe is non-local in the sense that a wavefunction can say something about more than one object, or to say it in a more shocking way, the two particles are actually the same "object", despite their physical separation. This is how I understand quantum entanglement. If I am misinformed please let me know.

I was under the impression that both these concepts were well established by experiment. Am I wrong about that?

 

[DrChinese]

I am joining this discussion as an undergrad with no mathematical knowledge of QM thus far, so keep that in mind. However, I am having trouble seeing why this issue is so difficult to resolve. First off, it seems readily apparent to me that signal-locality is true of QM. I don't think anybody except Careful has said that QM violates special relativity. I can concieve of no method by which a signal can be sent via quantum entanglement.

It also seems readily apparent to me that Bell locality is falsified by experiment. I did not think there was any controversy as to what happens when one member of an entangled pair is observed. You instantaneously know information about the other particle. This does not mean information has traveled outside the light cone, but it does mean you know information about an object outside of the lightcone. In this way the universe is non-local in the sense that a wavefunction can say something about more than one object, or to say it in a more shocking way, the two particles are actually the same "object", despite their physical separation. This is how I understand quantum entanglement. If I am misinformed please let me know.

I was under the impression that both these concepts were well established by experiment. Am I wrong about that?

No, I think you have it right.

For me personally, the confusion begins when you talk about an object outside the lightcone. Alice makes a measurement, which causes collapse of the shared wave function. Now you know something about something somewhere else, true, and that is outside the lightcone.

But what has happened that is really so weird? We project the knowledge we have back to the point at which the entangled particle pair was created. This is the same thing that happens when only one particle is involved, nothing strange about that. The particle acts as if it had that orientation from the last point something happened.

Say Alice sees a V orientation with a polarizer at 0 degrees. Naturally, all subsequent measurements will be consistent in EVERY WAY with this knowledge AS IF it was always that way from the creation of the particle. So in that sense there is absolutely nothing happening outside any light cone.

In other words, all quantum measurements find a particle in an eigenstate and its eigenvalue is consistent with the quantum measurement rules. Entangled particles are no different in this respect. So the real question to me is: why does a measurement at time T2 cause the particle to assume a specific value as if it had that value at time T1 (where T1 is before T2) ? Does that make oQM non-local? Or is that a case of backwards causality? I am not sure that anything physical occurs along with the collapse, and I think that is a relevant question too.

Naturally, some of these issues show up in our definition of locality. You can see that there is no information transfer which is FTL, and there is no clear causal effect which is FTL. Yet the Bell Locality condition is violated with a strict application of its definition. So what does that condition actually tell us? Of course, it fits with the Bell Inequality too so that is very important.

Inquiring minds want to know...

 

[Careful]

Huh? I saw a mention of Sorkin and have no idea what it is about or in reference to. I certainly made no comment about it since there is nothing to comment on. Hopefully it is somehow relevant to this thread, perhaps if that is so you will explain it to us (so we will understand how it relates). Disagreement between QM and experiment does NOT belong in this thread, please start a new one if you want to discuss that. This thread is about the non-local nature of QM, and it is poor etiquette to hijack threads for your own purposes. Your participation is very welcome and encouraged at PhysicsForums, but you will not find poor manners tolerated here for long.
As to science being a religion, again I have no idea what you are talking about. And I am pretty sure there are plenty who would get a kick out of the idea that I avoid issues like this.

Sorkin' paper treats the following issue: if you accept QFT and accept that measurment of all gauge invariant observables can be made (local or non local - examples of non local observables are so called Wilson Loops) then the Wightman axiom that two spacelike separated field operators do either perfectly commute or anticommute does indeed imply that two spacelike separated observers cannot signal faster than with the speed of light, but you can carefully select a situation with three obervers A in the past of B , C in the future of B but A not in the past of C in which a measurement at A is going to influence the signalling from B to C. Since this is unacceptable, you need to exclude by hand this kind of situations (note that non local observables are physical and belong to this world) which is the most uninsightful thing you could ever do. To someone with common sense, it seems almost impossible and extremely contrived that nature provides us with non local correlations (ie correlations beyond the lightcone) but forbids us to use them actively. There is much more to say still about the wightman AXIOMS but I leave it for now here. The paper is : Sorkin, impossible measurements on quantum fields, and you can find it on gr-qc (written in 1984).

 

[Sherlock]

Sorkin' paper treats the following issue: if you accept QFT and accept that measurment of all gauge invariant observables can be made (local or non local - examples of non local observables are so called Wilson Loops) then the Wightman axiom that two spacelike separated field operators do either perfectly commute or anticommute does indeed imply that two spacelike separated observers cannot signal faster than with the speed of light, but you can carefully select a situation with three obervers A in the past of B , C in the future of B but A not in the past of C in which a measurement at A is going to influence the signalling from B to C.

A and B are particles which have interacted. We detect A. Since the motions of A and B are somewhat related subsequent to their interaction, then it follows that the detection of A can tell us something about how B might interact with C.

To someone with common sense, it seems almost impossible and extremely contrived that nature provides us with non local correlations (ie correlations beyond the lightcone) but forbids us to use them actively.

But they are used actively, aren't they? Quantum computing? :-)

The view that Nature is local fits the data. It's the simplest, reasonable explanation for why we can't communicate (ie., use the correlations actively in the sense that I take you meant this) superluminally.

Correlations beyond the lightcone are conceptually understood in terms of the conservation laws.

The paper you reference sounds interesting ... I must check it out.

 

[CJames]

No, I think you have it right.
For me personally, the confusion begins when you talk about an object outside the lightcone. Alice makes a measurement, which causes collapse of the shared wave function. Now you know something about something somewhere else, true, and that is outside the lightcone.
But what has happened that is really so weird? We project the knowledge we have back to the point at which the entangled particle pair was created. This is the same thing that happens when only one particle is involved, nothing strange about that. The particle acts as if it had that orientation from the last point something happened.

To me that is the weird part, that the same thing that happens when one particle is involved is what happens when two particles are involved. This is an actual physical example of one thing being in two places at once. That's what I think of when I hear the words non-local. I don't think of superluminal travel or violating special relativity, I think of the total breakdown of the classical concept that an object cannot be in two places at once.

Say Alice sees a V orientation with a polarizer at 0 degrees. Naturally, all subsequent measurements will be consistent in EVERY WAY with this knowledge AS IF it was always that way from the creation of the particle. So in that sense there is absolutely nothing happening outside any light cone.
In other words, all quantum measurements find a particle in an eigenstate and its eigenvalue is consistent with the quantum measurement rules. Entangled particles are no different in this respect. So the real question to me is: why does a measurement at time T2 cause the particle to assume a specific value as if it had that value at time T1 (where T1 is before T2) ? Does that make oQM non-local? Or is that a case of backwards causality? I am not sure that anything physical occurs along with the collapse, and I think that is a relevant question too.

But it shouldn't be possible even in principle to demonstrate whether the collapse is physical, since once you observe it the whole thing collapses.

Naturally, some of these issues show up in our definition of locality. You can see that there is no information transfer which is FTL, and there is no clear causal effect which is FTL. Yet the Bell Locality condition is violated with a strict application of its definition. So what does that condition actually tell us? Of course, it fits with the Bell Inequality too so that is very important.
Inquiring minds want to know...

To me what it actually tells us is that, like I said earlier, something can be in two places at once. But perhaps this is wrong and somehow it is all predetermined at T1. But then wouldn't that violate QM to begin with?

 

[CJames]

carefully select a situation with three obervers A in the past of B , C in the future of B but A not in the past of C in which a measurement at A is going to influence the signalling from B to C.

I'm not going to immediately say you are wrong because I don't know the math well enough. However, if A is in the past of B but the future of C, wouldn't ANY form of communication following the path B->A->C be potentially superluminal? I think that it would and so I don't think that such a situation is allowed by special relativity in the first place.

 

[Careful]

Hi CJ, you must realize that there is usually much more to the things you learn than you teachers tell you. Let me first make some comments and then argue why the predictions of QM are very strange indeed.
(a) It is not SCIENTIFICALLY correct to state that Bell experiments refute local hidden variables/objective local theories. About this issue there is written an impressive book by Franco Selleri, and many papers about how Bell type experiments can be made to violate the orginal Bell and the CHSH inequality have appeared since 1964. Amongst these is the paper of Pearle (1970) in which the key result is that if your detector efficiency does not exceed 70% at relative angles of 90 degrees and 87% at relative angles of zero or 180 degrees, you can reproduce exactly the QM correlations for the pairs which are observed (these efficiencies are extremely high, experiments in which the efficiency is higher such as Rowe 2002 cannot maintain the locality assumption). Many subsequent papers have been written by Caeser (1984-1987), the eminent and late A.O. Barut, Santos, Marshall, Vaidman (and even Wigner before 1970) and many others. Up to date, there exists no experiment which excludes LHVT/OLT theories, you should scan the web upon the number of good papers written in 2005 which propose the next generation of loophole free tests (realize that experimentators are putting their best efforts in this for 35 years now!)
(b) the issue of the thread should be specified to : * is STANDARD QM inherently non local?*. It is actually very easy and much more natural to construct a Wave mechanics in which there is no entanglement at all: these are the Hartree equations. Actually there are at least four types of quantum theories present up to date : Standard, Hartree, GRW spontaneous reduction models, Consciousness nonsense... I do not mention different interpretations (although Bohm differs slightly from standard but makes the same predictions where standard can make them).
(c) I have mentioned before that you can construct causal, but not Bell local, hidden variable models where the predictions match exactly those of QM (so, no dector inefficiencies here), such a theory is backwards causation.

Now, let me argue why the predictions of Standard QM are very weird indeed:
the entangled state remains rotation invariant : what does this mean when both particles are clearly separated (you need a parallelism here)?? That is, the rotional invariance of the source is ``remembered´´ by the *individual* entangled pairs. Make the exercise by assuming that every pair has a definite allignment (ie. spin vector s and - s with the length of s equal to hbar/2), and suppose the v's are uniformly distributed over the sphere (this is what I would mean by rotation invariance). Now figure out a deterministic detection rule and suppose symmetry between both detectors and you will see that your correlations are a straight line if you assume separability of probabilities. So either the particles know (before they leave the cavity) what the dector settings will be in the future, meaning that you assume the world to be fully deteministic and that the particles have the ability to figure out all relevant parameters in this game and compute it, which is of course CRAZY. Or, the particles signal faster than with the speed of light, but we can only see the effect of those signals and not use them (something which I doubt very much - see my previous message). So, (a) a physical process which occurs faster than with the speed of light or (b) which travels backwards in time HAS to occur for the correlations of QM to come out right. This is very strange indeed... (note that standard QM says nothing intelligent about the measurement process which is supposed to provide us with these correlations)

 

[Careful]

A and B are particles which have interacted. We detect A. Since the motions of A and B are somewhat related subsequent to their interaction, then it follows that the detection of A can tell us something about how B might interact with C.
But they are used actively, aren't they? Quantum computing? :-)
The view that Nature is local fits the data. It's the simplest, reasonable explanation for why we can't communicate (ie., use the correlations actively in the sense that I take you meant this) superluminally.
Correlations beyond the lightcone are conceptually understood in terms of the conservation laws.
The paper you reference sounds interesting ... I must check it out.

Hi sherlock, the point is that the interaction between A and B has measurable consequences out of the lightcone of A.
Concerning quantum computing: it is impossibe to tell wether it are CLASSICAL correlations one is using or not (see my latest post).

Cheers,

Careful

 

[ttn]

No, I think you have it right.
For me personally, the confusion begins when you talk about an object outside the lightcone. Alice makes a measurement, which causes collapse of the shared wave function. Now you know something about something somewhere else, true, and that is outside the lightcone.

Wait, that isn't what Bell Locality says. You're forgetting that Bell Locality is defined in terms of conditionalizing all the probabilities on a *complete* description of the state. So it isn't just that the conditional probability changes when you learn something. That's trivial. Consider a simple example. Put a marble into a shoebox, then split the shoebox in half and carry the two halves (one of which contains the marble, but you don't know which one) to distant locations. What's the probability for alice to open her box and find the marble? 50% But suppose we specify that Bob has already looked in his box and has not found the marble. Suddenly the conditional probability for Alice's result (viz, conditional on Bob's outcome) jumps to 100% Is this a violation of Bell Locality? NO! Because the original probabilities (like the 50%) weren't of the right sort. They weren't conditional on the exact complete state of the system before the measurements -- which, for a simple classical example like this, obviously would contain the actual location of the marble. And if you include that, then the probabilities *don't* change when you add this information about Bob's result. If the marble was in Alice's box all along, the probability that she'll find it is 100% whether we specify Bob's outcome or not. So Bell Locality is respected here.

But what has happened that is really so weird? We project the knowledge we have back to the point at which the entangled particle pair was created. This is the same thing that happens when only one particle is involved, nothing strange about that. The particle acts as if it had that orientation from the last point something happened.

You might do this, but then this contradicts OQM, specifically the compelteness doctrine. Say a spin 1/2 particle is in the state |+z>. Then, later, you measure its x-spin-component and get +. Does that mean, "really", the particle had a positive x-spin-component all along, even during that time when the quantum state was |+z>? Not according to OQM! (But maybe according to some kind of hidden variable theory.)

Say Alice sees a V orientation with a polarizer at 0 degrees. Naturally, all subsequent measurements will be consistent in EVERY WAY with this knowledge AS IF it was always that way from the creation of the particle. So in that sense there is absolutely nothing happening outside any light cone.

In other words, all subsequent measurements will be consistent in every way with the assumption that the particle possesses definite spin hidden variables even when not being measured.

No wonder you're having trouble believing that OQM is non-local. You aren't willing to actually *accept* the completeness doctrine! You pay it lip service but then think as if the completeness doctrine were *false*!

In other words, all quantum measurements find a particle in an eigenstate and its eigenvalue is consistent with the quantum measurement rules. Entangled particles are no different in this respect. So the real question to me is: why does a measurement at time T2 cause the particle to assume a specific value as if it had that value at time T1 (where T1 is before T2) ?

Who says it does this? Certainly not OQM! In OQM the state just *is* defined by the quantum state, the wf. That's what the completeness doctrine *means*. And so the state changes -- the particle acquires some definite value for the property measured -- just when the wf collapses, i.e., just when the measurement is made. Not before. To say that this happens before is to say that it had a certain property *before* the measurement was made, i.e., before it was in an *eigenstate* of the operator corresponding to the property in question. And that is to posit hidden variables.

Does that make oQM non-local?

It doesn't make OQM anything, because you're not *talking* about OQM anymore!

Naturally, some of these issues show up in our definition of locality. You can see that there is no information transfer which is FTL, and there is no clear causal effect which is FTL. Yet the Bell Locality condition is violated with a strict application of its definition. So what does that condition actually tell us?

I think you need to get clearer on the definition of Bell Locality. Its violation *does* (at least according to Bell and many others) signal a "clear causal effect which is FTL". That's the whole point of that locality condition -- to test whether a given theory is "locally causal". You *really* need to read Bell's article "La Nouvelle Cuisine" if you want to understand this stuff.

 

[Careful]

A and B are particles which have interacted. We detect A. Since the motions of A and B are somewhat related subsequent to their interaction, then it follows that the detection of A can tell us something about how B might interact with C.
But they are used actively, aren't they? Quantum computing? :-)
The view that Nature is local fits the data. It's the simplest, reasonable explanation for why we can't communicate (ie., use the correlations actively in the sense that I take you meant this) superluminally.
Correlations beyond the lightcone are conceptually understood in terms of the conservation laws.
The paper you reference sounds interesting ... I must check it out.

Moreover, I am glad you state explicitely that you can use faster than light signalling in an operational sense. So standard QM is not local neither in the Bell nor operational sense. Moreover, I do not know precisely which conservation laws you are referring to. If you mean conservation of probability, then this has nothing to do with correlations over the lightcone. You can easily cook up non linear wave theories which still have a probablility current conservation law but give rise only to wave functions with support in the lightcone.

Cheers,

careful

 

[vanesch]

(a) It is not SCIENTIFICALLY correct to state that Bell experiments refute local hidden variables/objective local theories.

This is a correct statement. However, what is usually meant with the statement is that in those situations where quantum mechanics makes idealised predictions that DO violate the Bell inequalities, if we add to that the (quantum-mechanically sound, even though often not derived from first principles) *usual* experimental corrections of apparatus and detectors, then it would be highly surprising that quantum mechanics being wrong, it would have in it a kind of self-correcting mechanism where its ideal predictions are wrong, but its experimental corrections are just as wrong in the opposite sense such as to result in agreement between realistic QM predictions of the experiment (including corrections) and actual experimental results, and that is what is observed: agreement between realistic QM predictions and experimental results. So, barring a conspiracy, it is - to all experimental and scientific standards - a very reasonable working hypothesis that the QM predictions in this area are experimentally verified.

You are however, right, that many of these experiments do not have RAW DATA violating some Bell inequality. The assumption is that when the theory (QM) makes correct predictions concerning its realistic experimental predictions, combined with the overall success of QM in several other domains, that it is very plausible to take as established that QM is also correct in the ideal predictions (which do not correspond to actual experimental setups because of the non-idealities in the apparatus). This is the working hypothesis that is taken in this thread.

Amongst these is the paper of Pearle (1970) in which the key result is that if your detector efficiency does not exceed 70% at relative angles of 90 degrees and 87% at relative angles of zero or 180 degrees, you can reproduce exactly the QM correlations for the pairs which are observed (these efficiencies are extremely high, experiments in which the efficiency is higher such as Rowe 2002 cannot maintain the locality assumption).

This is correct. The experimental data of many experiments (I am not really following up all the latest developments), as such, as raw data, do not violate Bell's inequalities, and as such leave the door open to LR theories - often made up for the purpose.

Many subsequent papers have been written by Caeser (1984-1987), the eminent and late A.O. Barut, Santos, Marshall, Vaidman (and even Wigner before 1970) and many others. Up to date, there exists no experiment which excludes LHVT/OLT theories, you should scan the web upon the number of good papers written in 2005 which propose the next generation of loophole free tests (realize that experimentators are putting their best efforts in this for 35 years now!)

Although I have often seen the argument of LR proponents this way (that experimenters have tried a long time and STILL have no data violating Bell's inequalities), I think they miss the point - understandably, because the "publicity" of these experiments ALSO misleads. True, no RAW DATA excludes the possibility of a future LR theory.
However, ALL these raw data are IN AGREEMENT with the experimental predictions of an overall VERY SUCCESSFUL theory, quantum mechanics, and these experiments are challenging the QM predictions each time, in different situations. Each time, combining STANDARD experimental corrections (also rooted in QM) and predictions of QM, one arrives at agreement. So isn't it very reasonable to presume that after these gazillions of agreements between QM and experiment, the ideal predictions of QM ALSO are correct ?

This is the working hypothesis taken in this thread (and in fact in most of the QM threads here): QM makes experimentally correct predictions, also in those cases where the experiment has not been carried out.

It is actually very easy and much more natural to construct a Wave mechanics in which there is no entanglement at all: these are the Hartree equations.

I don't know exactly what you mean ? Do you mean, the effective potential models ?

 

[DrChinese]

No wonder you're having trouble believing that OQM is non-local. You aren't willing to actually *accept* the completeness doctrine! You pay it lip service but then think as if the completeness doctrine were *false*!
...

And so the state changes -- the particle acquires some definite value for the property measured -- just when the wf collapses, i.e., just when the measurement is made. Not before. To say that this happens before is to say that it had a certain property *before* the measurement was made, i.e., before it was in an *eigenstate* of the operator corresponding to the property in question. And that is to posit hidden variables.

Of course I accept that the WF is a complete description, and of course I don't believe in HV.

(I keep saying that it is "AS IF" and I am not trying to make a literal description. There are several different ways of visualizing what is happening. These are just images, clearly in oQM it is the formalism that rules.)

But there is a mystery about collapse that it would be desirable to know more about. You touch on it above. You say that the WF collapses upon measurement, and sure, this is standard. So when there are 2 entangled particles, which one cause the collapse - measurement of Alice or of Bob? Sure, the results are apparently the same regardless of which one "causes" the collapse. But again, that's the mystery. We have no specific rule that defines this. And again, that is what I am referring to when I say "I am confused about whether WF collapse is physical" etc.

As cjames mentions, the WF is in 2 places at once. So is that a non-local phenomenon? To me, I am not sure it is "non-local" in the sense of Bell Locality. But it might be non-local in another sense.

But I am a bit confused about your marble box example. I think what you are saying is: this example does not violate Bell Locality because adding the information about Bob's outcome does not actually change the probability of the outcome at Alice. Am I close? (Or maybe the example isn't that important, not sure.)