Bell Locality: New Paper Clarifies Arguments

[ttn]

I don’t find the measurement solution any more satisfying than the QM case. At least not until the two theories can predict different results that experiments can select between.

I wonder if you're not sure exactly what is meant by "the measurement problem." This is actually a serious problem for Orthodox QM, which literally gives two different dynamical rules for the evolution of wave functions (depending on whether or not a "measurement" is being made). The *problem* is that the theory does not say what constitutes a "measurement", so it is, to use Bell's phrase "unprofessionally vague and ambiguous." This is the problem that is supposed to be raised by Schroedinger's cat: if you follow the time evolution described by Schroedinger's equation, you get nonsense results like cats being in superpositions of alive and dead. Since we never *see* such states, that description must be wrong. The wave function must have collapsed, at some point in the development, to a more definite state. But where did this collapse occur? When we consciously registered the state of the cat? Or when some photons flew from the cat and interacted with our eyeballs? Or when some poison molecules interacted with the cat? Or when the hammer interacted with the vial of poison? etc. The theory just doesn't tell us where along this chain the wf collapses, i.e., where the normal Schroedinger time evolution gives way to the alternate "collapse" dynamics. Put another way, the problem is that OQM doesn't seem able to explain why experiments have definite outcomes; or rather, the only way it can explain why experiments have definite outcomes, is by importing some very dubious concepts (such as "measurement") into the fundamental laws of nature where they don't seem to belong. This is a serious foundational problem for the theory.

In Bohmian Mechanics, we simply do not have this problem. Because particles always have definite positions (even when not being "measured"), there is no problem whatsoever associated with measurements having definite outcomes. The needle on your detector ends up in some definite spot (registering some definite outcome) because it's made of *particles* and particles are always in some definite spot. So, there simply is no problem associated with measurement in Bohm's theory. Measurement is just another ordinary physical process, the same in principle (meaning, obeying the same dynamical laws) as any other "non-measurement" physical process.

Your worry that the two theories make the same predictions (and that it is therefore difficult to tell which one is right) is a completely different issue. Yes, it would be nice if the various theories made different predictions so we could just do the experiment and rule some of them out. But it's not so, so, if we're going to have an opinion about which theory is better, it has to be based on some criteria other than agreement with experiment. (For example, whether a given theory is plagued by the measurement problem, or whether it asks us to believe in gazillions of copies of unobservable parallel universes inhabited by mindless hulks, etc...) And by the way, this is not at all an abnormal thing in the history of science. Lots of times people have been confronted with differing theories which make the same predictions (at least for the time being).

 

[RandallB]

what is meant by "the measurement problem." ... a serious problem for QM, ... to use Bell's phrase "unprofessionally vague and ambiguous."
raised by Schroedinger's cat:

Schroedinger's cat issues have always seemed to me as a joke taken way too serious. Once everyone was having so much fun coming up with various scenarios too many started to take it as something real and the ‘cat was out of the bag’. The idea that HUP measurement issue can somehow be scaled up to be real at the macro view of an overly ego-centric / self-centric observer is to me pointless. The idea is more for popular book titles than serious science.
No rational reason to expect the HUP to factor up this way to a macro level. Or to expect someone must be obligated to look at the moon to be sure it remains in orbit for fear it may disappear were it failed to be observed. It’s just too silly to give that kind of extension serious consideration IMO.

Just as a non-local universal wave function running in some other dimension(s) so as to align real measurements to create apparent “weird-action-at-a-distance”, cannot be expected to show itself in some detectable “measurable” way in our 3-D experience to prove its viability.

Both theories have these things as non-local parts that at least to date can not been shown in detail in our local reality. If you insist on “seeing” that HUP shows itself in a cat rather than in phenomena. Than BM should produce the demonstration the reveals the universal wave function directly and standing alone, without relying on an after the fact explanation of a phenomena to infer it existence.

In this regard I still see no difference in them even with the “measurement problem”. Therefore, I don’t think either can “prove” the other wrong. (Even if the QM approach at least “seems” to have been the more practical in application so far.)

Remember Bell’s bias was FOR a deterministic solution to replace both BM & QM non-local theories like Einstein. And that his 'test' help point the way-- So far it has not.

 

[ttn]

Schroedinger's cat issues have always seemed to me as a joke taken way too serious. Once everyone was having so much fun coming up with various scenarios too many started to take it as something real and the ‘cat was out of the bag’. The idea that HUP measurement issue can somehow be scaled up to be real at the macro view of an overly ego-centric / self-centric observer is to me pointless. The idea is more for popular book titles than serious science.
No rational reason to expect the HUP to factor up this way to a macro level. Or to expect someone must be obligated to look at the moon to be sure it remains in orbit for fear it may disappear were it failed to be observed. It’s just too silly to give that kind of extension serious consideration IMO.

Just as a non-local universal wave function running in some other dimension(s) so as to align real measurements to create apparent “weird-action-at-a-distance”, cannot be expected to show itself in some detectable “measurable” way in our 3-D experience to prove its viability.

Both theories have these things as non-local parts that at least to date can not been shown in detail in our local reality. If you insist on “seeing” that HUP shows itself in a cat rather than in phenomena. Than BM should produce the demonstration the reveals the universal wave function directly and standing alone, without relying on an after the fact explanation of a phenomena to infer it existence.

In this regard I still see no difference in them even with the “measurement problem”. Therefore, I don’t think either can “prove” the other wrong. (Even if the QM approach at least “seems” to have been the more practical in application so far.)

Remember Bell’s bias was FOR a deterministic solution to replace both BM & QM non-local theories like Einstein. And that his 'test' help point the way-- So far it has not.

I can't make any sense of any of this, but I don't think it's worth pursuing anymore. Let's just agree to not understand each other.

 

[Sherlock]

ttn, I'm not sure what the most recent exchanges in this thread were about. Anyway, I just read Bohm and Bub's, "A Proposed Solution of the Measurement Problem in Quantum Mechanics by a Hidden Variable Theory".

I'm not finished rereading your papers yet, and am boinking this thread in the hope that some more knowledgeable (or at least less confused) people than myself will weigh in with detailed analyses and evaluations of your papers, especially the latest one on the nonlocal character of nature.

Wrt the Bohm and Bub paper, I finally feel that I'm on a track to understanding why the measurement problem is indeed a problem and why it can't be solved via quantum theory alone. I especially like their idea that, via quantum theory (at least the orthodox interpretation of it), one might get conceptually trapped without realizing that one is so trapped. However, I'm not sure I understand the nonlocal mechanism in the theory they propose.

 

[RandallB]

I can't make any sense of any of this, but I don't think it's worth pursuing anymore. Let's just agree to not understand each other.

OK, by me.
With the exception of not buying the claim that Bohm can solve a local measurement with a non-local solution, I’m satisfied with your explanation of BM and your consistent positions inside BM philosophy.
See-ya in another thread someday

 

[straycat]

mindless hulks?

You are free to just stipulate that all the different Alices living at different places along the 5th dimension are all equally real, equally conscious; none of them are mindless hulks. No problem.

The problem is then that statements about probability (which are rather important in QM) don't seem to have any meaning. That's what I was getting at before. So then it is not at all obvious how a model like yours can be said to agree with the QM predictions. This is a long-standing problem for MWI people (and is exactly why people like Patrick want to say that only one of the copies is the genuine article, and that which one this is is *random* according to Born's rule. This solves the problem of the meaninglessness of "probability" at the price of introducing mindless hulks).

Ahh! Perhaps we are getting somewhere in this discussion

I agree completely with the notion, as expressed by Patrick elsewhere, that the interpretation of probability from the perspective of the MWI is a long-standing problem -- perhaps its greatest problem. And I agree that if we use the Born rule to assign probabilities, then we are forced to the conclusion that only one of the copies is the genuine article; thus the rest must be mindless hulks.

But I think that the "mindless hulk" argument does NOT apply to Patrick's "alternate projection postulate" (APP). (Not sure if you or Patrick agree with this.)

So why does the mindless hulk argument apply to the Born rule and not to the APP? Here's why. Let's assume that all of the copies in all of the parallel worlds are equally conscious, equally real. If we apply the Born rule, then we find that most of the observers in most of the universes will "observe" the Born rule to be false. But if all observers are conscious, and most of them observe the Born rule to be false, then we must conclude that QM is just "wrong" for most observers. Which of course it's not. So the only way around this difficulty is to assert that the ones that observe the Born rule to be false are hulks; the only non-hulk copies are the ones that exist in those few worlds where the Born rule is true. iow we are forced to give up our assumption that they are equally conscious.

But if we assume the APP, we don't have that problem, because the measure of the worlds in which an observer will conclude that the APP is true approaches unity in the limit of a large number of measurements. iow, "most" observers observe that the APP is valid. So there is no need to go around claiming that most of them are mindless.

This is precisely the reason that my "saxion-esque" scheme -- the one you recall that we are assuming hypothetically for the sake of this thread to be successful in the derivation of the Schrodinger equation by imposing a few mathematical constraints over the M in my toy model -- takes the APP as the fundamental "rule" for calculating probabilities. The Born rule is understood to be a valid coarse-grained approximation to a situation which is fundamentally governed by the APP at the fine-grained level. I would also point out that, in addition to the hypothetically successful saxionesque development of my toy model, there are at least two independent proposals in the literature of ways to take the MWI, assume the APP at the fine-grained level, and show that the Born rule is valid at the coarse-grained level. (These are by Weissman and Hanson -- see discussions on some of Patrick's threads.)

So the ontological point to be made here is that it is, at the least, hypothetically possible to have an MWI model that does NOT require mindless hulks. So on this one issue, I see MWI and Bohmian mechanics as being on equal footing. (To be honest I have not considered the issue of mindless hulks and BM, but for the sake of argument I am willing to concede that BM does not require MH's.)

David

 

[straycat]

Except that a causal effect propagating at c in 5 dimensions could lead to superluminal actions at a distance as seen from our everyday 4 dimensions.

How exactly do you come at that conclusion?

Let's suppose that we have a causal effect propagating at c in 4 dimensions -- iow, assume classical relativity. Does this lead to superluminal actions at a distance as seen via consideration of a 3 dimensional hypersurface? I don't think so, unless I'm missing something. If going from 4 to 3 doesn't pose a problem, then I don't see why going from 5 to 4 does.

 

[straycat]

This phrase is confusing. Does it mean that your model violates Bell Locality (i.e., is not Bell Local)?

I see that there is some confusion over semantics. So let me try to clarify my terms so we can communicate.

This contradicts the way you were talking about it earlier. Before, you said that this "sudden change in the wave function" is *really* only an updating of knowledge. (Basically, there existed local hidden variables which determined the outcomes.) But such a model does *not* violate Bell Locality. A sudden change in the wave function at a distant location only involves a violation of Bell Locality if the wf is a physically real thing; if (as I thought you claimed earlier) the wf is merely a summary of our (incomplete) knowledge of the real physical state of affairs, then its change does not involve any nonlocality.

So which is it?

[...]

Either the wf is (what bell called) a "beable", or it isn't. If it is, then a sudden change in its value over there caused by something you did over here, means that Bell Locality is violated. If not, not.

[...]

Then why do you say this model violates Bell Locality? Sounds to me like it doesn't. But then my earlier question remains: how exactly do you think you're going to explain the QM predictions for correlations b/w entangled particles?

Here's what I am claiming about my model:

1. It is local in exactly the same way that GR is local, except that we have 5 dimensions instead of 4. Randall has argued that GR is not "local," but it is "background independent." Gosh darnit, you guys are a tough crowd. OK, maybe I'm using the wrong terminology here; I'll address that in a different post. Let me at least stick to the minimal claim: there are no signals FTL in my model. Whatever you say about GR, you can say about my model.

2. The wf is demoted to the status of not-a-beable. So a sudden change is no big deal. Nevertheless, it is still a useful mathematical entity that helps us do calculations.

3. Experiment tells us that Bell's inequality is violated, and my model is consistent with this experimental observation. Earlier, when I said that my model is "consistent with Bell nonlocality," I was referring to the content of the former sentence. IOW I was using this definition:

a model is consistent with Bell nonlocality <==> the model makes predictions that match the outcomes of Aspect experiments

But I see now that you are retaining the term "Bell nonlocality" to apply only to models in which the wf is real. So let me retract my claim of "consistency with Bell nonlocality," and replace it with the claim: "my model predicts the outcome of the Aspect experiments"

4. My model does not require mindless hulks.

5. My model is complete. As I stated earlier, Alice knows exactly where she is in M, and that tells us everything there is to know about her state.

6. My model employs a nonlocal HV. As stated earlier, the "variable" in quesion represents the 4-d hypersurface W. It's "hidden" to Alice because there is an ensemble of them relative to her state, none of which are priveleged over the others in the ensemble.

7. My model is a simplified version of some of the attempts at QG, like LQG, as I stated in an earlier post.

The only reason any sane person takes MWI seriously at all is that it seems to be the only way to explain (well, pseudo-explain) the data without accepting nonlocality and thus rejecting relativity. You said your model violates Bell Locality. If that's right, then Occam's razor desperately wants to slash off your model.

You are correct, I believe, that one argument in favor of the MWI is that the MWI does not ask us to reject relativity. Correct me if I'm wrong, but doesn't BM ask us to reject relativity since wf collapse propagates FTL? Are you still a proponent of Lorentzian relativity because of this issue -- despite the absence of evidence for LR? Isn't this issue just a little bit confusing in the BM world?

Using your terminology, my model does not violate Bell locality; therefore (according to what you said above) my model escapes Occam's razor on this point. Actually I don't follow what you said above. If a model violated BL, then Occam's razor wants to slash it off ...?? I don't follow. BM violates BL, so does Occam's razor want to slash it off??

Assuming that my model can in fact be seen as a pared-down model of some versions of QG (at least the background-independent ones, like LQG), then my model is superior to BM in this sense. BM equivocates on whether relativity should be abandoned; my model does not. By Occam's razor, if we have one model -- mine -- that gives us relativity and QM in one package, and we have another -- BM -- in which QM and GR must be delivered separately, then Occam's razor would favor the former.

 

[straycat]

From reading Bell himself instead of interpretations about him (most of those neglect to point out that Bell believed in “unknown variables”) I see no real difference in Einstein vs. Bell local.

Again, I have seen many people state (erroneously) that Bell ruled out hidden or "unknown" variables, and I am glad to see Travis leading the charge to correct this misunderstanding.

There is a big difference between Bell and Einstein local, though, by my way of understanding. OK, maybe I'm using the word "local" to mean something different that what you mean. So I'll tell you the content of what I see as a very significant difference between two very different concepts: Einstein tells us that signals cannot propagate FTL. QM tells us (and Bell proved it) that collapse of the wf CAN and DOES propagate FTL -- even though that cannot be used to transmit information FTL. I don't know if you have a name for these features of GR and QM -- if you do, please tell me; if you don't, then perhaps you should, because they are (imho) important enough concepts to have a name. I have always defined the terms like this:

Einstein locality = signals cannot propagate FTL

Bell nonlocality = collapse of the wf DOES propagate FTL, even though it cannot be used to transmit signals FTL

As you can see from the definitions, it is not contradictory to claim that nature is Einstein-local and Bell-nonlocal at the same time. Einstein locality and Bell nonlocality (as I defined above) are not mutually exclusive. Nevertheless, they are DIFFERENT concepts, and it behooves us to understand the difference. If we don't, then we find ourselves wanting to do something crazy like advocate for Lorentz relativity even though there is no experimental evidence in its favor <straycat ducking in anticipation of ttn's counterattack ...>

Having said that, I am perfectly willing to consider that my above-stated definitions are nonstandard, that I should redefine my terminology. How about this:

background independence = signals cannot propagate FTL

Bell nonlocality = collapse of the wf DOES propagate FTL, even though it cannot be used to transmit signals FTL. Furthermore, as Travis stipulates, the wf is real.

No I don’t.
I accept GR is background independent (Ref: “The case for background independence” Lee Smolin/ Perimeter)

SR and Minkowski space-time (As a flat 4-D representation of Classical SR) as classical theories are background dependent (Although Minkowski I believe disagrees that his was actually classical) are able to hold the “unknown variable”. Einstein and Bell both always hoped that variable would be able to be demonstrated as real & local somehow (Speakable – Unspeakable; Bell).

Other than to incorrectly claim “local”, I really don’t see where GR applies here at all.
As soon as you introduce anything in an additional dimension that can collapse or link between two otherwise space-like separated events you are by definition not using a local realist, a requirement for both Einstein and Bell Local. Suggest you review Bell’s own writings again.

OK, the difference between "local" and "background independent" is something that I have not learned to appreciate, but perhaps I should. So for the time being let's state that GR is background independent. All I am claiming is that my toy model is likewise background independent, in the same sense and for the same reasons that GR is background independent. I do not see how or why the addition of a 5th dimension should change that.

David

 

[ttn]

How exactly do you come at that conclusion?

Let's suppose that we have a causal effect propagating at c in 4 dimensions -- iow, assume classical relativity. Does this lead to superluminal actions at a distance as seen via consideration of a 3 dimensional hypersurface? I don't think so, unless I'm missing something. If going from 4 to 3 doesn't pose a problem, then I don't see why going from 5 to 4 does.

Consider a sheet of paper folded (almost) in half. A signal propagating through 3D at speed c can get from one side to the other in hardly any time at all. Observers who don't know about the 3rd dimension would assume that the effect had propagated the long way round, in the sheet of paper, which would of course require it to be superluminal.

 

[ttn]

Correct me if I'm wrong, but doesn't BM ask us to reject relativity since wf collapse propagates FTL?

"wf collapse" plays no role whatsoever in bohm's theory. It doesn't even really happen; it's just something physicists are entitled to do for their own convenience under certain circumstances. Yes, Bohm's theory is explicitly nonlocal (as must be any theory agreeing with experiment) and thus suggests that relativity must be, in some sense, given up. But to understand what this nonlocality consists of in Bohm's theory, I guess you need to understand better how that theory actually works.

All the stuff about your "model" is just word salad... not even wrong, as they say. Sorry. But I don't see any point in discussing it further.

 

[straycat]

Consider a sheet of paper folded (almost) in half. A signal propagating through 3D at speed c can get from one side to the other in hardly any time at all. Observers who don't know about the 3rd dimension would assume that the effect had propagated the long way round, in the sheet of paper, which would of course require it to be superluminal.

You are assuming the existence of some sort of large-scale nontrivial topology to M, whereby a region "over here" is folded over and connected to some distant region "over there." I never stated my model works that way.

 

[RandallB]

There is a big difference between Bell and Einstein local,
Einstein locality = signals cannot propagate FTL

Bell nonlocality = collapse of the wf DOES propagate FTL,

Are you joking? Comparing Einstein vs. Bell Local by defining Bell Non-local??
Is this some kind of slight of hand with words or are you just fooling yourself?
You need to do much better than that. Like define Bell Local, not something it is not (Nno-L). Have you read Bell? He was not making an argument for a collapsing wave function.
Define “Bell Local” then compare that directly with your idea of Einstein Local.

Other than what they have not shown,
for you what exactly have the Bell tests shown if anything?

As to ‘background independence’ <> FTL. Not on point.
I’d recommend you Google Scholar it with ‘Smolin’ & ‘Perimeter’ to find his paper(s) and take some time with it before making random speculations about independence. It’s not that simple a concept.