OK Corral: Local versus non-local QM

[wm]

Preamble: I have yet to find a reason to abandon my support of a LOCAL interpretation of QM; especially in relation to EPRB. It is therefore my hope that we might here bring the issue to a head. Given the frequent verbal misunderstandings across the LOCAL--NONLOCAL divide, it is also my hope that we might resolve the issue in the beautiful (and usually clearer) language and logic of mathematics.

PS: I suggest we constrain our discussion on this thread to the EPR-Bohm experiment with spin-half particles; it being the source of Bell's (1964) theorem. So:

On another thread, I (wm) wrote:

Dear vanesch, I appreciate your defence of MWI but it still does not make sense to me (though I am a hardened LOCALIST):

1. You say: such as non-conspiracy in nature which requires random events to be statistically independent if there is no causal link somewhere, in one way or another.

I thought that the experimental results reveal statistical (= LOGICAL) dependence? Which is what I would expect, given that the ''entangled particles'' (created by the conservation of angular momentum) represent some of the highest correlations that we can generate (ie, the particle-state is spherically symmetric). Given then that the measuring-devices are also HIGHLY correlated by their differential setting: the detected correlations appear to be LOCAL and non-mysterious to me. What am I missing?

2. To help my understanding of MWI: Let us say that I come to Paris (to discuss MWI with you) and we agree to toss a coin to decide who pays for dinner. As the coin arcs through the air, I guess we agree that it is in a superposition? On coming to rest, the coin reveals a definite result.

It seems to me that you and I remain physically in the world with the definitive physical result (the world where we expect to have dinner): AND that no other real world exists. Why then the need for MW in MWI?

Thanks, wm

DrChinese replied:

The measuring devices in Bell tests are sometimes set while the entangled particles are mid-flight. Therefore their settings cannot be causally correlated.

The mystery is this: was the outcome of a particular detector setting determined when the particle pair was created? If you say YES, then you run afoul of Bell: because then the particles must be carrying enough "answers" to match all possible detector settings (Bell shows this cannot be true). If you say NO, then how does the answer at one point get transmitted to the other point?

(Emphasis added.)

1. Causally correlated?? DrC, note that I said that the detector settings are correlated by their diifferential setting. If detector A on the left is set at a (a unit vector) and detector B' on the right is set at b' (another unit vector), then the detector settings are correlated by a function of (a, b'): that is, by a function of the differential setting.

For example: If we took cos(a, b') as the correlation function; +1 indicates a parallel setting; -1 indicates an antiparallel correlation; etc.

Of course the correlation that counts is that established by the respective detector-settings at the instant of arrival of each particle: set in any manner of your choosing.

2. Was the outcome of a particular detector setting determined when the particle pair was created?? The outcome was determined when the particle and the detector interacted LOCALLY.

So could I ask you to provide the detailed maths by which you derive the EPRB correlation function? So that I might see where you believe that nonlocality enters (in the language of maths.)?

3. How does the answer at one point get transmitted to the other point? ??

Well, in that we have not introduced Alice and Bob to the discussion thus far: it does not get transmitted! How could it?

Trust this all helps to convince you of the need for your mathematical derivation of the EPRB correlation function,

With best regards, wm

 

[Hurkyl]

Why then the need for MW in MWI?

I can answer this one.

There is a lot of information in the universe. There is information about Alice's laboratory, there is information about Bob's laboratory, there is information about the Andromeda galaxy, etc.

Alice can measure anything in her laboratory, but she can't (directly) measure anyting in Bob's laboratory. Nor can she directly measure anything in the Andromeda galaxy.

When we take the state of the universe and discard the information that isn't (directly) accessible to Alice, what remains is what we call Alice's world.

In particular, it isn't like Star Trek where we have a parallel "universe" in which our evil twins live.


If the entire universe consisted of the pair of photons in the experiment, then the state of the universe might be

(3/5) |01> + (4/5) |10>

Alice's world is the first coordinate. If we take the partial trace along the second coordinate to obtain the state of Alice's world, we get the statistical mixture

36% chance of |0>
64% chance of |1>

Similarly, Bob's world is the statistical mixture

64% chance of |0>
36% chance of |1>

 

[wm]

Clarifying the MW in MWI?

I can answer this one.

There is a lot of information in the universe. There is information about Alice's laboratory, there is information about Bob's laboratory, there is information about the Andromeda galaxy, etc.

Alice can measure anything in her laboratory, but she can't (directly) measure anyting in Bob's laboratory. Nor can she directly measure anything in the Andromeda galaxy.

When we take the state of the universe and discard the information that isn't (directly) accessible to Alice, what remains is what we call Alice's world.

In particular, it isn't like Star Trek where we have a parallel "universe" in which our evil twins live.


If the entire universe consisted of the pair of photons in the experiment, then the state of the universe might be

(3/5) |01> + (4/5) |10>

Alice's world is the first coordinate. If we take the partial trace along the second coordinate to obtain the state of Alice's world, we get the statistical mixture

36% chance of |0>
64% chance of |1>

Similarly, Bob's world is the statistical mixture

64% chance of |0>
36% chance of |1>

Dear hurkyl, I appreciate this answer very much!

BUT I wonder: How long before someone comes to clobber us in this common belief. Dare I summarise that belief as: Many private local worlds, one common local universe?

AND I wonder that I've not seen such a clear delineation of the MW in MWI before? That is: I wonder if dedicated MWIers will accept it?

For, if I'm not mistaken, you allow that the overlap in Alice and Bob's private worlds may increase when they get on the phone to discuss their respective results: in full accord with locality.

And in full accord with the common-sense response to DrC's question re information (answers) about Alice's result (in Alice's world) being transferred to Bob (in Bob's world); and vice-versa: There being no virtual/mysterious/magical/non-local/Star-Trek universes built from virtual Alices and Bobs ... ad infinitum.

Hoping I've understood you correctly, many thanks, wm

 

[Demystifier]

There are local and nonlocal *interpretations* of quantum mechanics (QM).
But what can we say about QM (non)locality independently on the interpretation?
We can say that a (many-particle) state is described by a wave function of the form psi(x_1,...,x_n), i.e., that, in general, there is no wave function for each particle, but that there is only one wave function describing all particles together. Isn't that an interpretation-independent sign of NONlocality?

Of course, if you start to deal with interpretations, you may say that quantum physics is only about correlations, and that only the correlations are the entities which are nonlocal. But if the only entities quantum mechanics is about are nonlocal, then obviously quantum mechanics IS nonlocal, isn't it?

And if QM is *not* only about the correlations, then what is it about? Whatever *clear* answer to that question you choose, it seems that you cannot avoid the conclusion that this thing must be nonlocal. The only way to save locality is to advocate an interpretation that avoids a clear answer to the question - what is QM about?!

 

[Demystifier]

Let me also explain why I claim that correlations are nonlocal. If two particles spacially separated are correlated, then this correlation is nonlocal. Clearly, in the EPR(B) setup this is the case. It is also clear that the Schrodinger equation allows entangled wave functions without local interactions between particles. In that regard see also
http://arxiv.org/abs/quant-ph/0304031

 

[vanesch]

I've been giving my PoV on this issue at least a dozen of times, so I'm not going to do so here again...

 

[Demystifier]

I've been giving my PoV on this issue at least a dozen of times, so I'm not going to do so here again...

Then let me put a summary of my view of your PoV. It is MWI and local and is in agreement with what I said above. Namely, it involves a fundamental role of conscious observers, but their role is not *clearly* explained. Instead, their role remains somewhat mysterious. In other words, just as I said above, you save locality by avoiding a *clear* answer to the question - what is QM about?!

But of course, it does not mean that your PoV is wrong. It may be a fundamental property of consciousness that it is uncomprehensible and logically unexplainable. In that case, QM can indeed be local, while apparent nonlocality can be merely an artefact of misleading attempts to find a non-existing complete comprehensible model of nature.

 

[Demystifier]

Let me also say the following. If QM is both local and comprehensible, then there must exist a formulation of QM in which this locality is manifest in *every* equation. In particular, it should not contain many-particle wave functions that cannot be reduced to a collection of single-particle wave functions. But no one ever constructed such a formulation of QM. Thus, I conclude that QM is either nonlocal or noncomprehensible (or possibly both). This conjecture is nothing but a somewhat stronger variant of the well-known *theorem* (Bell, Hardy, ...) that QM is either nonlocal or nonrealistic (or possibly both).

But if you accept the option that QM is noncomprehensible, then you come to the teritory of religion-like arguments. Whenever scientists give a logical argument against some religion dogma, religious people reply that the logical argument does not disprove anything, because God is not comprehensible by humans. I respect such arguments as long as they are used consistently, but then physicists advocating locality of QM should not pretend that they find QM fully comprehensible. Locality is compatible with QM in the same way as religion is compatible with science.

 

[Demystifier]

What I cannot understand is WHY some physicists so desperately seek for a local interpretation of QM? What is so sacred about locality? Before getting familiar with Einstein theory of relativity, (almost) nobody had problems with accepting the Newton nonlocal theory of gravity or the Coulomb nonlocal law of electrostatics. So, why locality could not be only an approximative principle valid only in a restricted domain? Why many physicists find so dangerous or even unacceptable to have nonlocal laws of physics?

 

[DrChinese]

Preamble: I have yet to find a reason to abandon my support of a LOCAL interpretation of QM; especially in relation to EPRB. It is therefore my hope that we might here bring the issue to a head. Given the frequent verbal misunderstandings across the LOCAL--NONLOCAL divide, it is also my hope that we might resolve the issue in the beautiful (and usually clearer) language and logic of mathematics.

1. Causally correlated?? DrC, note that I said that the detector settings are correlated by their diifferential setting. If detector A on the left is set at a (a unit vector) and detector B' on the right is set at b' (another unit vector), then the detector settings are correlated by a function of (a, b'): that is, by a function of the differential setting.

I am not defending non-locality. I simply defend Bell's Theorem. I would be willing to discuss the topic in this light. Basically, Bell says that:

No physical theory of local Hidden Variables can ever reproduce all of the predictions of Quantum Mechanics.

Further, experimental results are within the range of the predictions of QM. If you are a proponent of locality, then you should reject the existence of hidden variables. If you are asserting that there ARE local hidden variables, then you should demonstrate that Bell's Theorem is wrong. The proof of Bell's Theorem has been presented many times, and I provide a couple of versions of it on my website:

Bell's Theorem (following Bell)

Bell's Theorem (following Mermin)

 

[vanesch]

What I cannot understand is WHY some physicists so desperately seek for a local interpretation of QM? What is so sacred about locality? Before getting familiar with Einstein theory of relativity, (almost) nobody had problems with accepting the Newton nonlocal theory of gravity or the Coulomb nonlocal law of electrostatics. So, why locality could not be only an approximative principle valid only in a restricted domain? Why many physicists find so dangerous or even unacceptable to have nonlocal laws of physics?

The one and only reason is the one you give yourself: relativity. Its basic principles require locality, and it is difficult to restore all the confirmed results of relativity in a natural way (I mean: following from some principle, and not one by one put in by hand to make things come out AS IF) if you violate locality in interaction.

After all, the idea is to distill some fundamental principles on which to build physics. If we do not do this, and allow for a "deus ex machina" each time that we *seem* to have a principle, but which we'd like to violate somehow, which makes things come out nevertheless AS IF that principle were valid, then you can just as well go all the way: there are no laws of physics. There's just a big bag of events, and there happen to be correlations between them, but there's no real reason for that, there's no structure behind it, and if we find some, then that's a funny coincidence. What we think are cause-effect relationships are just coincidences, and their apparently systematic relationship is just by good (or bad) fortune. Tomorrow, things may be totally different. Or not.

So, if we want to avoid the above viewpoint (which, after all, could be "true"), we'd better try to find rigorous principles, and be wary each time we find funny relationships that don't seem to rely on such a principle. This quest is not guaranteed to work, of course. But when we HAVE such a principle, and we have no absolute reason to reject it, we shouldn't reject it carelessly! Now, the relativity principle seems to be ok: there's no known, observed violation of anything that's derived from it, and it is extremely powerful in explaining a lot of stuff. So that's why people are very reluctant to dump it for no good reason but just some bad feeling about certain interpretations of quantum theory.

Now, that said, Newton himself had a lot of troubles with his own "action-at-a-distance" ! In his principia, he states:

That one body may act upon another at a distance through a vacuum without the mediation of anything else, by and through which their action and force may be conveyed from one another, is to me so great an absurdity that, I believe, no man who has in philosophic matters a competent faculty of thinking could ever fall into it

So, locality does have some attractiveness, apart from its necessity in relativity.

 

[Demystifier]

Vanesch, I agree, relativity looks as a nice and general physical principle with which we could start.
But the existence of objective reality also looks as a nice and even *more general* physical principle with which we could start as well.
On the other hand, the Bell theorem proves that at least one of these two nice principles is wrong.
Isn't it more natural to retain the more general principle and to crucify the less general one?
In addition, if you abandon the general principle of the existence of objective reality, then the existence of spacetime may not be objectively real as well, which means that even the principle of relativity (or locality) may not be objectively real, so you may lose *both* nice principles, which is certainly not what you want. (For an expanded version of this idea see also
http://arxiv.org/abs/quant-ph/0607057 )

 

[Hurkyl]

That is: I wonder if dedicated MWIers will accept it?

I'm not sure about the semantics anymore: I (thought I) had gotten it straight from Wikipedia, but I don't see it there anymore. MWIers will certainly agree with what I said, but they might disagree that's the definition of a "world".

 

[Schrodinger's Dog]

There are local and nonlocal *interpretations* of quantum mechanics (QM).
But what can we say about QM (non)locality independently on the interpretation?
We can say that a (many-particle) state is described by a wave function of the form psi(x_1,...,x_n), i.e., that, in general, there is no wave function for each particle, but that there is only one wave function describing all particles together. Isn't that an interpretation-independent sign of NONlocality?

Of course, if you start to deal with interpretations, you may say that quantum physics is only about correlations, and that only the correlations are the entities which are nonlocal. But if the only entities quantum mechanics is about are nonlocal, then obviously quantum mechanics IS nonlocal, isn't it?

And if QM is *not* only about the correlations, then what is it about? Whatever *clear* answer to that question you choose, it seems that you cannot avoid the conclusion that this thing must be nonlocal. The only way to save locality is to advocate an interpretation that avoids a clear answer to the question - what is QM about?!

Or to presume that your interpretation of QM is wrong in some as yet undefined way. Not that I think it is definitely, but it's a possibility; that only our practically limited understanding of what exactly is going on, is clouding our theoretical approach to what is actually going on, in other words we're making incorrect assumptions.

On the other hand of course we could just assume that QM is correct and there is something enigmatic(not necessarily a hidden variable as we might think of it, but something intrinsic we're missing) Which with our current technology we cannot reveal, that would make sense of the seemingly unusual or non common sense ideas.

Perhaps we need to decouple ourselves from current reason and accept that probability or energy and matter may well work in as yet incomprehensible ways we cannot yet grasp at our current level. Maybe we're unfortunate enough to be living in a time where our reach exceeds our grasp?

I honestly have no idea, and probably will not have any more the more I come to learn about QM in depth.

 

[Demystifier]

Schrodinger's Dog, but there *is* something very general that is proved rigorously by the Bell theorem:
If the results of measurements are manifestation of *some* objective reality (*whatever* this reality might be), then this reality must be nonlocal. This generality of the Bell theorem expressed by my words "some" and "whatever" is what makes this theorem so strong.

 

[Schrodinger's Dog]

Schrodinger's Dog, but there *is* something very general that is proved rigorously by the Bell theorem:
If the results of measurements are manifestation of *some* objective reality (*whatever* this reality might be), then this reality must be nonlocal. This generality of the Bell theorem expressed by my words "some" and "whatever" is what makes this theorem so strong.

Oh I agree given what we currently know the Bell theorem makes a pretty strong case for non locality, that's not something I would presume to argue about as it's a pretty air tight thought experiment, the only thing I suggested is that it might not necessarily be based on a true picture, and therefore we could be making flawed assumptions, which of course is entirely speculatory.

We assume that there is something based on our logic system at work, without knowing if there is something else going on we can't possibly account for given our current technology, for example, it's like Newton looking at light without understanding it's duality so he claims it's a particle, and then Young claims it must be a wave because of his experiment, both are wrong, because both don't have the bigger picture.

This doesn't mean there is some hidden variable or locality exists with QM, but I think it's possible that QM could be mistaken in some key area so I don't discount anything without having the 100% picture. Obviously this is just an example where nothing is 100% air tight, particularly when we don't have all the books in our collection.

I'm not disagreeing with anything, just keeping an open mind.

If Bell's theorem is not falsifiable then it is not a theorem

 

[Tomsk]

I have a question- how can you test for realism? Is it falsifiable? You would say, realism is a false assumption if everybody does not 'see the same elephant', but if 'everybody hears everybody else stating that they see the same elephant they see', then you can't then know if the elephant has an objective reality. It certainly seems that the elephant is an objective thing, say from A's point of view, especially if B is saying it's an elephant, but a completely different consistent set of obsevations is perfectly possible- B seeing a wombat, and hearing A say that it's a wombat.

Locality is certainly falsifiable, and has been tested, so it would surely make more sense to get rid of realism than locality.

 

[complexPHILOSOPHY]

I have a question- how can you test for realism? Is it falsifiable? You would say, realism is a false assumption if everybody does not 'see the same elephant', but if 'everybody hears everybody else stating that they see the same elephant they see', then you can't then know if the elephant has an objective reality. It certainly seems that the elephant is an objective thing, say from A's point of view, especially if B is saying it's an elephant, but a completely different consistent set of obsevations is perfectly possible- B seeing a wombat, and hearing A say that it's a wombat.

Locality is certainly falsifiable, and has been tested, so it would surely make more sense to get rid of realism than locality.

Is the elephant pink?

 

[vanesch]

Schrodinger's Dog, but there *is* something very general that is proved rigorously by the Bell theorem:
If the results of measurements are manifestation of *some* objective reality (*whatever* this reality might be), then this reality must be nonlocal. This generality of the Bell theorem expressed by my words "some" and "whatever" is what makes this theorem so strong.

Yes, this is about correct (if you make the implicit hypothesis of no super-determinism, but which I think is necessary to make if we are going to draw ANY conclusion from observation): one has to choose between giving up the objective (and unique!) existence of outcomes of measurement, OR non-locality. Now, given that we can more or less cope with the first without hurting the known principles of physics (namely, MWI), but that, when accepting non-locality, we have to throw out the whole formal machinery of relativity, I would say that the requirement of objective and unique outcomes is for sure an esthetic and philosophically very satisfying one, but not absolutely strictly required by what we know, formally, about physics. However, the second (locality), is.

So it seems that Bell + experimental confirmation (which is suggestively strong, but I'm not sure it is 100% watertight) makes us choose between:
- a philosophically satisfying concept, that "what we see, IS there, and ONLY that" OR
- the formal principle of relativity.

I would say that as a human being, I'd go for the first, but as a physicist, I go for the second. Because the first principle doesn't bring me any FORMAL confort, while the second one is formally useful... at least as long as there isn't another principle from which we can get the *results* of relativity without its founding principle (which is the 4-dim spacetime manifold).

 

[vanesch]

I have a question- how can you test for realism? Is it falsifiable? You would say, realism is a false assumption if everybody does not 'see the same elephant', but if 'everybody hears everybody else stating that they see the same elephant they see', then you can't then know if the elephant has an objective reality. It certainly seems that the elephant is an objective thing, say from A's point of view, especially if B is saying it's an elephant, but a completely different consistent set of obsevations is perfectly possible- B seeing a wombat, and hearing A say that it's a wombat.

I love that :!)
It is exactly my critique on Rovelli's RQM.

 

[complexPHILOSOPHY]

I love that :!)
It is exactly my critique on Rovelli's RQM.

The argument regarding 'the pink elephant,' deconstructs objectivity into a collective agreeance of subjective perceptions and it seems rather sound.

I hate you Berkely.

 

[vanesch]

The argument regarding 'the pink elephant,' deconstructs objectivity into a collective agreeance of subjective perceptions and it seems rather sound.

The point is: in order for the :"we all agree over the pink elephant" to make any progress in the EPR-Bell dilemma, we need to say that at a certain point, Alice and her friends all agree on a pink elephant, Bob and his friends all agree on a blue elephant, and Alice and her friends all agree on the fact that Bob hasn't seen an elephant. Because if they do, they can assign a probability distribution to a hidden variable ("what Bob and his friends have seen"), and Bell bites us.

 

[complexPHILOSOPHY]

The point is: in order for the :"we all agree over the pink elephant" to make any progress in the EPR-Bell dilemma, we need to say that at a certain point, Alice and her friends all agree on a pink elephant, Bob and his friends all agree on a blue elephant, and Alice and her friends all agree on the fact that Bob hasn't seen an elephant. Because if they do, they can assign a probability distribution to a hidden variable ("what Bob and his friends have seen"), and Bell bites us.

I actually never thought about it in that fashion. That is actually pretty intriguing, although my exposure and knowledge of QM is amatuer at best.

 

[DrChinese]

...So it seems that Bell + experimental confirmation (which is suggestively strong, but I'm not sure it is 100% watertight)...

Just a note for anyone who thinks this means that QM might not be "right": the leeway on this is VERY small - multiple standard deviations of agreement between QM and actual results.

Also, the comparison point for local realistic theories is severely problematic as well; because there are no such theories currently in existence that come close to actual results. I.e. the closest you can come to actual results - and still be within the Bell Inequality - does not match to any existing local realistic interpretation. That would be (more or less) a straight line function which agrees with QM at theta=0, 45 and 90 degrees but is slightly different at all other angles. (QM, of course, predicts the COS^2 function and that is what is observed.) So, a local realistic theory not only needs to match the Bell Inequality, it also needs to show how it gets there as a function of theta. This is not so easy, as the Bell Inequality is essentially 1-(theta/90 degrees) for correlated Alice and Bob. You then need "experiment error" and bias to account for the difference between this and actual results.

So don't think for a second that experimental problems in Bell tests will allow you to return to local realistic positions. This avenue creates as many problems as it purports to solve.

 

[wm]

Just a note for anyone who thinks this means that QM might not be "right": the leeway on this is VERY small - multiple standard deviations of agreement between QM and actual results.

Also, the comparison point for local realistic theories is severely problematic as well; because there are no such theories currently in existence that come close to actual results. I.e. the closest you can come to actual results - and still be within the Bell Inequality - does not match to any existing local realistic interpretation. That would be (more or less) a straight line function which agrees with QM at theta=0, 45 and 90 degrees but is slightly different at all other angles. (QM, of course, predicts the COS^2 function and that is what is observed.) So, a local realistic theory not only needs to match the Bell Inequality, it also needs to show how it gets there as a function of theta. This is not so easy, as the Bell Inequality is essentially 1-(theta/90 degrees) for correlated Alice and Bob. You then need "experiment error" and bias to account for the difference between this and actual results.

So don't think for a second that experimental problems in Bell tests will allow you to return to local realistic positions. This avenue creates as many problems as it purports to solve.

DrC, surely you have mis-spoken? Or are badly mistaken? Even just plain wrong?

You write: ''So, a local realistic theory not only needs to match the Bell Inequality, it also needs to show how it gets there as a function of theta.''

Here is the background to my questioning:

1. QM is correct.

2. Bell-tests are correct.

So, in response to your claim above, I say:

3. The only theories that need to match the Bell Inequality are those which meet the Bellian realism conditions.

4. Bellian realism is satisfied by such stable classical objects as dirty socks, down-hill skiers, books in libraries.

5. So could I repeat an earlier question and again request your personal derivation of the EPRB (spin-half) correlations? That is, a derivation that you understand intimately. (Noting that this is NOT a request for you or anyone else to derive the Bellian inequalities.)

Thanks, wm

 

[wm]

In the context of EPRB, please?

The point is: in order for the :"we all agree over the pink elephant" to make any progress in the EPR-Bell dilemma, we need to say that at a certain point, Alice and her friends all agree on a pink elephant, Bob and his friends all agree on a blue elephant, and Alice and her friends all agree on the fact that Bob hasn't seen an elephant. Because if they do, they can assign a probability distribution to a hidden variable ("what Bob and his friends have seen"), and Bell bites us.

Help please! The OP requested the discussion here be in the context of EPRB with spin-half particles.

Is there some reason why we need to consider elephants and wombats? The subject is already beset and riddled with linguistic problems. And, given that, I personally tend to avoid further confusion via extraneous literary elements.

PS: For the record, I understand that wombats in Australia anciently were as big as elephants. Did John Bell know that?

wm

 

[wm]

I love that :!)
It is exactly my critique on Rovelli's RQM.

Can you point me to your critique (or a summary thereof), please?

wm

 

[wm]

I have a question- how can you test for realism? Is it falsifiable? You would say, realism is a false assumption if everybody does not 'see the same elephant', but if 'everybody hears everybody else stating that they see the same elephant they see', then you can't then know if the elephant has an objective reality. It certainly seems that the elephant is an objective thing, say from A's point of view, especially if B is saying it's an elephant, but a completely different consistent set of obsevations is perfectly possible- B seeing a wombat, and hearing A say that it's a wombat.

Locality is certainly falsifiable, and has been tested, so it would surely make more sense to get rid of realism than locality.

I'd like to encourage you in the view that: It surely makes more sense to get rid of pseudo-realism (= limited realism = Bellian realism) than locality.

Thus Bell once strongly endorsed a derivation of his inequalities by d'Espagnat (Sci. Am. November 1979). In it you find this move:

"These conclusions require a subtle but important extension of the meaning assigned to the notation A+. Whereas previously A+ was merely one possible outcome of a measurement made on a particle, it is converted by this argument into an attribute of the particle itself.'' (Emphasis added.)

In my view, most quantum objects are perturbed by ''measurement'' and that is why Bellian Inequalities are breached by quantum objects! Bellian realism being of very limited validity.

PS: As I recall, Bell said he could do no better than d'Espagnat!

wm

 

[DrChinese]

DrC, surely you have mis-spoken? Or are badly mistaken? Even just plain wrong?

You write: ''So, a local realistic theory not only needs to match the Bell Inequality, it also needs to show how it gets there as a function of theta.''

Here is the background to my questioning:

1. QM is correct.

2. Bell-tests are correct.

So, in response to your claim above, I say:

3. The only theories that need to match the Bell Inequality are those which meet the Bellian realism conditions.

4. Bellian realism is satisfied by such stable classical objects as dirty socks, down-hill skiers, books in libraries.

5. So could I repeat an earlier question and again request your personal derivation of the EPRB (spin-half) correlations? That is, a derivation that you understand intimately. (Noting that this is NOT a request for you or anyone else to derive the Bellian inequalities.)

Thanks, wm

I still don't understand what you are asking, especially about the derivation of EPRB. What is the issue about photons vs. electrons? Photons are a lot easier to discuss and that is what all tests are about.

Bell's theorem is NOT satisified by dirty socks and library books (specifically those things do NOT violate Bell's Inequality because they are classical in nature). A Bell local + realisitic theory will never match Bell test experimental results.

You say 1. and 2. above are things you accept, so why are you still talking about local realism?

 

[wm]

Maths is the best logic!

I still don't understand what you are asking, especially about the derivation of EPRB. What is the issue about photons vs. electrons? Photons are a lot easier to discuss and that is what all tests are about.

Bell's theorem is NOT satisified by dirty socks and library books (specifically those things do NOT violate Bell's Inequality because they are classical in nature). A Bell local + realisitic theory will never match Bell test experimental results.

You say 1. and 2. above are things you accept, so why are you still talking about local realism?

1. In that EPRB was the source of Bell (1964), and in that we both accept that the experimental results would agree with QM, if the experiment was done, I am interested in your personal derivation of the EPRB correlation.

2. Can you provide such? Because your words are (to me) so unclear and confusing that I get lost. I am much less likely to get lost when I see you derive the EPRB correlation in mathematical terms that you understand and commit to.

3. See here (again) how my wording is twisted by you: I said that Bell realism is satisfied by dirty socks and library books. THAT IS: They satisfy Bellian Inequalities. Then YOU say: Bell's theorem is not satisfied by dirty socks and library books.

4. Bellian inequalities are satisfied by the Bellian realism of dirty socks (= the realism from which it was derived) and Bellian inequalities are breached by quantum particles because they are not like dirty socks.

5. YOU ASK: Given my acceptance of QM and Bell-test results, why am I still talking about local realism? BECAUSE it is Bellian (limited, constrained) realism that we should reject; it being not valid in general. (See my recent note on this thread about the shifty (''subtle'') move by d'Espagnat regarding A+; a move endorsed by Bell and many others ... and rejected by me. Do you personally accept it for quantum particles?)

6. So let's see your derivation of the EPRB (spin-half) correlations and take it from there. OK?

I might be wrong, and my words no better than yours: BUT Maths is the best logic (so let's see yours)! wm

 

[JesseM]

wm, would you agree that it would be impossible to violate the Bell inequalities classically if one obeyed all the conditions specified in the proof of Bell's theorem (including the condition that the state of the objects/signals emitted by the source be statistically independent of the detector settings), even if one allowed the measurements to modify the state of the objects/signals received by the two experimenters?

 

[DrChinese]

1. In that EPRB was the source of Bell (1964), and in that we both accept that the experimental results would agree with QM, if the experiment was done, I am interested in your personal derivation of the EPRB correlation.

2. Can you provide such? Because your words are (to me) so unclear and confusing that I get lost. I am much less likely to get lost when I see you derive the EPRB correlation in mathematical terms that you understand and commit to.

3. See here (again) how my wording is twisted by you: I said that Bell realism is satisfied by dirty socks and library books. THAT IS: They satisfy Bellian Inequalities. Then YOU say: Bell's theorem is not satisfied by dirty socks and library books.

4. Bellian inequalities are satisfied by the Bellian realism of dirty socks (= the realism from which it was derived) and Bellian inequalities are breached by quantum particles because they are not like dirty socks.

5. YOU ASK: Given my acceptance of QM and Bell-test results, why am I still talking about local realism? BECAUSE it is Bellian (limited, constrained) realism that we should reject; it being not valid in general. (See my recent note on this thread about the shifty (''subtle'') move by d'Espagnat regarding A+; a move endorsed by Bell and many others ... and rejected by me. Do you personally accept it for quantum particles?)

6. So let's see your derivation of the EPRB (spin-half) correlations and take it from there. OK?

I might be wrong, and my words no better than yours: BUT Maths is the best logic (so let's see yours)! wm

What EPR-B correlations are you taking about?

You need to make a specific statement and let's discuss that. I will be glad to discuss any side of Bell's Theorem you want to discuss. I have already referenced several derivations of Bell on my web pages, so I am not sure what you are asking.

To make it clear: I advocate a standard reading of EPR/Bell/Aspect. Bell realism is as limited - or not - as you care to view it. There are those who refer to it as "naive realism" but I personally reject that description (as would Einstein, who was a realist of the same vein).

The math of the Bell realism assumption is simple: assume the simultaneous existence of pre-determined values for 3 non-commuting spin operators (A, B and C). Then prepare a table which shows these 8 permutations when measured as up/down (electrons) and the relative percentages. You will find that it is not possible to create such a table AND have it agree to experiment UNLESS you put negative percentages in some spots.

If you think we should reject "Bellian realism" and accept "locality": I think that is a perfectly sensible interpretation and have no issue with it. But I doubt that most folks will conclude that local realism is still a viable option just because "Bellian realism" is too "limited". If you can come up with an acceptable alternative definition of realism, I would be interested in seeing it.

 

[DrChinese]

In my view, most quantum objects are perturbed by ''measurement'' and that is why Bellian Inequalities are breached by quantum objects! Bellian realism being of very limited validity.

So what if an observation perturbs a system under study? That in no ways explains anything, and it certainly does not explain Bell test results. This is pure hand-waving, and is just as true in the classical world.

 

[wm]

wm, would you agree that it would be impossible to violate the Bell inequalities classically if one obeyed all the conditions specified in the proof of Bell's theorem (including the condition that the state of the objects/signals emitted by the source be statistically independent of the detector settings), even if one allowed the measurements to modify the state of the objects/signals received by the two experimenters?

Hi JesseM,

Would I agree that it is impossible to violate the Bell inequalities classically if one obeyed all the conditions [see* below] specified in the proof of Bell's theorem (including the condition that the state of the objects/signals emitted by the source be independent of the detector settings), even if one allowed the measurements to modify the state of the objects/signals received by the two experimenters?

No; I would not agree.

*But (to be sure we agreeing on the question), I would surely like you to spell out all the conditions, especially any that you see relating to the move in Bell's (1964) maths which no experiment can confirm. I refer to the unnumbered equations between his (14) and (15).

PS: (1) I have not forgotten an old question of yours and have been waiting a reply from a central authoritative source.

(2) Is it not fascinating that Bell should leave unnumbered the most crucial equations in his paper? Did he have doubts? Remember (as I understand the position): He [like me] did not like his theorem!

So let's be sure of the conditions; regards, wm

 

[wm]

Epr-bohm

What EPR-B correlations are you taking about?

You need to make a specific statement and let's discuss that. I will be glad to discuss any side of Bell's Theorem you want to discuss. I have already referenced several derivations of Bell on my web pages, so I am not sure what you are asking.

To make it clear: I advocate a standard reading of EPR/Bell/Aspect. Bell realism is as limited - or not - as you care to view it. There are those who refer to it as "naive realism" but I personally reject that description (as would Einstein, who was a realist of the same vein).

The math of the Bell realism assumption is simple: assume the simultaneous existence of pre-determined values for 3 non-commuting spin operators (A, B and C). Then prepare a table which shows these 8 permutations when measured as up/down (electrons) and the relative percentages. You will find that it is not possible to create such a table AND have it agree to experiment UNLESS you put negative percentages in some spots.

If you think we should reject "Bellian realism" and accept "locality": I think that is a perfectly sensible interpretation and have no issue with it. But I doubt that most folks will conclude that local realism is still a viable option just because "Bellian realism" is too "limited". If you can come up with an acceptable alternative definition of realism, I would be interested in seeing it.

(Emphasis added.)

1. DrC, EPRB, EPR-B stands here for EPR-BOHM. (EPR-Bell is written for EPR-BELL; and I am not seeking a derivation of Bellian Inequalities.) So I am sincerely seeking your derivation of the related correlation:

(1) CORRELATION (EPR-Bohm; spin-half particles) = -a.b'

per terms in OP.

2. Thank you for the realisation that one can drop Bellian-realism and RETAIN LOCALITY. That exactly summarises my position.

3. That alternative definition of realism (called CLR = Common-sense local realism) is on my website (known to you). I'd welcome some critique of it before throwing it in here.

4. So: Could I see your maths for EPR-Bohm, please?

Thanks, wm

 

[wm]

So what if an observation perturbs a system under study? That in no ways explains anything, and it certainly does not explain Bell test results. This is pure hand-waving, and is just as true in the classical world.

Hand-waving?

I had the impression that Bell thought (counter-factually) that an unmeasured system had the property that would have been revealed IF that system had been measured.

That is why he endorsed ''the d'Espagnat move'' mentioned by me here earlier. Thus:

I'd like to encourage you in the view that: It surely makes more sense to get rid of pseudo-realism (= limited realism = Bellian realism) than locality.

Thus Bell once strongly endorsed a derivation of his inequalities by d'Espagnat (Sci. Am. November 1979). In it you find this move:

"These conclusions require a subtle but important extension of the meaning assigned to the notation A+. Whereas previously A+ was merely one possible outcome of a measurement made on a particle, it is converted by this argument into an attribute of the particle itself.'' (Emphasis added.)

In my view, most quantum objects are perturbed by ''measurement'' and that is why Bellian Inequalities are breached by quantum objects! Bellian realism being of very limited validity.

PS: As I recall, Bell said he could do no better than d'Espagnat!

wm

DrC, Are you saying that Bellian Inequalities are based on measurement perturbation?

Regards, wm

 

[JesseM]

Hi JesseM,

Would I agree that it is impossible to violate the Bell inequalities classically if one obeyed all the conditions [see* below] specified in the proof of Bell's theorem (including the condition that the state of the objects/signals emitted by the source be independent of the detector settings), even if one allowed the measurements to modify the state of the objects/signals received by the two experimenters?

No; I would not agree.

*But (to be sure we agreeing on the question), I would surely like you to spell out all the conditions, especially any that you see relating to the move in Bell's (1964) maths which no experiment can confirm. I refer to the unnumbered equations between his (14) and (15).

Well, on the previous thread I already attempted to spell out all the conditions I thought were relevant, in post #133:

do you agree or disagree that if we have two experimenters with a spacelike separation who have a choice of 3 possible measurements which we label A,B,C that can each return two possible answers which we label + and - (note that these could be properties of socks, downhill skiers, whatever you like), then if they always get opposite answers when they make the same measurement on any given trial, and we try to explain this in terms of some event in both their past light cone which predetermined the answer they'd get to each possible measurement with no violations of locality allowed (and also with the assumption that their choice of what to measure is independent of what the predetermined answers are on each trial, so their measurements are not having a backwards-in-time effect on the original predetermining event, as well as the assumption that the experimenters are not splitting into multiple copies as in the many-worlds interpretation), then the following inequalities must hold:

1. Probability(Experimenter #1 measures A and gets +, Experimenter #2 measures B and gets +) plus Probability(Experimenter #1 measures B and gets +, Experimenter #2 measures C and gets +) must be greater than or equal to Probability(Experimenter #1 measures A and gets +, Experimenter #2 measures C and gets +)

2. On the trials where they make different measurements, the probability of getting opposite answers must be greater than or equal to 1/3

I guess I should note that when I say "the assumption that their choice of what to measure is independent of what the predetermined answers are on each trial", this refers to the assumption that the detailed state of the objects/signals sent out by the source, a state which we assume implies a predetermined answer to every measurement (otherwise I don't see any way of explaining how both experimenters always get the same answer when they make the same measurement), is not in any way correlated with or informed by the experimenters' choice of detector settings on that trial.

Also, note that I am not making any assumption that when they make a measurement of a property, they are simply revealing a property which was already present in the state before measurement. I only assume that the state before measurement + the choice of detector setting determines the outcome of the measurement completely. For example, if the experimenter measures a particle on axis A and gets the result "spin-up", this need not imply the particle was somehow in a spin-up state on axis A before it was measured; it just implies that the state of the particle before measurement was such that it was guaranteed that if the detector was on setting A on the measurement, the result would come back "spin-up". Again, without assuming this sort of determinism, there seems to be no way that you could explain how both experimenters always get the same result when they make the same measurement, and still satisfy all the conditions I describe above. Would you agree, at least, with this necessity for determinism in the outcome given both the state of the object/signal emitted by the source on a trial and the choice of detector setting, if the experimenters do indeed get the same result on every trial where they choose the same setting, and the object/signal is a purely classical one, and all my conditions above are being obeyed?

PS: (1) I have not forgotten an old question of yours and have been waiting a reply from a central authoritative source.

Ultimately it is not really important whether any given physicist remembered to include the condition I mentioned in their statements of Bell's theorem or not (although I've shown that several do in their papers); all that's really important is my claim that if you include that condition, along with others I mention, then it is impossible to violate any Bell inequalities classically, but possible to violate them in quantum physics (we are, I hope, debating the physical question of whether quantum results are compatible with local realism, not the historical question of whether Bell or any other particular physicist remembered to state all the relevant conditions in their proofs). If you disagree, then you should be able to come up with a classical experiment where this condition and the other ones I mentioned are all obeyed, yet some Bell inequality is violated; your previous example involving classical polarized light and the source being "yoked" to Alice's detector setting obviously does not obey all my conditions.

 

[DrChinese]

DrC, Are you saying that Bellian Inequalities are based on measurement perturbation?

I am saying that perturbation from measurement has nothing to do with explaining why quantum systems violate Bell Inequalities. It is either because a) realism is a bad assumption; or b) locality is a bad assumption.

Your statement about A+ has nothing to do with this. What Bell thought about his theorem in later years does not prove anything anyway.

 

[DrChinese]

1. DrC, EPRB, EPR-B stands here for EPR-BOHM. (EPR-Bell is written for EPR-BELL; and I am not seeking a derivation of Bellian Inequalities.) So I am sincerely seeking your derivation of the related correlation:

(1) CORRELATION (EPR-Bohm; spin-half particles) = -a.b'

per terms in OP.

2. Thank you for the realisation that one can drop Bellian-realism and RETAIN LOCALITY. That exactly summarises my position.

3. That alternative definition of realism (called CLR = Common-sense local realism) is on my website (known to you). I'd welcome some critique of it before throwing it in here.

4. So: Could I see your maths for EPR-Bohm, please?

Thanks, wm

1. I know what EPR-B stands for. I have no idea of the context.

2. Good.

3. I do not know what common sense realism is. Bell's Realism is pretty common sense to most people.

4. What math are you talking about? Are you talking about the predictions of QM for spin 1/2 particles?

 

[vanesch]

I only assume that the state before measurement + the choice of detector setting determines the outcome of the measurement completely. For example, if the experimenter measures a particle on axis A and gets the result "spin-up", this need not imply the particle was somehow in a spin-up state on axis A before it was measured; it just implies that the state of the particle before measurement was such that it was guaranteed that if the detector was on setting A on the measurement, the result would come back "spin-up". Again, without assuming this sort of determinism, there seems to be no way that you could explain how both experimenters always get the same result when they make the same measurement, and still satisfy all the conditions I describe above.

Indeed. One could illustrate this with the following observation:
Imagine two people, Alice in New York, and Bob in Tokyo, throwing each 1000 times a dice in the following way. They can choose, for each of their 1000 trials, to use a red, a blue or a green box at there disposal ; then they throw the dice in the box of their choice, and write down the outcome and the color of the box they chose.

Note that, if Bob picks the red box for his 52th throw, then he will never know what he would have gotten if instead he'd have picked the green one. And if he next picks the green one, that will not be his 52th, but his 53th throw.

The funny thing now, is when Bob and Alice come together, that they find out that each time that, by coincidence, they picked the same color, well, they also got the same outcome ! Of course, in advance, they cannot know for which throws they will pick the same color, and they cannot determine the outcome. But they simply see that in those particular cases WHEN they pick the same color, then they ALWAYS obtain the same outcomes.

Now, this funny correlation would be totally incomprehensible if there were not some "common origin" or "some action at a distance" between the dice, right ? And if we exclude the last possibility, then we would be looking at some very funny phenomenon. It would be black magic, until we proposed some MECHANISM by which both dice would somehow, in advance, know what to set as a result as a function of the color of the box. One would go and look at the producer of the dice: maybe he put some very complicated mechanism inside each of them.

People who dismiss any "a priori" determinism of the outcomes in an EPR experiment, ought to feel totally comfortable with the above situation, under the motto: "correlations happen".

 

[wm]

1. I know what EPR-B stands for. I have no idea of the context.

2. Good.

3. I do not know what common sense realism is. Bell's Realism is pretty common sense to most people.

4. What math are you talking about? Are you talking about the predictions of QM for spin 1/2 particles?

In reply, by number:

1. EPR-Bohm has two spin-half particles in the singlet state; correlation as previously given here.

2. Good.

3. I told you where to find a definition of CLR (= common-sense local realism) but you do not look? Bell's realism is common-sense? Particles unperturbed ... that A+ ''d'Espagnat move'' again? Are you saying that this Bell-endorsed move is of no consequence?

I say: Drop such ''nonsense'' and such ''fiddles'' and retain LOCALITY?

4. I am talking about you deriving the EPR-Bohm correlation of -a.b' (per terms in OP) so that I can better understand the realism that you hold to; or the locality that you reject; and the terminology that you support mathematically.

If you do the maths, I am presuming we might agree re the terms and come to some agreement about the validity (or otherwise) of LQM (Local QM).

Was it Feynman who said: Do the maths, or risk parrotting the errors of others?

wm

 

[Demystifier]

Wm was right: This really is OK corral.

 

[DrChinese]

3. I told you where to find a definition of CLR (= common-sense local realism) but you do not look? Bell's realism is common-sense? Particles unperturbed ... that A+ ''d'Espagnat move'' again? Are you saying that this Bell-endorsed move is of no consequence?

I say: Drop such ''nonsense'' and such ''fiddles'' and retain LOCALITY?

If you want to push a new definition of realism, bring it out where we can discuss it. But I don't see the purpose of a new definition when the current one is so well accepted. After all, it is exactly what Einstein would have expected.

Bell's realism is common sense, that is why Bell's Theorem is so important. If it did not match up to something most people can understand, it would not be as important.

I don't know why you are making a point about some "move" you are saying Bell endorsed. For all I know, he endorsed Richard Nixon (this is a joke, because he was not an American). The point is that Bell's Theorem stands as written and is generally accepted as such. There has been much debate about whether "hidden variables" exist or not, and if so, whether they are intrinsic particle attributes. If you deny Bell realism (as I am prone to do), none of that matters.

1. As to the math of the realism requirement, the usual presentation is essentially as follows:

1 >= P(A, B, C) >= 0

where A, B and C are 3 simultaneously "real" hidden variables (or attributes, or measurement setting outcomes).

2. As to the correlation in an EPR-B setup with electrons, the usual formula for matches (both up, or both down) is:

p(Match) = sin^2(\Theta/2)

The only significant difference for electrons versus photons being that there is the factor of 1/2 applied to electrons to adjust for being a spin 1/2 particle, while photons have a factor of 1 being a spin 1 particle. Also, entangled photons pairs are usually created by either Type I or Type II PDC. Type II gets a sin^2 function for matches while Type I has the cos^2 function.

 

[complexPHILOSOPHY]

Allow me to interject for a moment to pose a question for philosophical clarity. If the term 'common sense realism' is being used in the context that I suppose it might, are you referencing Thomas Reid?

That is the only 'common sense realism' that I am familiar with.

EDIT: Now that I read through some of the posts, I don't think you are referencing the philosopher. However, when I search for "Common Sense Local Realism" on google, I get returned back to physicsforums.

This leads me to believe that you should just tell me what it is, since google directed me here MY FRIEND! <3333

PAYCEEEE HOMIES.

 

[wm]

Clarifications

Well, on the previous thread I already attempted to spell out all the conditions I thought were relevant, in post #133: I guess I should note that when I say "the assumption that their choice of what to measure is independent of what the predetermined answers are on each trial", this refers to the assumption that the detailed state of the objects/signals sent out by the source, a state which we assume implies a predetermined answer to every measurement (otherwise I don't see any way of explaining how both experimenters always get the same answer when they make the same measurement), is not in any way correlated with or informed by the experimenters' choice of detector settings on that trial.

Also, note that I am not making any assumption that when they make a measurement of a property, they are simply revealing a property which was already present in the state before measurement. I only assume that the state before measurement + the choice of detector setting determines the outcome of the measurement completely. For example, if the experimenter measures a particle on axis A and gets the result "spin-up", this need not imply the particle was somehow in a spin-up state on axis A before it was measured; it just implies that the state of the particle before measurement was such that it was guaranteed that if the detector was on setting A on the measurement, the result would come back "spin-up". Again, without assuming this sort of determinism, there seems to be no way that you could explain how both experimenters always get the same result when they make the same measurement, and still satisfy all the conditions I describe above. Would you agree, at least, with this necessity for determinism in the outcome given both the state of the object/signal emitted by the source on a trial and the choice of detector setting, if the experimenters do indeed get the same result on every trial where they choose the same setting, and the object/signal is a purely classical one, and all my conditions above are being obeyed?

Ultimately it is not really important whether any given physicist remembered to include the condition I mentioned in their statements of Bell's theorem or not (although I've shown that several do in their papers); all that's really important is my claim that if you include that condition, along with others I mention, then it is impossible to violate any Bell inequalities classically, but possible to violate them in quantum physics (we are, I hope, debating the physical question of whether quantum results are compatible with local realism, not the historical question of whether Bell or any other particular physicist remembered to state all the relevant conditions in their proofs).

If you disagree, then you should be able to come up with a classical experiment where this condition and the other ones I mentioned are all obeyed, yet some Bell inequality is violated; your previous example involving classical polarized light and the source being "yoked" to Alice's detector setting obviously does not obey all my conditions.

Dear JesseM,

1. It seems to me that some of your parenthetic comments (''otherwise I don't see any way of explaining how both experimenters always get the same answer when they make the same measurement'') would be helped by your doing the maths in detail so that you understand every step.

2. This is not to say that my maths will always be correct; nor that your words are unintelligible. But more and more I find that those who offer ''almost non-intelligible maths'' (or none at all) are those whose words I struggle most to comprehend.

3. As for determinism: I am most certainly that way inclined! Take any anti-parallel detector settings in EPRB and the detectors punch out identical (++) XOR (--) results till kingdom come.

4. You have completely missed the question that I await an answer on. Let me answer it now, without the external input that I was hoping for: It is my view that Bell (dissatisfied with his own theorem) was open to any hidden-variable theory; local or non-local. However, in my view, non-local hidden-variables are so trivial as to be unworthy of the great man. For (it seems to me) one postulates that a measurement reveals a non-local hidden-variable in one wing of the experiment AND THEN that revelation is non-locally transmitted to the other wing. UGH!

5. Please recall that my earlier CLASSICAL experiment classically refuted Bellian Inequalities, not Bell's theorem. That is, that CLASSICAL experiment complied with the (plus/minus one) conditions used to derive the CHSH etc Inequalities.

6. I'd hoped that DrC would have detailed his maths for the EPRB-correlations. Then my proposed more general CLASSICAL refutation of Bell's Theorem (meeting the more general conditions that you point to, and responding thereto) could have been posted as a counter-point. The hope was that we could see and discuss where DrC needed non-locality and where I thought that I did not!

PS: Is there a simple spot on PF where I can pick up on LaTeX? (My search revealed too much.) Though I'll probably post in a simpler but wholly adequate fashion.

7. So (to be clear): That earlier classical experiment of mine was directed at Bellian Inequalities only. The next is also wholly classical, but seeks to meet the more general Bellian conditions and establish the EPRB correlation -a.b' (per OP) in response to our discussion.

Thanks, and cheers, wm

 

[wm]

Clarification

If you want to push a new definition of realism, bring it out where we can discuss it. But I don't see the purpose of a new definition when the current one is so well accepted. After all, it is exactly what Einstein would have expected.

Bell's realism is common sense, that is why Bell's Theorem is so important. If it did not match up to something most people can understand, it would not be as important.

I don't know why you are making a point about some "move" you are saying Bell endorsed. For all I know, he endorsed Richard Nixon (this is a joke, because he was not an American). The point is that Bell's Theorem stands as written and is generally accepted as such. There has been much debate about whether "hidden variables" exist or not, and if so, whether they are intrinsic particle attributes. If you deny Bell realism (as I am prone to do), none of that matters.

1. As to the math of the realism requirement, the usual presentation is essentially as follows:

1 >= P(A, B, C) >= 0

where A, B and C are 3 simultaneously "real" hidden variables (or attributes, or measurement setting outcomes).

2. As to the correlation in an EPR-B setup with electrons, the usual formula for matches (both up, or both down) is:

p(Match) = sin^2(\Theta/2)

The only significant difference for electrons versus photons being that there is the factor of 1/2 applied to electrons to adjust for being a spin 1/2 particle, while photons have a factor of 1 being a spin 1 particle. Also, entangled photons pairs are usually created by either Type I or Type II PDC. Type II gets a sin^2 function for matches while Type I has the cos^2 function.

1. I'll advance my definition of CLR (common-sense local realism) in the hope that you (and Einstein) will find it agreeable.

2. I'm heartened in this regard, noting with some cameraderie that you too are ''prone to deny Bell realism''. Let me add (from my perspective) you are moving in the direction of a growing band.

3. Thus: In my opinion: If you study the Bell-endorsed d'Espagnat move (re the subtle change re A+ -- see earlier posts), then you'll see more clearly what Bellian realism entails. I think it involves a clear denial of measurement perturbation; and I'm not sure that I know where Bell moved away from that position.

4. I'm not too happy with your maths here; though I too am not a mathematician. I'll post my maths ideas (based on my view of common-sense) to see if I can move you further in that direction.

wm

 

[DrChinese]

Dear JesseM,

1. ... would be helped by your doing the maths in detail so that you understand every step.

6. I'd hoped that DrC would have detailed his maths for the EPRB-correlations...

We already understand the math of Bell. If you have something to say, say it. It is getting old wondering when you are going to drill into a specific point. I have presented standard treatments of Bell, and so far there is no point of disagreement other than you don't seem to like Bell's realism assumption. If you want to replace his with your own, you will need to make a convincing argument of the benefit of doing so because otherwise there will be no interest in pursuing the matter.

 

[DrChinese]

1. I'll advance my definition of CLR (common-sense local realism) in the hope that you (and Einstein) will find it agreeable.

2. I'm heartened in this regard, noting with some cameraderie that you too are ''prone to deny Bell realism''. Let me add (from my perspective) you are moving in the direction of a growing band.

3. Thus: In my opinion: If you study the Bell-endorsed d'Espagnat move (re the subtle change re A+ -- see earlier posts), then you'll see more clearly what Bellian realism entails. I think it involves a clear denial of measurement perturbation; and I'm not sure that I know where Bell moved away from that position.

4. I'm not too happy with your maths here; though I too am not a mathematician. I'll post my maths ideas (based on my view of common-sense) to see if I can move you further in that direction.

wm

1. Einstein said of realism: "I think that a particle must have a separate reality independent of the measurements. That is: an electron has spin, location and so forth even when it is not being measured." That is Bell realism to a tee, and I don't see how you are going to do Bell & Einstein one better. I can tell you for sure that I am not interested in a semantics debate on realism. There needs to be a connection to the physics.

2. I haven't moved anywhere on this subject.

3. This has nothing to do with anything. I assume there is measurement perturbation, the question is what is the significance of it? Is it physical? Does it have non-local impact? etc.

4. Sorry professor! You are always welcome to add your improvements.

-DrC

 

[DrChinese]

3. Re: ''no point of disagreement other than you don't seem to like Bell's realism assumption''. I just wanted to check that you ALSO are moderately sympathetic to this point of disagreement? Yes?

No, not in the least. Bell's realism is well-defined, and it is reasonable to reject it as an assumption (while keeping locality as an assumption). That does NOT mean it is a poor definition - it is a good one and that is a big part of why Bell's Theorem is so strong!

A lot of people reject realism as an assumption, and I am hardly the first. Don't confuse the "assumption" with the "definition of the assumption" - they are entirely different.

 

[vanesch]

That is the only 'common sense realism' that I am familiar with.

I don't think it is a technical term. Look at my post #42. The "common sense realism" is the idea that these correlations between dice throws of Alice and Bob "just cannot be" if there is no common causal origin.
It comes down to the idea that, when correlations occur where they are not a priori expected for some logical reason, there must be some causal link (by common origin, or by influence), and that this causal link must be found in some ontologically existing mechanism.
The reason to adhere to it, is that it is our main (and only) technique of inference and empirical enquiry.

Indeed, if you throw a switch, and a light goes on, and you throw it again, and the light goes out, and you do this 20 times, you expect there to be some kind of ontological mechanism to exist which explains this. It doesn't necessarily mean that the switch causes the light to go on. A technician might observe you through a camera, and steer the light as you throw a non-connected switch. Or worse, he might switch on and off a light, and "send you some brain waves" which make you throw the switch. Or even better, when you walked into the room, a scanner might have found out the exact state of your brain, and calculated at what moments you will throw the switch. One extravagant explanation even worse than the other. But you prefer that, over: well, correlations happen. There's no relationship. This desire of explaining correlations is, I think, what is meant by "common sense realism": there must be something real, which is the mechanism which explains the correlations.