Absolute versus Relative Angles in the Bell-CHSH Scandal

[Joy Christian]

I want to bring out something that usually gets obscured in the discussions about the so-called Bell's theorem (we shall not be concerned about trivial variants of such "theorems" by lesser brains---the proponents of such trivialities are advised to promote their nonsense elsewhere).

The dying breed of the Bell mafia (when they are not engaged in spreading despicable violence and terror) routinely demand things from their chastisers that are not predicted by quantum mechanics, since they have been systematically deprived of valid predictions of quantum mechanics they can pretend to play their game with.

Take, for example, the familiar predictions of the four pairs of Bell-test angles in an EPRB-type experiment:









On the face of it there seem to be nothing wrong with these predictions. The expectation value of the products of measurement results when Alice has her polarizer oriented at 0 degrees and Bob has his polarizer oriented at 45 degreees is -0.7071..., and the expectation value of the products of measurement results when Alice has her polarizer oriented at 0 degrees and Bob has his polarizer oriented at 135 degreees is +0.7071..., and so on. But look closely and you will find that this is not really what quantum mechanics actually predicts. If we call Alice's polarization angle and Bob's polarization angle , then quantum mechanics actually predicts



where is the singlet state. In other words, quantum mechanics neither knows nor predicts anything about the absolute polarization angles and . To be sure experimenters are free to choose them as they will, but quantum mechanics couldn't care less about them. It only cares (i.e., makes predictions) about quantities dependent on the relative angle between the polarization directions. Thus the above stated four quantitative predictions are actually an overstatement of the facts. The valid predictions of quantum mechanics are only the following two predictions, reducing the above four to two:





This effectively means that, without violating any predictions of quantum mechanics, we can always set one of the polarization angles to zero and measure the variations in the correlation with respect to that setting. This does not compromise the free will of the experimenter. It only sets a convention for a coordinate system. Thus, for definiteness, we can either set or set . In the following simulation I show the correlations these two possible conventions lead to:

http://rpubs.com/jjc/16415.

Let me stress again: To demand control over absolute values of and in this simulation is to demand something beyond the confines of quantum mechanics, even if experimenters may have such a control. Thus when the Bell mafia make such demands they are as usual being themselves---i.e., being sore losers and dishonest.

Moreover, it is important to note that in the above simulation there is no filter of any kind---i.e., no post-selection of any kind. The initial states are defined by a set of vectors , each one of which is accounted for in the calculation of the correlations.

They say there is no point in flogging a dead horse, but Bell's theorem seems to be an exception. It needs to be obliterated from the consciousness of the community.

[FrediFizzx]

Yep, and with the advent of these two recent classical experiments that show strong correlations, it shouldn't be long now hopefully that Bell's theorem is relegated to the dustbin of other failed no-go theorems.

http://arxiv.org/abs/1406.3338
http://arxiv.org/abs/1406.6239

As I have said many times, experimenters are very clever and it shouldn't be long before a Weihs, et al, type experiment is checked in the classical domain. Of course that would be a stepping stone to the ultimate test with a mechanical experiment.

[Joy Christian]

It is worth adding here that even the demand of a control over the absolute values of angles have been met in this simulation, despite the fact that such a demand cannot be justified on the quantum mechanical grounds.

Thus, even the pseudo-demands of the Bell mafia have been successfully met by the correlation surface generated in the simulation:

[gill1109]

Yep, and with the advent of these two recent classical experiments that show strong correlations, it shouldn't be long now hopefully that Bell's theorem is relegated to the dustbin of other failed no-go theorems.

http://arxiv.org/abs/1406.3338
http://arxiv.org/abs/1406.6239

As I have said many times, experimenters are very clever and it shouldn't be long before a Weihs, et al, type experiment is checked in the classical domain. Of course that would be a stepping stone to the ultimate test with a mechanical experiment.

The Weihs et al experiment suffers severely from the well-known loopholes, in particular, the detection loophole.

After 50 years of trying hard, the experimentalists still haven't succeeded in doing the good experiment in the quantum domain. No-one has been trying to do this experiment in the classical domain, except for the army of believers in local hidden variables models who have tried to implement their models in computer simulations. So far they were only succesful by exploiting the known loopholes, ie by performing the wrong experiment.

[FrediFizzx]

The Weihs et al experiment suffers severely from the well-known loopholes, in particular, the detection loophole.

After 50 years of trying hard, the experimentalists still haven't succeeded in doing the good experiment in the quantum domain. No-one has been trying to do this experiment in the classical domain, except for the army of believers in local hidden variables models who have tried to implement their models in computer simulations. So far they were only succesful by exploiting the known loopholes, ie by performing the wrong experiment.

And once again... Gill is on the total wrong track.

[gill1109]

The Weihs et al experiment suffers severely from the well-known loopholes, in particular, the detection loophole.

After 50 years of trying hard, the experimentalists still haven't succeeded in doing the good experiment in the quantum domain. No-one has been trying to do this experiment in the classical domain, except for the army of believers in local hidden variables models who have tried to implement their models in computer simulations. So far they were only successful by exploiting the known loopholes, ie by performing the wrong experiment.

And once again... Gill is on the total wrong track.

Well ... one of us is on the total wrong track.

But it is easy for you, Fred, to prove that I am wrong: just write a computer simulation of one of those experiments which you claim show classical violation of Bell. Make sure it satisfies the time and space and freedom (of choice of settings) constraints of a proper Bell-CHSH type experiment -- not the experiment Weihs et al actually did, but the kind of experiment which they always wanted to do! The kind of experiment described very carefully by Bell (1981). The kind of experiment which the top experimenters think might well get done within one year from now (we know that the top experimental groups are racing all out to be the first).

If you succeed then you not only have proven that I am on the wrong track, but you have also shown the intellectual bankruptcy and moral corruption of the whole Bell / entanglement / quantum information maffia. You'll even get the Nobel prize since a computer network running your programs is a classical physical system generating those "classically impossible" strong "quantum" correlations without cheating in any way. You post the programs to internet and the whole world will know about it in no time. No way the Establishment can suppress the news. You will be a great Hero. I will eat my hat (my professorial beret,to be precise), I will publicly apologise to you and to a whole lot of other people, and I will withdraw about 15 of the papers I am most proud of and worked most hard on. I'll tweet it on Twitter and post it on Facebook and Google plus. I know a whole lot of good science journalists all over the world. This news will spread like wildfire ...

[Joy Christian]

Off topic. Stay out of phoney and macho rhetoric and stay on the topic of this thread: Absolute versus Relative Angles in the Bell-CHSH Scandal.

[gill1109]

Off topic. Stay out of phoney and macho rhetoric and stay on the topic of this thread: Absolute versus Relative Angles in the Bell-CHSH Scandal.

Talking about macho rhetoric and off-topic postings, what about

Experimenters are very clever and it shouldn't be long before a Weihs, et al, type experiment is checked in the classical domain

A Weihs et al experiment uses absolute angles, selected at random by both Alice and Bob, independently, on both sides of the experiment. And experimenters are still far from having done this experiment properly in the quantum domain. Strange that no-one even tried to do it in the classical domain, in all these 50 years?

Your own simulation (sorry, you yourself went off-topic when bringing it up) - Pearle's detection loophole model from 1970 - does the same: it uses absolute angles. The computer code removes a subset of the runs from the experiment, depending both on the current settings and on the originally generated hidden variable. The probability distribution of the hidden variables within the subset of "accepted" states depends on the two settings in force on each side of the experiment.

This could be arranged physically by using the conspiracy loophole: the experimenter has no free choice in picking those settings: they were determined long in advance hence known in advance at the source. No violation of locality.

But if you don't like conspiracy, you can do it with non-locality: the two settings are instantaneously known at the source.

One can just as well think of the selection as being pre-selection as being post-selection. Pre-selection requires conspiracy and/or non-locality. Post-selection is simply against the rules of the experiment - ie it is cheating. The computer simulation is just a mathematical model. The mathematical model of Pearle (1970). It can be interpreted physically in many different ways.

There is no way to violate a true theorem. You can only try to circumvent it. One can say that Christian's simulations yet again beautifully confirm Bell's theorem.

[gill1109]

Back on topic: if you are only interested in relative angles you might as well keep Alice's angle fixed and only vary Bob's. This means that Alice's angle can be taken as known, only Bob's varies, and on Bob's side of the experiment, both angles are known. Not difficult to create "quantum correlations" by classical means.

Remember EPR? In each wing of the experiment, there are two different measurements which can be made. Position or momentum. No EPR paradox when the way one of the particles is measured, is fixed. Seems we have to blame Einstein for this obsession of the Bell mafia with absolute angles, varying in both wings of the experiment.

[Joy Christian]

Back on topic: if you are only interested in relative angles you might as well keep Alice's angle fixed and only vary Bob's. This means that Alice's angle can be taken as known, only Bob's varies, and on Bob's side of the experiment, both angles are known. Not difficult to create "quantum correlations" by classical means.

Remember EPR? In each wing of the experiment, there are two different measurements which can be made. Position or momentum. No EPR paradox when the way one of the particles is measured, is fixed. Seems we have to blame Einstein for this obsession of the Bell mafia with absolute angles, varying in both wings of the experiment.

Perhaps you are suffering from reading difficulty. Let me repeat at least a part of what I said above. Perhaps the point will become clearer second time around:

It is worth adding here that even the demand of a control over the absolute values of angles have been met in this simulation, despite the fact that such a demand cannot be justified on the quantum mechanical grounds.

Thus, even the pseudo-demands of the Bell mafia have been successfully met by the correlation surface generated in the simulation:

[Ben6993]

You have replied to a post by Richard Gill who says that he has since been barred from this site. I am mentioning this for anyone who might expect a reply to you from Richard. He cannot reply here.

[Joy Christian]

You have replied to a post by Richard Gill who says that he has since been barred from this site. I am mentioning this for anyone who might expect a reply to you from Richard. He cannot reply here.

Thank you, Ben. Why was he barred from this site, if I may ask?

[Ben6993]

Hi Joy,
You would need to find out from Fred.
(I have a new computer and am struggling to type with it!)
Ben

[FrediFizzx]

Guys, this is off topic; let's stay on topic. Please discuss this via PM or email.

[Joy Christian]

Over on the FQXi community blog Jonathan Dickau made some nice comments about my work which may be of interest here:

Well said Tom,

Your post ... sums up the salient features of Joy Christian's model nicely. I think that the existence of an extra-dimensional framework is essential to understanding the quantum correlations problem, and that Joy goes to the core of the issue. Peter's approach seems mainly to deal with the phenomenology, which may be more like the conventional approach to Physics in some ways, but JC points out that the assumption of an underlying reality with a topological twist provides an elegant answer why some of those intriguing phenomena occur in the first place.

The thing is; Joy is simply citing facts of geometry and topology that are indisputable realities, and asserting that they explain the Physics we observe. However; showing that the rudiments of geometry dictate the properties of space is difficult, in a world where the majority of physicists feel that the Physics is determining the properties of space, rather than the other way around.

Regards,

Jonathan