Remaining open problems with Bryan Sanctuary’s approach

[Richard Gill]

Nobody responded to this remark by me: I mentioned that in Bryan’s video he seemed to suggest that whether a particle pair ended up having their polarizations or their helicities measured depended on the measurement setting(s). It seems to me that this would have to be a nonlocal process. However, the papers and the simulation don’t have any information about this at all. The simulation does not simulate a single Bell type experiment. It simulates two, one for the helucities, one for the polarizations, then adds the correlations. (Everyone but Bryan thinks this is nonsensical.)

Maybe I’m the only one who watched the video?

Overall, the present situation is that nobody in the group is actively supporting Bryan. If I remember correctly, Mark Hadley asked him if he knew of any scientists who agreed with his analysis and/or were building on it. We never got an answer.

PS Austin Fearnley asked for a new thread on Bryan Sanctuary’s model. So this could be it!

Actually Austin’s problem would disappear if, when replying to an email from the group, people would not quote the previous email, with a quote of the previous at the end,, with a quote of the previous …)

[Jan-Åke Larsson]

I certainly did not watch the video.

Bryan's claimed violation is just a calculation error that would cause him to fail the exam in Probability 101, badly.

His whole "theory" is built on that error. No need to spend time on that.

/Jan-Åke

--
You received this message because you are subscribed to the Google Groups "Bell inequalities and quantum foundations" group.
To unsubscribe from this group and stop receiving emails from it, send an email to Bell_quantum_found...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/f1dc13f1-59b5-4678-ab96-717fd1c39eefn%40googlegroups.com.

--
Jan-Åke Larsson
Professor, Head of Department

Department of Electrical Engineering SE-581 83 Linköping Phone: +46 (0)13-28 14 68 Mobile: +46 (0)13-28 14 68 Visiting address: Campus Valla, House B, Entr 27, 3A:512 Please visit us at www.liu.se

[Richard Gill]

PS. I believe that any group member can start a new topic; just go to our Google groups webpage https://groups.google.com/g/Bell_quantum_foundations  , will probably work best on Chrome and if you use a gmail account. You don’t have to ask the managers (me and Alexandre) or the owner (Alexandre). If you think you are getting too many emails, just ask to only get a weekly digest. There are many more features. Take a look! Take control!

[Richard Gill]

I agree with your remark about Probability 101. I don’t think there is any scientific need to spend any time at all on Bryan’s model. It is based on several serious misconceptions and on elementary logical errors. However I have a scientific interest in finding out why very intelligent people are so good at fooling themselves and why they can dedicate their lives to a science fantasy.

[Mark Hadley]

I have not watched the video. Bryan has given no explanation of how he can avoid the conclusion of Bell. Claims that Bell is classical or cannot account for complementarity are just nonsense. There are no such assumptions in the derivation of CHSH.

When presented with claims to explain QM I don't usually even read the papers unless they give a succinct credible explanation of how it relates to BI.

Bryan, just for the record the E is the expectation value of the correlation. It is defined as (N_eq - N_ne)/N_tot. That is a definition. The experimenter counts the ticks in each box and does that simple sum. The fact that you do soe manipulations are get a different equation is proof that you have a calculation error. YOu are luck that people have studied your work well enough to find the error for you.

CHeers

Mark

 

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/b7ee2316-1c7e-36fc-71a2-28ed60599833%40liu.se.

[Mark Hadley]

It is very common for scientists to hold onto their pet theories despite mounting contrary evidence. I'm doing that myself to some extent.

They deny the facts or modify their theories. 

It's been said that scientific revolutions don't happen because people change their minds. But rather that old ideas die with their supporters. 

Having an alternative idea to work on is probably the best exit route. 

Cheers

Mark 

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/754a431f-94a6-402e-b38b-f97aac921c08n%40googlegroups.com.

[Bryan Sanctuary]

Hi Richard

I should point out that the latest discussion is not used in my papers.   It was a recent idea to see what people thought and if it will work,  I agree that it does not work, so I will put that idea aside and go back to what I did in the paper, and just add the two correlations..

Although I very much appreciate the comments and discussion we just had, please note that Richard's title, "remaining problems with my approach..." does not recognize that my theory works, is consistent from QFT up, answers the questions,  and resolves old problems.  Except for this last point we just discussed, no objective problems have been raised which show my approach is wrong, in fact the opposite.  I get a mechanism for the correlation which makes a lot of sense.   Rather than the disingenuous comment about "remaining problems.." I suggest there is only one problem and you guys should help by trying to answer that, rather than throw up your hands and kill the whole theory because we are not yet clear on how the data can be rationalized.  

Although some say the two correlations should not be added, it appears to me that that is exactly what Nature is telling us.  Again I bring it back to complementarity.  The question, how much correlation is there in a complementary system might be answered like this.  We have two independent properties.  At the max settings in CHSH (45 degrees) one attribute gives a correlation of CHSH = 2 and the other gives a correlation of CHSH = 1, and so the maximum correlation in the system is the sum of the two.  I note that correlation is added in the CHSH formula from four experiments, so what prohibits adding correlation from two complementarity attributes?  I also note the BI are not violated in my work since neither pol nor coh exceed CHSH = 2.  Bell's work does not consider complementarity.  He says the max correlation from pol equals two and the extra is due to non-locality. I find it comes from coherence which exactly accounts for the violation. 

I would be interested in your views of this. 

Bryan

--

[Bryan Sanctuary]

Hi Mark

Well I do not hold onto pet theories.  I am flexible and let the ideas evolve.  I change as logic dictates. 

Bryan

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/CAN%3D2%2Bo0UJqSQT-77mF9qbxnGztUX6w_ODJbz15OkAwUPe%2BJ1Jg%40mail.gmail.com.

[Richard Gill]

About what I was saying about whether helicity or polarization is observed:

Look at Bryan's video:

https://www.youtube.com/watch?v=nQdtBRxL8Mk

Notice the links in the ”info” at a couple of places:

29:44 SPIN DECOUPLING

38:16 The simulation. Different filter positions favour polarization or coherence

Here Bryan is saying things which he didn't write in his papers. If he would implement these ideas for a *pair* of particles at two separated detectors he will run into a problem. He wants either both particles' helicities to be measured, or both particles' polarizations to be measured. But he also wants this to depend on the two relative filter positions which the particles encounter.

Bryan's ideas and Bryan's maths are not in sync. He has graphic ideas about what is going on, expresses them in (for physicists) emotive words and phrases, but can't connect them faithfully to maths.

I'm not a physicist. I don't take much notice of the imagery, I can't use the associations with fundamental physics concepts and relations between them. I have to look carefully at the mathematical definitions and look carefully at the mathematical relations and derivations. It is totally irrelevant what "complementarity" ought to mean, or what all kinds of famous (and now mostly, dead) guys said about it. Successful physics gets mathematized and axiomatized. That is not going to happen with Bryan's theory.

[Richard Gill]

Correlation is not *added* in Bell-CHSH. Three correlations are added and one is subtracted. The 8 one-sided Bell-CHSH inequalities delineate the sets of four correlations which are allowed by theories which can be expressed as local hidden variable theories. Sets of correlations which violate any of these inequalities cannot be generated by a local hidden variables theory. Sets of four correlations which satisfy all of them (together with the no signalling equalities on the marginal distributions of outcome given both settings) can be generated by. LHV. Quantum mechanics allows, theoretically,  sets of correlations which cannot be produced by ab LHV theory. These results have been known for getting on for 60 years, but Bryan is blissfully unaware of them. This is a bit like someone starting completely afresh, inventing a new theory of planar geometry, ans who has only the vaguest of ideas what Pythagoras theorem is about and confident that it will not be true in his theory. It's a bit ... foolish.

[Jan-Åke Larsson]

For clarity: You do emphatically *not* generate a new correlation from any of the expressions used in Bell inequalities. You add particular correlations to form Bell expressions that are bounded by Bell inequalities.

Also, Bell's inequality is the triangle inequality of Local Realist models.

/JÅ

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/ef983dce-3d0e-480e-a862-4d5b11002d8cn%40googlegroups.com.

[Austin Fearnley]

Hi Bryan

As I noted previously, one does not need to average correlations, instead accumulate the paired counts and find the overall correlation in one step.  If the model cannot give S = 2.8 overall than it has fallen behind real experiments which far exceed S= 2 overall.  That is irrespective of separate correlation coefficients for coh and pol.

Hi Richard

I have looked again at the video where you have indicated.  I learned a touch of GA to cope with Joy's model but dropped out when he moved to quaternions.  That means I am not trying to cope with Bryan's quaternions.  It seems to me though that Bryan is merely pointing out under what condition (i.e., differences in a-b vectors) one gets clicks dependent on coh or pol or some combination.  And the fact that the simulation uses both coh and pol correlations means that Bryan is not manipulating a and b vector settings to enhance his results via dependence of particle properties on detector settings?  

Hi Mark

You have given me an interesting target to link e and h.  Thanks.
My starting point is that according to my preon model and to my string theory learnings from Susskind, electric charge lives in its own 4D space (actually 12D).  This is the Kaluza-Klein fifth dimension.  We can only access that independent 4D space by clicks (rather like S-G measurements are clicks).  That explains a lack of precision so we cannot delve within a click.  This is also because the contents of the KK dimension are moving at relativistic speeds compared to our 4D.

Also, Ahranov has papers where he says that his TSVF model completely recovers QM.  And (if I remember accurately) he has a paper using TSVF to model two-slit interference patterns using particle at a time.

You suggest an alternative idea for Bryan's to target.  I suggest that the model be  used to simulate Malus's Law and S-G measurements.
   
Austin

[Richard Gill]

Dear Bryan

You said you would be interested in my views on the idea of adding correlations from experiments in which complementary variables are measured. 

My view is that I see no point to it at all. No experimenter ever did it. The experimenters who observed statistically significant violations of Bell inequalities did not do it. So your idea doesn’t solve any existing problem. Does not predict anything exciting. Does not change our understanding of physics. Does not contradict any existing theory.

Why don’t you take a break and contribute to some other threads? 

e.g. Let me hear *your* views on the Marian Kupczynski - Richard Gill arguments. My paper with Justo Pastor Lambare https://arxiv.org/abs/2209.00702  is held up for an editorial investigation by Quantum Reports (MDPI). 

Or on Inge Helland’s ideas.

Or - in a new thread - on my new paper on optimal statistical analysis of Bell experimental data. It’s on arXiv, https://arxiv.org/abs/2209.00702 , I’m currently correcting the proofs for its publication in an MDPI journal called Applied Mathematics.

Richard

Sent from my iPad

[Austin Fearnley]

Hi Richard

The link to your paper with Justo is not correct as it duplicates the other link.

I briefly looked at the other link and found a typo on page10, section 6.3, line 3, where 'my' should be 'by'.  Sorry, not much help!

I am glad you commented on meaningful significance, which is preferable to mere p value significance.

I did not read closely as that would take me a very long time, and so am very unclear how in principle your work identified anything about time drift?

Best wishes

Austin

[Richard Gill]

Dear Austin

Here is the missing link - my. long paper with Justo on one of Marian’s works:

https://arxiv.org/abs/2208.09930

History: we wrote this first and submitted it to Frontiers. It was refereed and accepted but Frontiers is much too expensive for us. So we wrote a very short version for Frontiers, which we could just afford.

But: I had been given an invitation for an APC free article in MDPIs Quantum Reports, so we sent the long version there. It got three referee reports, two were very positive, one was very negative. It was accepted and we submitted a revision. But the revision has got held up because the editors received a complaint, presumably from the reviewer who submitted the “minority report”. C’est la vie. 

The time drift issue is subtle. If both the parameters of the setting choice randomisers, and of the whole internal stuff of the experiment (source, transmission times, and detectors), drift over time, then one can see violation of no-signaling. That means that deviations from the no-signalling equalities are large. Now these deviations are also correlated, if the theoretical perfect symmetry of the four 2x2 tables is disturbed, with the sampling error in S.  Reducing the error by least squares makes it much smaller and makes its variance much smaller and hence makes the statistical test of S .le. 2 more powerful. Hence, in experiments which are blemished by these drifts, the improvement which my methodology gives can be large. If on the other hand an experiment has no appreciable drifts and moreover if the four tetranomial distributions exhibit the ideally expected symmetries, and the four sample sizes are all equally large, then the least squares procedure does not change the p-value at all. Moreover, it turns out to be the same as the p-value of the martingale test based on the Bell game. 

Ie when the experiment is perfect, the Bell game martingale p-value and the p-value based on the standard error of CHSH computed in the usual elementar way are more or less the same. 

The martingale test protects you from drifts. It needs less assumptions. But it does not lose statistical power!

[Inge Svein Helland]

Dear all,

On the Bell experiment, I can only recommend Richard's latest papers, my own commentary just posted, and my recent contribution arXiv:2305.05299 [quant-ph]. For those that really want to go into the mathematics of my views, there is a longer paper on arXiv: 2305.06727 [quant.ph].

Inge

---

From: bell_quantum...@googlegroups.com <bell_quantum...@googlegroups.com> on behalf of Richard Gill <gill...@gmail.com>
Sent: 12 May 2023 22:13:42
To: Austin Fearnley
Cc: Bell Inequalities and quantum foundations
Subject: Re: [Bell_quantum_foundations] Remaining open problems with Bryan Sanctuary’s approach

 

--

You received this message because you are subscribed to the Google Groups "Bell inequalities and quantum foundations" group.
To unsubscribe from this group and stop receiving emails from it, send an email to Bell_quantum_found...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/A015DDBB-6755-4FBF-8AE0-C4BBF5F7CDBE%40gmail.com.

[Richard Gill]

Dear Inge

I would be delighted to report my own opinions on pros and cons of your innovative and interesting approach. How about you start a new topic dedicated to it? Start off with the key references and an "executive summary". And a nice topic title.

Richard

[Bryan Sanctuary]

On the LHS we have, say, N_eq with a max of the total number of EPR pairs (one coincidence per pair).  When these are separated into N^pol_eq + N^coh_eq they add to 2N_eq, twice as many particles as you really have.  So Jan-Ank needs to learn to count rather than dismiss my work by stating I do not understand simple probabilities.  You can use crayons to prove this.  

So with that factor of one half, the Pol and Coh correlations are added, not averaged.  This is consistent with complementarity, something perhaps along the lines of change of ensemble.

Bryan

Bryan

Bryan

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/b7ee2316-1c7e-36fc-71a2-28ed60599833%40liu.se.

[Jan-Åke Larsson]

No they don't.

According to your own claims, the data you count is just one data point per pair, not two.

EITHER polarization OR coherence, you said so yourself.

Then N^pol_eq+N^coh_eq add to N_eq. Not 2 N_eq.

One count per pair is also what experimenters have. They never count two pairs in a single timeslot.
These counts are then used to calculate a point estimate of the correlation.

/Jan-Åke

[Mark Hadley]

N is the number of measurements not the number of particles.

The correlation value is the difference of eq and ne  divided by the sum

Jan makes perfect sense. And agrees with Richard and myself. 

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/CALLw9YwO0gZxqRN8cLJWU9aaEHZ2VYW_fc4XfpuArx5y3ug65g%40mail.gmail.com.

[Jan-Åke Larsson]

Bryan is attempting to count each pair twice, once as coherent and once as polarization.

Experimenters count each pair once.

[Richard Gill]

Dear Bryan, are you never going to learn how to spell Jan-Åke’s first name correctly?

If you want to be taken seriously, you have to learn how to spell as well as how to count

[Are you by any chance dyslectic?]

Richard

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/CALLw9YwO0gZxqRN8cLJWU9aaEHZ2VYW_fc4XfpuArx5y3ug65g%40mail.gmail.com.

[Bryan Sanctuary]

Dear Jan-Åke

Clearly you do not follow my argument. You simultaneously count a click from Pol and a click from Coh so the sum is 2N_eq unless you put in some restrictions which you don't.. 

Bryan

[Jan-Åke Larsson]

No I am following the argument. It is just that it is self-contradicting.

With your logic 1=2.

[Richard Gill]

Dear Bryan, your argument cannot be followed because it depends on two mutually contradictory assumptions.

Seems you must have had a lot of practice in believing impossible things. Remember Alice in Wonderland? "I daresay you haven't had much practice,' said the Queen. 'When I was your age, I always did it for half-an-hour a day. Why, sometimes I've believed as many as six impossible things before breakfast. There goes the shawl again!”

Richard

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/CALLw9YwMmn8yrmW4VvmAK-7swgFpcCH9mz0U05vAP3yMXgE77w%40mail.gmail.com.

[Chantal Roth]

How about a trivial example?

Bryan, what numbers would you put in here?

(Left ist before the experiment. Right after measurement)

We have 2 sets of entangled photons (pair P1 and pair P2)

The first pair registers a Pol click.

The second pair registers a Coh click.)

Best wishes,

Chantal

[Bryan Sanctuary]

no I say that 1 = 1/2 x2

[Chantal Roth]

Bryan,

What about the counts in this example (maybe we can all agree and then move on :-))? 

There are 2 pairs of photons (total 4).

The first pair registers as pol.

The second pair registers as coh.

What are your click counts?

What is your total in this case?

Best wishes,

Chantal

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/CALLw9YztnNNHQh1PjNtzpemN_Sgh0PZ6xYaObd2Oav6MuJWoAw%40mail.gmail.com.

[Richard Gill]

Chantal, are the settings held constant in your example? And what are the possible outcomes on each side?

Are you taking them to be elements of the Cartesian product  {-1, 0, +1} x {pol, coh} ?

Maybe Bryan can give us a list of all possible outcome pairs which he envisages for Alice and Bob

There are clicks +/-1 for each party and they may be “pol” or “coh” clicks

Alice and Bob either both have a pol click or both have a coh click

Since Bryan never gives a clean mathematical description of his model nobody can ever say that his analysis is correct or incorrect because it is simply incomplete

I believe that it is wrong because it cannot be completed in a coherent way.

On 24 May 2023, at 06:53, Chantal Roth <cr...@nobilitas.com> wrote:

Bryan,

What about the counts in this example (maybe we can all agree and then move on :-))? 

There are 2 pairs of photons (total 4).

The first pair registers as pol.

The second pair registers as coh.

What are your click counts?

What is your total in this case?

<image.png>

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/9cbae5f8-2b07-4c5f-a4c7-e74c88fac753%40app.fastmail.com.

[Chantal Roth]

Richard,

This is the simplest possible (I think) toy example :-). Can't get any easier :-).

So yes, fixed settings.

Yes, -1, 1 or 0 (no click).

And yes, both have pol measurement, or both have a col measurement (XOR)

[Richard Gill]

OK, thanks Chantal! So we do have the same guess as to what Bryan is thinking the data would look like, if one could experimentally distinguish “pol” from “coh”

It seems that he has the weird idea that as it is not yet possible to do that, we should double all correlations. I would say that by his own logic he should halve them, and then give those numbers as the correlation coming from each of his imagined sources.

I don’t see this idea catching on

[Pierre Leroy]

Hi,

By noting:
p => event of type 'pol'
c => event of type 'coh'

With one emitted pair, the possibles combinations of states are as
follows: (some may be impossible, it's to Bryan to say it)

+p +p
+p -p
+p +c
+p -c

-p +p
-p -p
-p +c
-p -c

+c +p
+c -p
+c +c
+c -c

-c +p
-c -p
-c +c
-c -c

The types of possible experimental measurements are:

+ +
+ -
+ 0
- +
- -
- 0
0 +
0 -
0 0

(noting 0 => no detection)

It is enough that Bryan define for each type of possible experimental
measurement, the states of the first list which can produce the measurement.

example:

+ +   +p +p / +c +c / ..
+ -   ..

It should be clearer after that.

Pierre

[Bryan Sanctuary]

Jan Ank is wrong. I am saying the opposite, to count each event once, not twice, the way experiments are done.

Bryan

[Bryan Sanctuary]

Hi Chantal,

How many particles are there in your example? The total number of coincidence clicks from all sources cannot exceed the total number of EPR pairs-- one coincidence per pair, one click per particle.  In your example it seems you get 4 clicks from 2 particles.

Bryan

To view this discussion on the web visit https://groups.google.com/d/msgid/Bell_quantum_foundations/9cbae5f8-2b07-4c5f-a4c7-e74c88fac753%40app.fastmail.com.