Re: Conference summary ... and what is measurement?

9 views
Skip to first unread message

Richard Gill

unread,
Feb 12, 2023, 12:21:28 PM2/12/23
to Jarek Duda, Marc Fleury, nature of time, models-of...@googlegroups.com, Bell Inequalities and quantum foundations
Dear all

After some email exchanges with Tim Palmer I think I understand at last what he is doing.

He restricts QM to a "thin subset” of states and (joint) measurements which is everywhere dense but which does not violate Bell inequalities in the sense that at least one of the four precise sub-experiments is forbidden. If Alice wants to use angles a and a’, and Bob wants to use angles b and b’,  then at least one of the joint experiments (combining one of Alice’s with one of Bob’s angles) does not exist.

The theory is by design incompatible with Bell's LHV concept! It does allow classical (deterministic) description with local and real hidden variables. There is lambda and there are functions (A(a, lambda), B(b, lambda)) but the pair is not defined on a whole Cartesian product of sets of settings a, settings b, and hidden variables lambda. The allowed settings come from a dense subset and the allowed pairs from another dense subset.

I think that this theory is irrelevant to the evaluation of actual Bell experiments. We do not "assume QM" in order to evaluate such experiments. We assume LHV and show the data is incompatible with LHV. We do not assume anything concerning angles or quantum states, when we analyse the data from such an experiment. Of course, we try to engineer according to our usual QM picture. That's allowed, is it not? So Palmer’s model escapes Bell’s theorem by not describing the experiments which Bell is interested in. Palmer's theory does not say anything whatsoever about the data actually observed in actual successful Bell experiments (which does turn out to be compatible with ordinary QM, but not with LHV).

One cannot simulate data according to Palmer’s model in a local realistic way since Alice's measurement angle is restricted by Bob's (or vice versa). Of course, one can simulate from a close-by model if Alice’s chosen setting is allowed to “nudge” Bob’s chosen setting by a tiny amount. That does require one bit of near instantaneous communication from Alice to Bob’s place (or vice versa), if we suppose that Alice and Bob have already agreed to perform something very close to a usual QM experiment for pre-chosen a, a’, b, b’. If they are planning to use many more angles, then much more communication will be needed.

Thus the model can describe EPR-B data to any required degree of accuracy, but in order to do this, action at a distance has to be used.

Richard

PS you can find Tim’s talk here https://youtu.be/lZP0EFTNPu0

,
On 11 Feb 2023, at 21:32, Jarek Duda <dud...@gmail.com> wrote:

Dear Colleagues,

Thank you very much for great talks and discussion. The recordings should be available in a day or two on youtube - till the end of last of talk, let us know if somebody doesn't want to be there so we will cut.

Marc,

This was slide 2 of https://www.dropbox.com/s/a8yqfabq3gxsjth/Bell%20mini.pdf

Sergey article: https://www.preprints.org/manuscript/202210.0478/v1

From evolution equation (3) he gets this super simple up/down alignment condition (25):

O8tLBgdNZXkatbMP.png

Looks nice, but I doubt it is sufficient for Bell violation (?)


Generally, bispinor encodes 3D spin direction (a,b,c) below: https://en.wikipedia.org/wiki/Bispinor#Construction_of_Dirac_spinor_with_a_given_spin_direction_and_charge

ip2f0lMxBZq9EjAG.png

So non-polarized particles with spin I basically imagine as tiny magnets pointing a random direction (?)

In external magnetic field tiny magnets have tendency to align - e.g. in Stern-Gerlach.

With this alignment there should come difference in energy (kinetic of Larmor precession) and angular momentum (toward spin) - which should be radiated.

Precessing magnet is cylindrically symmetric kind of antenna - so I would expect the differences are radiated as cylindrically symmetric EM impulse (delocalized in contrast to optical photons).

The big question is if such EM impulse could be observed?

I think it should turn into random noise - heat ... we had some exchange with Sergey and he agreed.

Best wishes,
Jarek

ps. Another big question we have started discussing (also with John Bush) is if electron's de Broglie clock/zitterbewegung is static (always ticking), or only an effect of resonance (normal mode)?

The "always ticking" way has problem of propulsion (pendulum has tendency to calm down to zero).

The resonance way should lose its strength with distance e.g. in Mach-Zehnder, what is not true (?)


-- 
dr Jarosław Duda
Institute of Computer Science and Computer Mathematics,
Jagiellonian University, Cracow, Poland
http://th.if.uj.edu.pl/~dudaj/

--
You received this message because you are subscribed to the Google Groups "nature of time" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nature-of-tim...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nature-of-time/2041549c-bc53-1fb6-2322-da3bcbcb338c%40gmail.com.

Jan-Åke Larsson

unread,
Feb 12, 2023, 12:33:12 PM2/12/23
to Richard Gill, Jarek Duda, Marc Fleury, nature of time, models-of...@googlegroups.com, Bell Inequalities and quantum foundations
Dear all,
Well that's disappointing.

This idea is not even all that new, it's formally equivalent to my 1999 paper.

Best
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/E875412C-E1A3-47CC-ABB9-CE7FBD29B9F7%40gmail.com.

--
Jan-Åke Larsson
Professor, Head of Department


Linköping University
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

Austin Fearnley

unread,
Feb 12, 2023, 3:55:52 PM2/12/23
to Bell inequalities and quantum foundations
Hi Richard

Thank you for showing the link to Tim's talk.  I watched most of it and recognised Tim, you and Jarek.

Combinations of detector angles which do not actually exist reminds me of Joy's model and provides a dual between not-allowed detector settings and disallowed pairs of simulated results, or trimmed data sets.

I agree that comparisons with macro trials on rats etc is unfair criticism of superdeterminism, at least judging against my retrocausal model.  In my model, macro effects cannot reproduce micro effects because of large numbers of particles in the macro objects.  And also in my model the enhanced correlations of Bell rely on retrocausal properties of antiparticles (e.g. positrons) which would average out to zero for macro objects.  (My preon model shows what an anti-photon comprises.)

Twenty+ years ago I programmed some fractal diagrams on an Amstrad computer: very basic sets of points rather than the lovely photos in books.   So I know that you need complex quadratic equations and that the end points in a pattern can be, as was said, a single dot, or a jump to near infinity or the more interesting fractal pattern.  In my programs the pattern evolved additively over time or during a FOR:NEXT loop of commands. I am having trouble seeing a fractal pattern as not having evolved over time. But I appreciate that the state space might have no time dimension in it.  But U and U' were defined as the state spaces of the universe and of universe prime.  Defining the state space U as a point seems to me to overlook that the whole universe is non-local despite being described as a point.

Tim's removal of 'time' off the table sidesteps my own reason for not liking superdeterminism.  I am merely an amateur and I probably have a strawman model of superdeterminism in my mind. To me, superdeterminism relies on retrocausality, so time for me is not off the table.  I have no problem with retrocausality as that is my model and I have a working retrocausal simulation of the Bell experiment with the enhanced correlation outputted. The maths is very simple in my simulation.  Also in my preon model, all elementary particles have both preons and antipreons within them. In one of Susskind's online lectures he said that particles do not set off in flight unless they can be sure of a later target interaction.  Well, he may not have said it quite so retrocausally.  But in my preon model, all particles, including bosons, have forwards and backwards-in-time component preons.  So, when a particle is emitted, its future interaction is already secure.  I thought of superdeterminism in these terms.  Particles cannot be emitted or detected in a Bell experiment unless future interactions are secure.  This could limit the pool of available detector settings of particle flight paths.  ...

Richard Gill

unread,
Feb 12, 2023, 11:24:21 PM2/12/23
to Jan-Åke Larsson, Jarek Duda, Marc Fleury, nature of time, models-of...@googlegroups.com, Bell Inequalities and quantum foundations
Dear Jan-Åke

Indeed, wasn’t the idea of restricting QM to only talk about measurement bases in some countable dense subset earlier used to “disprove” the Kochen-Specker theorem? That is a no-go theorem for non-contextual hidden variables for measurements on a single quantum system of dimension 3 or more. Traditional proofs (Bell, Kochen-Specker) are based on construction of specific finite sets of measurement bases in C^3 including numerous incompatible but overlapping sets of three orthogonal vectors.

See Meyer, D. A., 1999, “Finite Precision Measurement Nullifies the Kochen-Specker Theorem”, Physical Review Letters, 83: 3751–54.

The long history of this idea is discussed in the section called “the question of empirical testing of the Kochen Specker theorem” in the Stanford Encyclopedia of Philosophy article devoted to the Kochen-Specker theorem.

The idea is that finite precision of implementation of measurement bases nullifies experimental proof of such no-go theorems. I have always thought that such arguments are fallacious.

Richard



Sent from my iPad

On 12 Feb 2023, at 18:33, Jan-Åke Larsson <jan-ake...@liu.se> wrote:

 Dear all,
Well that's disappointing.

This idea is not even all that new, it's formally equivalent to my 1999 paper.

Best
Jan-Åke



On 2023-02-12 18:21, Richard Gill wrote:
Dear all

After some email exchanges with Tim Palmer I think I understand at last what he is doing.

He restricts QM to a "thin subset” of states and (joint) measurements which is everywhere dense but which does not violate Bell inequalities in the sense that at least one of the four precise sub-experiments is forbidden. If Alice wants to use angles a and a’, and Bob wants to use angles b and b’,  then at least one of the joint experiments (combining one of Alice’s with one of Bob’s angles) does not exist.

The theory is by design incompatible with Bell's LHV concept! It does allow classical (deterministic) description with local and real hidden variables. There is lambda and there are functions (A(a, lambda), B(b, lambda)) but the pair is not defined on a whole Cartesian product of sets of settings a, settings b, and hidden variables lambda. The allowed settings come from a dense subset and the allowed pairs from another dense subset.

I think that this theory is irrelevant to the evaluation of actual Bell experiments. We do not "assume QM" in order to evaluate such experiments. We assume LHV and show the data is incompatible with LHV. We do not assume anything concerning angles or quantum states, when we analyse the data from such an experiment. Of course, we try to engineer according to our usual QM picture. That's allowed, is it not? So Palmer’s model escapes Bell’s theorem by not describing the experiments which Bell is interested in. Palmer's theory does not say anything whatsoever about the data actually observed in actual successful Bell experiments (which does turn out to be compatible with ordinary QM, but not with LHV).

One cannot simulate data according to Palmer’s model in a local realistic way since Alice's measurement angle is restricted by Bob's (or vice versa). Of course, one can simulate from a close-by model if Alice’s chosen setting is allowed to “nudge” Bob’s chosen setting by a tiny amount. That does require one bit of near instantaneous communication from Alice to Bob’s place (or vice versa), if we suppose that Alice and Bob have already agreed to perform something very close to a usual QM experiment for pre-chosen a, a’, b, b’. If they are planning to use many more angles, then much more communication will be needed.

Thus the model can describe EPR-B data to any required degree of accuracy, but in order to do this, action at a distance has to be used.

Richard

PS you can find Tim’s talk here https://youtu.be/lZP0EFTNPu0

,
On 11 Feb 2023, at 21:32, Jarek Duda <dud...@gmail.com> wrote:

Dear Colleagues,

Thank you very much for great talks and discussion. The recordings should be available in a day or two on youtube - till the end of last of talk, let us know if somebody doesn't want to be there so we will cut.

Marc,

This was slide 2 of https://www.dropbox.com/s/a8yqfabq3gxsjth/Bell%20mini.pdf

Sergey article: https://www.preprints.org/manuscript/202210.0478/v1

From evolution equation (3) he gets this super simple up/down alignment condition (25):

<O8tLBgdNZXkatbMP.png>

Looks nice, but I doubt it is sufficient for Bell violation (?)


Generally, bispinor encodes 3D spin direction (a,b,c) below: https://en.wikipedia.org/wiki/Bispinor#Construction_of_Dirac_spinor_with_a_given_spin_direction_and_charge

<ip2f0lMxBZq9EjAG.png>

Jan-Åke Larsson

unread,
Feb 13, 2023, 1:20:07 AM2/13/23
to Richard Gill, Jarek Duda, Marc Fleury, nature of time, models-of...@googlegroups.com, Bell Inequalities and quantum foundations
Dear Richard,

I was thinking about writing about that but did not have the time to do so.

Indeed that this idea nullifies the results is just wrong. What matters is that there is a discrete set of inputs to the measurement device, or if you will, a discrete set of labels for the inputs. The measurement takes a "setting" as input and generates an "outcome".

If the outcome is generated by a local hidden-variable model, Bell's inequality applies. The finite precision becomes part of the local hidden-variable model.

(If the outcome is generated by a noncontextual hidden-variable model, the KS theorem applies. The finite precision becomes part of the noncontextual hidden-variable model.)

To make these models appear to violate the bounds, these models must always use a more "proper" loophole: they must refuse to answer for certain setting combinations. They must use the detection loophole. And that can be controlled for.

/Jan-Åke
Reply all
Reply to author
Forward
0 new messages