Superdeterminism makes a comeback

Foundations of physics and/or philosophy of physics, and in particular, posts on unresolved or controversial issues

Superdeterminism makes a comeback

https://arxiv.org/abs/2108.07292

Supermeasured: Violating Statistical Independence without violating statistical independence
T.N. Palmer, S. Hossenfelder, J.R. Hance

Bell's theorem is often said to imply that quantum mechanics violates local causality, and that local causality cannot be restored with a hidden-variables theory. This however is only correct if the hidden-variables theory fulfils an assumption called Statistical Independence. Violations of Statistical Independence are commonly interpreted as correlations between the measurement settings and the hidden variables (which determine the measurement outcomes). Such correlations have been discarded as "finetuning" or a "conspiracy". We here point out that the common interpretation is at best physically ambiguous and at worst incorrect. The problem with the common interpretation is that Statistical Independence might be violated because of a non-trivial measure in state space, a possibility we propose to call "supermeasured". We use Invariant Set Theory as an example of a supermeasured theory that violates the Statistical Independence assumption in Bell's theorem without requiring correlations between hidden variables and measurement settings.

I’ve challenged Sabine er al. to implement this model in a computer simulation and win 65 thousand Euro (and Nobel prizes)
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

gill1109 wrote:https://arxiv.org/abs/2108.07292

Supermeasured: Violating Statistical Independence without violating statistical independence
T.N. Palmer, S. Hossenfelder, J.R. Hance

Bell's theorem is often said to imply that quantum mechanics violates local causality, and that local causality cannot be restored with a hidden-variables theory. This however is only correct if the hidden-variables theory fulfils an assumption called Statistical Independence. Violations of Statistical Independence are commonly interpreted as correlations between the measurement settings and the hidden variables (which determine the measurement outcomes). Such correlations have been discarded as "finetuning" or a "conspiracy". We here point out that the common interpretation is at best physically ambiguous and at worst incorrect. The problem with the common interpretation is that Statistical Independence might be violated because of a non-trivial measure in state space, a possibility we propose to call "supermeasured". We use Invariant Set Theory as an example of a supermeasured theory that violates the Statistical Independence assumption in Bell's theorem without requiring correlations between hidden variables and measurement settings.

I’ve challenged Sabine er al. to implement this model in a computer simulation and win 65 thousand Euro (and Nobel prizes)

As I understand it from the abstract, they only propose a new interpretation for measurement independence violation. This should mean nothing to you since you believe that locality can be retained by rejecting CFD.
There is no doubt the inequality can be violated through violation of measurement independence.
Those who understand that rejection of realism doest not mean anything (be it CFD or something else, see my comment to Griffith) consider rejection of MI very seriously, for instance, Gerard t' Hooft.
By the way your challenge cannot disprove violation of MI.
Justo

Re: Superdeterminism makes a comeback

Justo wrote:
gill1109 wrote:https://arxiv.org/abs/2108.07292

Supermeasured: Violating Statistical Independence without violating statistical independence
T.N. Palmer, S. Hossenfelder, J.R. Hance

Bell's theorem is often said to imply that quantum mechanics violates local causality, and that local causality cannot be restored with a hidden-variables theory. This however is only correct if the hidden-variables theory fulfils an assumption called Statistical Independence. Violations of Statistical Independence are commonly interpreted as correlations between the measurement settings and the hidden variables (which determine the measurement outcomes). Such correlations have been discarded as "finetuning" or a "conspiracy". We here point out that the common interpretation is at best physically ambiguous and at worst incorrect. The problem with the common interpretation is that Statistical Independence might be violated because of a non-trivial measure in state space, a possibility we propose to call "supermeasured". We use Invariant Set Theory as an example of a supermeasured theory that violates the Statistical Independence assumption in Bell's theorem without requiring correlations between hidden variables and measurement settings.

I’ve challenged Sabine er al. to implement this model in a computer simulation and win 65 thousand Euro (and Nobel prizes)

As I understand it from the abstract, they only propose a new interpretation for measurement independence violation. This should mean nothing to you since you believe that locality can be retained by rejecting CFD.
There is no doubt the inequality can be violated through violation of measurement independence.
Those who understand that rejection of realism doest not mean anything (be it CFD or something else, see my comment to Griffith) consider rejection of MI very seriously, for instance, Gerard t' Hooft.
By the way your challenge cannot disprove violation of MI.

Justo, as I understand their abstract, they do not propose to modify any of the standard definitions. They propose using some non standard form of measure theory on the state space, which must be the space of hidden variables. I think they think they can escape Bell’s theorem by using some abstract and uncomputable mathematical structures. I suspect that nobody will understand this paper. I think they are playing meaningless word games which nobody else can play except the authors.

I know that some famous people reject MI. I think they too are wasting their time. You can always appeal to superdeterminism. Present day experiments can be explained in that way. But the explanation is ludicrous. And it is useless. It does not lead to testable predictions. It is just a word game for insiders, a comfort blanket.
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

The invariant set postulate on which they rely boils down to an affirmation of modal logic. The choice of the term "superdeterminism" is very unfortunate because it is actually very easy to understand. In the final analysis "superdeterminism" is just a rejection of counterfactual definiteness.

The modal logic states that if it is true today that tomorrow you will eat a banana, then tomorrow you will eat a banana and there is nothing you can do to change it. But You are also an absolutely free agent since it is the truth value of the statement which results from your choice to eat a banana tomorrow. The fact that in the future you chose to eat a banana, causes the statement that you would eat a banana tomorrow to be true. It is NOT the statement that you will eat a banana tomorrow which causes you to eat a banana tomorrow without regard for your freedom. In other words, the occurrence of events causes the statement of their occurrence to be true and not the reverse. Nothing is super determined. All that has happened is that you can't go back and change what has happened already. It doesn't mean you could not have chosen otherwise before.

Translating to the Bell situation, Alice and Bob can't go back and chose different settings. This does not mean they could not have chosen different settings to begin with. The set of choices already made in the past is invariant and separated from the set of choices not made.
minkwe

Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Superdeterminism makes a comeback

minkwe wrote:The invariant set postulate on which they rely boils down to an affirmation of modal logic. The choice of the term "superdeterminism" is very unfortunate because it is actually very easy to understand. In the final analysis "superdeterminism" is just a rejection of counterfactual definiteness.

The modal logic states that if it is true today that tomorrow you will eat a banana, then tomorrow you will eat a banana and there is nothing you can do to change it. But You are also an absolutely free agent since it is the truth value of the statement which results from your choice to eat a banana tomorrow. The fact that in the future you chose to eat a banana, causes the statement that you would eat a banana tomorrow to be true. It is NOT the statement that you will eat a banana tomorrow which causes you to eat a banana tomorrow without regard for your freedom. In other words, the occurrence of events causes the statement of their occurrence to be true and not the reverse. Nothing is super determined. All that has happened is that you can't go back and change what has happened already. It doesn't mean you could not have chosen otherwise before.

Translating to the Bell situation, Alice and Bob can't go back and chose different settings. This does not mean they could not have chosen different settings to begin with. The set of choices already made in the past is invariant and separated from the set of choices not made.

Superdeterminism means that experimentalists are deluded by thinking that they choose which experiments they will perform. Physics is reduced to "what will happen, will happen". God has determined everything according to his own unknowable wisdom, at the time of the big bang. The idea that we can predict anything in the future by observing mere statistical regularities in the past is an illusion, since the future is already determined.

Suppose we perform a Bell experiment with Alice on Mars and Bob on Jupiter. Alice determines her measurement settings by literally tossing coins. Bob uses a state of the art pseudo random number generator with seed determined by his grandmothers birth date. According to Hossenfelder et al. the violation of Bell inequalities comes about because of an exquisite connection between Bob's grandmother's birthday and the prime numbers used in his RNG and Alice's photodetectors. Personally, I find this a ludicrous idea, but who knows, maybe this will enable those folk to unify relativity and QM at last, so from a mathematical point of view it could be very interesting. Maybe it will even lead to correct future physical predictions which nobody could have expected, so it is great that people explore such avenues, even if they are unpopular and at first sight ludicrous. Newton's theory was also thought at first sight, ludicrous. Changing the position of the moon would instantaneously change the forces felt in a glass of water. But the thing was, the water molecules would take some time to actually move. So the theory was not quite so crazy as long as you thought of the force of gravitation as being something virtual, not actually real. Whatever that means. Classical physics does allow CFD (counterfactual determinateness). Einstein used CFD in the EPR thought experiment to *justify* realism and hence to argue for the *incompleteness* (not the incorrectness) of quantum theory.
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

minkwe wrote:The invariant set postulate on which they rely boils down to an affirmation of modal logic. The choice of the term "superdeterminism" is very unfortunate because it is actually very easy to understand. In the final analysis "superdeterminism" is just a rejection of counterfactual definiteness.

The modal logic states that if it is true today that tomorrow you will eat a banana, then tomorrow you will eat a banana and there is nothing you can do to change it. But You are also an absolutely free agent since it is the truth value of the statement which results from your choice to eat a banana tomorrow. The fact that in the future you chose to eat a banana, causes the statement that you would eat a banana tomorrow to be true. It is NOT the statement that you will eat a banana tomorrow which causes you to eat a banana tomorrow without regard for your freedom. In other words, the occurrence of events causes the statement of their occurrence to be true and not the reverse. Nothing is super determined. All that has happened is that you can't go back and change what has happened already. It doesn't mean you could not have chosen otherwise before.

Translating to the Bell situation, Alice and Bob can't go back and chose different settings. This does not mean they could not have chosen different settings to begin with. The set of choices already made in the past is invariant and separated from the set of choices not made.

That is not what their paper is about.

If it were, it would have been the most trivial paper ever written.

I'm about halfway trough it, and so far they have taken a correlated distribution over an uncorrelated (uniform) measure space and Radon-Nikodym transformed it to an uncorrelated distribution over a correlated measure. Well thank you very much, but that solves absolutely nothing. I will read the last part with an open mind, but by skim reading it seems they will soon introduce some crazy unphysical fractal measure.
Heinera

Posts: 917
Joined: Thu Feb 06, 2014 1:50 am

Re: Superdeterminism makes a comeback

gill1109 wrote:Superdeterminism means that experimentalists are deluded by thinking that they choose which experiments they will perform.

This is hyper sensationalism. That's why I think the name "superdeterminism" like many others in Quantum Foundations like "non-locality", etc are unfortunate choices.

Physics is reduced to "what will happen, will happen".

Absolutely, that is what physics is. What will happen, will happen. What alternative do you prescribe:

1. What won't happen, will happen?
2. What will happen, won't happen?

You see, you are jumping to conclusions based purely on sensationalism without doing the required analysis.

God has determined everything according to his own unknowable wisdom, at the time of the big bang.

More unfounded sensationalism. Focus on the analysis, not the rhetorical devices.

The idea that we can predict anything in the future by observing mere statistical regularities in the past is an illusion, since the future is already determined.

Here you commit the modal fallacy. We can predict things because "what will happen, will happen". So contrary to your claim it's the alternative that is an illusion. Secondly, the future is in fact determined, conditionally. If Alice measures at "a" she will obtain A(a), but if she measures at " a' " she will obtain A(a'). So yes, the future is determined conditionally. However, in hindsight, the future is determined exactly. What has happened has happened. Therefore what will happen will happen.

Suppose we perform a Bell experiment with Alice on Mars and Bob on Jupiter. Alice determines her measurement settings by literally tossing coins. Bob uses a state of the art pseudo random number generator with seed determined by his grandmothers birth date. According to Hossenfelder et al. the violation of Bell inequalities comes about because of an exquisite connection between Bob's grandmother's birthday and the prime numbers used in his RNG and Alice's photodetectors.

More sensationalism. They are using invariant set theory to say "fine-tuning", "non-locality" or "conspiracy" are not correct interpretations of any statistical dependence that exists between measurement settings and hidden variables. Don't mischaracterize them in order to argue against them. The paper you mention is actually not about superdeterminism at all.
Anyone who understands basic probability theory can look at equation (1) in their paper:

$\rho(\lambda|X) = \rho(\lambda)$

And see that being able to freely pick settings X is not the same thing as being able to freely pick hidden variables for X. They show in this paper that it is possible to violate statistical independence and yet have $\rho(\lambda|X) = \rho(\lambda)$ for state spaces with non-trivial measure.
minkwe

Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Superdeterminism makes a comeback

.
It looks like Sabine Hossenfelder has unfriended Richard Gill on Facebook after he spewed a huge amount of nonsense about her paper on Facebook.
.
Joy Christian
Research Physicist

Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Superdeterminism makes a comeback

Joy Christian wrote:.
It looks like Sabine Hossenfelder has unfriended Richard Gill on Facebook after he spewed a huge amount of nonsense about her paper on Facebook.
.

Seems funny, how can we find those dicussions?
Justo

Posts: 83
Joined: Fri Aug 20, 2021 8:20 am

Re: Superdeterminism makes a comeback

Justo wrote:
Joy Christian wrote:.
It looks like Sabine Hossenfelder has unfriended Richard Gill on Facebook after he spewed a huge amount of nonsense about her paper on Facebook.
.

Seems funny, how can we find those dicussions?

You may be able to access her public post on Facebook: https://www.facebook.com/joyjch/posts/1 ... 06242856:8
.
Joy Christian
Research Physicist

Posts: 2793
Joined: Wed Feb 05, 2014 4:49 am
Location: Oxford, United Kingdom

Re: Superdeterminism makes a comeback

https://arxiv.org/pdf/2108.07292.pdf wrote:While Bell’s theorem is often said to imply that local causality (which is violated by standard quantum mechanics) cannot
be restored with a deterministic hidden variables theory, this is only correct if the hidden-variables theory respects Statistical
Independence. Violations of Statistical Independence are commonly interpreted as implying a correlation between the measurement settings and the hidden variables which determine the measurement outcomes. However, as we have shown here, one can
violate the Statistical Independence assumption in Bell’s theorem without any correlations between the measurement outcomes
and the hidden variables. The violations of Statistical Independence can instead come about by the geometry of the underlying
state space. We have argued that this is a simple way to see that violating Statistical Independence does not require fine tuning

The conclusion from this paper sounds familiar. I wonder where I read something like this before.
minkwe

Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Superdeterminism makes a comeback

minkwe wrote:
https://arxiv.org/pdf/2108.07292.pdf wrote:While Bell’s theorem is often said to imply that local causality (which is violated by standard quantum mechanics) cannot
be restored with a deterministic hidden variables theory, this is only correct if the hidden-variables theory respects Statistical
Independence. Violations of Statistical Independence are commonly interpreted as implying a correlation between the measurement settings and the hidden variables which determine the measurement outcomes. However, as we have shown here, one can
violate the Statistical Independence assumption in Bell’s theorem without any correlations between the measurement outcomes
and the hidden variables. The violations of Statistical Independence can instead come about by the geometry of the underlying
state space. We have argued that this is a simple way to see that violating Statistical Independence does not require fine tuning

The conclusion from this paper sounds familiar. I wonder where I read something like this before.

Violating Statistical Independence (‘SI’) does not require fine tuning. That is true, Sabine and Tim already proved that a couple of years ago. The paper distinguishes between ‘si’ and “SI”. I believe the authors don’t really appreciate the difference between a probability measure and a probability density. If a density factors with respect to a non-factoring measure, one can do useful things with that, but not explain quantum nonlocality (violation of Bell inequalities by QM predictions in certain space-time contexts).

Trouble is, they introduce a topology of space which is built around fractal sets. Their topology is not much use in practice. A numerical approximation of a fractal does not get closer and closer, in their topology, to the fractal as numerical accuracy increases. It is impossible to faithfully simulate their model.

The work is very interesting! Worth reading carefully.
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

Justo wrote:
Joy Christian wrote:.
It looks like Sabine Hossenfelder has unfriended Richard Gill on Facebook after he spewed a huge amount of nonsense about her paper on Facebook.
.

Seems funny, how can we find those dicussions?

Justo, become my FB friend and I can show you the discussions. The problem was, I think, that some of my "FaceBook friends" [incidentally, they are also FB friends of Joy!] started saying really nasty personal things about Sabine. I told them they should not do that but it was too late. Sabine was fed up, and rightly so. I've apologised to her.

Actually, I did not "spew a huge amount of nonsense" about her paper. I made some wise, critical comments. Unfortunately, some people take criticism of their work a bit too personally.
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

gill1109 wrote:Justo, become my FB friend and I can show you the discussions. The problem was, I think, that some of my "FaceBook friends" [incidentally, they are also FB friends of Joy!] started saying really nasty personal things about Sabine. I told them they should not do that but it was too late. Sabine was fed up, and rightly so. I've apologised to her.

Actually, I did not "spew a huge amount of nonsense" about her paper. I made some wise, critical comments. Unfortunately, some people take criticism of their work a bit too personally.

I don't believe you, I believe Joy. You are just addicted to making personal attacks for no reason. One thing is to attack scientific ideas another is to insult people.
Justo

Posts: 83
Joined: Fri Aug 20, 2021 8:20 am

Re: Superdeterminism makes a comeback

Justo wrote:
gill1109 wrote:Justo, become my FB friend and I can show you the discussions. The problem was, I think, that some of my "FaceBook friends" [incidentally, they are also FB friends of Joy!] started saying really nasty personal things about Sabine. I told them they should not do that but it was too late. Sabine was fed up, and rightly so. I've apologised to her.

Actually, I did not "spew a huge amount of nonsense" about her paper. I made some wise, critical comments. Unfortunately, some people take criticism of their work a bit too personally.

I don't believe you, I believe Joy. You are just addicted to making personal attacks for no reason. One thing is to attack scientific ideas another is to insult people.

Dear Justo, I’m very sorry to hear that you feel that way.

Oh well, your remarks will delight Joy and several other people.
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden

Re: Superdeterminism makes a comeback

Justo wrote: You [Gill] are just addicted to making personal attacks for no reason.

The reason is simple. He's a stone-cold narcissist that tries to raise himself up by lowering others. And he seems to be doing a lot of apologizing lately. You'd think he would have a personal epiphany about his behavior, but narcissists never see anything wrong in themselves. They may feign it but the behavior never changes.
local

Posts: 295
Joined: Mon Aug 05, 2019 1:19 pm

Re: Superdeterminism makes a comeback

Ok guys, we are getting a bit too personal here.
.
FrediFizzx
Independent Physics Researcher

Posts: 2905
Joined: Tue Mar 19, 2013 7:12 pm
Location: N. California, USA

Re: Superdeterminism makes a comeback

gill1109 wrote:
Justo wrote:
gill1109 wrote:Justo, become my FB friend and I can show you the discussions. The problem was, I think, that some of my "FaceBook friends" [incidentally, they are also FB friends of Joy!] started saying really nasty personal things about Sabine. I told them they should not do that but it was too late. Sabine was fed up, and rightly so. I've apologised to her.

Actually, I did not "spew a huge amount of nonsense" about her paper. I made some wise, critical comments. Unfortunately, some people take criticism of their work a bit too personally.

I don't believe you, I believe Joy. You are just addicted to making personal attacks for no reason. One thing is to attack scientific ideas another is to insult people.

Dear Justo, I’m very sorry to hear that you feel that way.

Oh well, your remarks will delight Joy and several other people.

I think it is possible to learn a thing or two even in old age so here for what it's worth, the second sentence did not need to be uttered. Absolutely no purpose to it than to troll. This is the problem. Stick to the analysis.
minkwe

Posts: 1441
Joined: Sat Feb 08, 2014 10:22 am

Re: Superdeterminism makes a comeback

Joy Christian wrote:.
It looks like Sabine Hossenfelder has unfriended Richard Gill on Facebook after he spewed a huge amount of nonsense about her paper on Facebook.

I apologised to Sabine for what some of *our* mutual FB friends [FB friends of all three of us: Joy, Sabine and me] had said about her on my feed. Sabine and I have agreed to further discuss the maths of her paper with Tim and Jonte by email.

minkwe wrote:I think it is possible to learn a thing or two even in old age … Stick to the analysis.

Sure.

As I told you Michel, your problems with Bell (1964) were (I think) already raised by quite a few people, including many eminent physicists, back in the late 60’s. Bell answered that criticism, quite adequately I think. Over the years his writing becomes clearer, in my opinion, probably because he also learnt from the tough criticism of other scientists. Please tell us what you think of Chapter 8 of his book. (Locality in quantum mechanics: Reply to critics). The closing passage, just half a page. https://cds.cern.ch/record/980330/files/CM-P00061609.pdf
gill1109
Mathematical Statistician

Posts: 2812
Joined: Tue Feb 04, 2014 10:39 pm
Location: Leiden