EPR, Bell and (none-)locality again

[joigus]

9 hours ago, Markus Hanke said:

Actually, I’m hoping that there might be some convincing reason to once and for all rule out superdeterminism (I don’t like the idea) - but I can’t find any, and a number of quite esteemed physicists seem to pursue this line of research.

I know. Likewise, I don't see a reason why it would not be possible in principle. If you take determinism really seriously, it's kind of a given really. Determinism is an inevitable logical consequence of: (1) generalised use of analytic functions, and (2) the physical law abiding by the principle of the differential equation.

If that paradigm is correct, you can always reconstruct the value of all physical parameters everywhere and at every time by just Taylor-expanding the functions, provided you could measure field inhomogeneities with infinite precision (derivatives of arbitrarily high order.)

It's just a corolary of our model (perhaps delusional?) of the world.

The alternatives always involve step-wise changes that are unpredictable, but have the unfortunate[?] character of being ad hoc, somewhat contrived... (Collapse of the wave function, fractal algorithmic "steps", discrete updating law of cellular automata...)

[Eise]

On 9/22/2023 at 2:18 PM, Genady said:

I can think of it as a conservation of something. Regardless of how many interactions occur, for how long, and how far away they spread, this something has to be conserved, and this conservation means that all the things involved continue to stay correlated.

But that would be plain vanilla determinism. Causal determinism is the default assumption of natural science, and QM seems to be the exception. But generally the  assumption holds very well: again and again we find laws of nature that show how events are causally connected to each other. So for me the 'proof' that determinism is a correct assumption, is that we can formulate how events are related, not just that they are related. Showing that (space-like separated) measurements of entangled particles in principle could be explained deterministically is far from formulating how this happens. With other words, to be convincing, it should be showed that e.g very different ways of switching the direction of the polarisators (researcher manually turns them, a mechanism turns them, (based on a pseudo random generator, or based on a quantum randomness generator, or based on the radio noise of two remote quasars), or using 2 pair of polarisators with an optical switch in front of them, etc), all lead to the same correlations between the measurements. I have no problem with the idea that everything is determined. But I have a problem that such different ways of 'choosing' the polarisation orientations always result in the same correlations. What kind of (hidden?) laws of nature can manage that?

In this article, the author comes not very far in my opinion:

Louis Vervoort, Bell’s Theorem: Two Neglected Solutions:

Quote

Now, if one takes the Big-Bang theory into account, it would appear that the idea of a universal causal ‘branching’ or ‘network’ between events, originating at the Big-Bang and connecting (almost) ‘anything to anything’, is a quite natural conclusion. In sum, measurement independence appears to be in contradiction with a simple ontology, the mentioned total or superdeterminism (actually, determinism would suffice as a term)

So plain vanilla determinism would do the trick? Simple causal and local determinism?

Quote

It points however to a very different kind of non-locality, namely a universal connectedness of virtually all systems (including human beings), due to a receding world cone. Of course, if S3 (S3 = Superdeterminism, Eise) would be true, the truly mind-boggling thing would be that quantum mechanics and Bell’s theorem allow us to discover this ancient and universal link between virtually all objects; and to corroborate a millennia old philosophy. We believe that the main argument in favor of superdeterminism is that it corresponds to the simplest worldview, based on the fewest concepts (and should one not adopt the simplest theory agreeing with the facts?). Its main drawback is that it may be difficult to directly transpose it in a physical theory. (At least this is Bell’s position, see [16]. The reason invoked is that theories that describe both our choices and Bell experiments by explicitly exhibiting common parameters are doubtlessly impossible. Other people are not impressed [9].)

[9] is 't Hooft's book on cellular automaton solution.

In bold: what? The author even gives a (very) short outline of Spinoza, to show this point, but Spinoza is talking about plain vanilla determinism. My position is that determinism is OK (with QM as the exception), but pre-determinism is not. Superdeterminism smells very strong like pre-determinism.

And yes, I tried to read 't Hooft's book, but as I expected, the technicalities are above my head. However, my impression is that there is not even an 'Ansatz' of a theory that could explain how such differently determined pathways to orient the polarisators all lead to the same result. Showing that (some interpretations of) QM allow for this, is wholly different as hypothesising a mechanism for it, (that could be tested?)

 

Edited September 25, 2023 by Eise

[joigus]

Although I think I totally understand @Genady's point, the problem with conservation laws as standalones to spell out determinism, is the following, brilliantly exposed, as I remember, in Landau's mechanics: If you manage to solve the differential equations of the system by brute force or sophisticated methods, you can always look at it like this "movie" which mathematically is represented by the explicit q(t) q'(t) (all coordinates and velocities of all degrees of freedom for all times). Once you've done that, you can always claim certain functions f₁(q,q´), ..., f_n-1(q,q´) are integrals of motion AKA conserved quantities.

I've just found the quote:

The problem is, of course, what Landau quickly points out next:

What @Eise calls "vanilla determinism", if I understood him correctly, might be related to this particular point. It is one thing to assert (perhaps have a theoretical principle to do so) that tells us that all information of relevance "is there". It really is there!!! Yes, but a very different thing altogether is to be able to predict, retrodict, etc. with that information. For that to be done, it might be necessary to have the system speak our language. That means integrals of motion that we can measure and find the correlations of.

What I would like to be done within my lifetime is a robust proof, both theoretical and experimental, that Nature is deterministic in principle, but undeterministic de facto, presumably in way similar to how it's done in classical mechanics as concerns chaos theory. 

I think it's been tried and it hasn't exactly worked as expected. Quantum chaos, AFAIK, is not chaos of the quantum, although there are interesting connections.

 

* Even a deterministic sub-level that we are not given to solve or use, or perhaps understand.

BTW, what I call n in my previous comment (number of coordinates and velocities) is what Landau calls 2s.

[Genady]

It seems to me that all efforts of spacetime interpretation of quantum states are akin efforts to prove Euclid's fifth axiom. The discovery of states which exist independently of spacetime sheds light on a deeper structure of the world of events.

[Markus Hanke]

On 9/25/2023 at 4:55 PM, Genady said:

The discovery of states which exist independently of spacetime sheds light on a deeper structure of the world of events.

Yes, very good point +1

[Eise]

12 hours ago, Markus Hanke said:

On 9/25/2023 at 4:55 PM, Genady said:

The discovery of states which exist independently of spacetime sheds light on a deeper structure of the world of events.

Yes, very good point +1

I agree too. I had another picture, instead of proving Euclid's fifth axiom: Plato's cave. But instead of seeing projections of eternal ideas, we have a glimpse on a dynamic reality that lies beneath our observations in space and time. You, @Markus Hanke, amplified this idea when you explained the singlet state in the thread-we-do-not-mention: distance (or space-time interval) between the measurements simply does not appear in the state of an entangled pair. So what we observe is the projection of a 'beneath quantum physics reality' on space-time.

My view is more or less Kantian: it might be that we are encountering the 'Ding-an sich' (thing-in-itself), meaning we cannot look further because of our limitation to observing events in space-time. Space and time may not be the fixed categories Kant originally thought, but our observations will still always be in space-time. We may try to tickle as much as we want beyond our limited cognitive capabilities, and maybe we will discover more 'EPR-Bell-Kochen-Specker-Clauser-Zeilinger' (in)equalities. But probably these will just astonish us just more, maybe will exclude the one or the other interpretation of QM, but not a model of reality as it is under the hood. 

We are touching the limit of empirical science here: it is so to speak the 'micro-equivalent' of the observable universe. But not because of a practical limit, but because of our cognitive limitations.

[Genady]

Mathematical models do not have such limitations, IMHO.

Edited September 27, 2023 by Genady

[joigus]

I totally agree with @Genady too. We seem to be fo the same mind in this respect, or very similar.

The way Kant analised Nature and science, and epistemology, blew my mind when I was a teenager. Didn't follow him quite equally well when he spoke about other things, moral in particular, or "practical reason."

In fact, I see many clues that the world we see and measure could be but a projection from some more "fundamental" level, perhaps involving things other than geometry. Watching for clues from pure mathematics could not be entirely misguided.

[Eise]

OK, I am trying to read (and understand) Superdeterminism: A Guide for the Perplexed, by Sabine Hossenfelder. Until now, I do not find it convincing, but maybe some of you (Genady, Joigus, Markus?) would like to read, comment and discuss it here? 

Here I have already a problem:

Quote

What does it mean to violate Statistical Independence? It means that fundamentally everything
in the universe is connected with everything else, if subtly so. You may be tempted to ask
where these connections come from, but the whole point of superdeterminism is that this is just
how nature is.

Bold by me.

Bell's inequalities are based on:

• locality
• realism
• statistical independence

Every classical theory should 'obey' Bell's inequality. QM does not, so at least one of the suppositions must be dropped for QM. Superdeterminism would be dropping statistical independence. But at the same time Hossenfelder gives up on locality, and adds 'this is just how nature is'. Sounds like giving up science. 

I also found this blog, by Scott Aaronson, which argues ferociously against Hossenfelder's position. A lot of reactions on it.

[joigus]

Thanks, Eise.

I'll have to look into it more deeply, but for the time beeing it checks with my understanding that superdeterminism is similar to the observation that bipartite, tripartite etc entangled states are correlated (strangely, non-realistically, unintuitively, so on) from the beginning. It's that statement but on steroids. Namely: Everything is entangled from the very beginning (or perhaps non-beginning, as the Mahayana teachers say) of time.

Something like that, but I would like to read more about it.

[Markus Hanke]

I agree with @joigus, that’s how I would look at it too. But I, too, must do some further reading when I’ve got the time.

And yes, SupD is certainly far from being an accepted consensus, and very much a minority view.

[Eise]

Well, if you suppose that after every interaction between particles, they are entangled, then yes, of course. But then, the history of those particles is not ended: they will interact with a lot of more particles, so 'diluting the entanglement' after this 'first' interaction beyond recognition. And that makes superdeterminism for me unacceptable. How could then the measurements in entanglement experiments work together in such a perfect way that it suggests that local realism is invalid? Due to all interactions, all parts of the experiment and the experimenters themselves consist of particles that can have wildly different histories, so are, FAPP, random.

[joigus]

Yes, @Eise. You've just kind of voiced my concerns here. I don't know in detail about those critical voices, but I'm sure they must sound something like this: So we have to accept that, even though everything in the universe thermalises very quickly, somehow this information, which is dynamical in nature, must be protected from thermalising so that one day in a laboratory in, say, Vienna, a physicists chooses a polarisation direction and the universe conjures up that information?

While not impossible, it rings totally wrong. It goes against everything else we know about entropy, the arrow of time, etc. Degrees of freedom thermalise, mix, get blurred out with time. What magical DoF's are these?

Edited October 4, 2023 by joigus
user-quote added

[Eise]

Well, I finished 'Superdeterminism: A Guide for the Perplexed', and it leaves me just as perplexed as I was before. Also 'Rethinking Superdeterminism' by Hossenfelder and Palmer did not help.

H seems to state that the influence of the orientation of the polarisers is still local (but how does this fit to her remark that "fundamentally everything in the universe is (subtly) connected with everything else"?). I simply do not get it. Anybody here that can explain H's and P's position?

[joigus]

4 hours ago, Eise said:

Well, I finished 'Superdeterminism: A Guide for the Perplexed', and it leaves me just as perplexed as I was before. Also 'Rethinking Superdeterminism' by Hossenfelder and Palmer did not help.

H seems to state that the influence of the orientation of the polarisers is still local (but how does this fit to her remark that "fundamentally everything in the universe is (subtly) connected with everything else"?). I simply do not get it. Anybody here that can explain H's and P's position?

Well, I cannot be sure 100% of what they mean. But I would say that they must be implying that all information that determines what polarisation direction the experimenters are going to measure is somehow "diluted" in the dynamical information that's been going around for billions of years, and that information can be "tapped" locally at the moment of the measurement actually being performed.

I cannot conceive of any other sense in which they can be speaking. IOW, information can propagate strictly locally, and yet have had plenty of time to reach every corner of the universe, so to speak.

[Eise]

Just another read: Superdeterminism is unscientific, by Mateus Araújo.

And obviously such discussions become so fiercely that people even refuse to discuss it further. See Sabine Hossenfelder's blog, where Mateus Araújo after a while is refused further reactions.

 

[joigus]

Yes, it's almost as if it's all a matter of how much you perceive the other side is clutching at straws really.

I have to say I do not really like this kind of physics of impossibility theorems, of what is possible and what isn't.

In pure mathematics, everything is crystal clear --technical difficulties apart--. Your premises are what you say they are. In physics, on the other hand, it seems as though it were always possible to relax the hypotheses some way or another, even at the cost of making extremely unnatural or strangely contrived assumptions.

IMO, people who are working on SD are living dangerously, while people who prefer to think in terms of multiple realities are too narrow-minded. Some synthesis will appear eventually and it will feel like "how could we not have thought of this before?" It's obvious to me that's dropping some implicit or hidden assumption that has been invisible to us so far.