Talk:Bell's theorem

Indicate subtleties in reasoning

I added the qualifier "apparently" (to collapses) and replaced "this means" by "this suggests" in the early sentence in the article describing collapse of the state. After all, we have interpretations of quantum mechanics which deny the physicaal reality of the quantum state and see the "collapse" as a computational device for updating predictions about future measurement results. When an agent's knowledge changes, their predictions about the future might well change too. But physical reality around them has not changed. Richard Gill (talk) 10:21, 20 May 2022 (UTC)

This is the sentence I am referring to, my new words in bold: "When they choose a measurement and obtain a result, the quantum state of the other particle apparently collapses instantaneously into a new state depending upon that result, no matter how far away the other particle is. This suggests that either the measurement of the first particle somehow also interacted with the second particle at faster than the speed of light, or that the entangled particles had some unmeasured property which pre-determined their final quantum states before they were separated." Richard Gill (talk) 11:07, 20 May 2022 (UTC)

Notice that Alice's idea of Bob's particle's state has changed after Alice does her measurement and sees her result, but Bob does not know anything yet. As long as no message from Alice reaches him, he is perfectly right to use the original joint state for his own local prediction purposes. In fact, he can also conveniently reduce the joint density matrix to the reduced state of his subsystem by tracing out the other. So what did collapse? Nothing. Alice's knowledge about the world changed because something happened close by, which she saw. Richard Gill (talk) 11:12, 20 May 2022 (UTC)

The result of Bob's measurement will still be correlated with Alice's in a manner that cannot be explained by a local influence. Talking about quantum states being subjective and people having the right to assign any state they want doesn't explain anything. It's just idle talk by people that insist in denying Bell's 1976 theorem. Tercer (talk) 11:52, 20 May 2022 (UTC)

I agree. I am not denying Bell’s theorem. Richard Gill (talk) 12:17, 20 May 2022 (UTC)

You wrote “ Talking about quantum states being subjective and people having the right to assign any state they want doesn't explain anything.” I never said quantum states are subjective and people have the right to use any state they like. They are objective. But they are not “real” in the sense of being located in space time. Richard Gill (talk) 12:20, 20 May 2022 (UTC)

So quantum states are not subjective, but at the same time they are not located in spacetime. What are they, then? Magical fairies that have any property you need in order to deny the existence of nonlocality? Tercer (talk) 12:27, 20 May 2022 (UTC)

No. Read the Wikipedia article on QBism. “both QBism and RQM [Rovelli’s relational QM] insist that quantum mechanics is a fundamentally local theory. MWI people say so, too. People are emotional about the words “local”, “locality”. I’m a mathematician who also works in Quantum Information and in Quantun Foundations. I try to keep an open mind. There is a central core to QM which everyone does agree on, and many interpretations, which seem to me to be optional extras , since they don’t influence the work of a physicist. Richard Gill (talk) 12:35, 20 May 2022 (UTC)

I'm familiar with QBism and RQM. I know that they claim to be local. It's just not true. They are just throwing up a smokescreen to hide the obvious fact that quantum mechanics is not local, as proven by Bell in 1976. MWI, on the other hand, is actually local, as Bell needs to assume that measurements have a single outcome in order to derive nonlocality. Tercer (talk) 13:23, 20 May 2022 (UTC)

In my opinion, MWI is the big smokescreen! Anyway, I suppose we agree that Bell's theorem is a true theorem. Bell defined a hidden variables model as an essentially deterministic model, and he defined "local hidden variables model" as a special case of "hidden variables model". Quantum mechanics is not a hidden variables model. There are good reasons that one may describe it as a local model, though you may not like Rovelli's reasons or the qBists' reasons. "What's in a name? A rose by any name would smell as sweet". I don't have an axe to grind on this (as I probably said, I'm a mathematician). Anyway: I hope we agree that there are QM predictions which cannot even be approximated by an LHV model, and more importantly, there are real experimental results which cannot be modelled by an LHV model. Richard Gill (talk) 13:40, 20 May 2022 (UTC)

That's a common misconception, and why I'm insisting on Bell 1976 instead of Bell 1964 or CHSH 1969. Bell does not assume determinism. The only assumption is local causality. Read the paper. It's not about hidden variable models, it's about any (single-world) theory at all, and it evidently applies to quantum mechanics, as that was the whole point. Tercer (talk) 15:41, 20 May 2022 (UTC)

Ah, I was wondering why you kept writing 1976, thanks! Cool. But: I have read all the papers, many times. (I even wrote quite a few much cited papers on the topic, too. I'm just saying that I have been working quite intensively in this field for 25 years now, collaborating with many physicists too). Yes, in 1976 he allows local randomness. We cannot distinguish randomness from deterministic chaos. His assumption of local causality allows him to write down a deterministic model, in which the randomness of measurement outcomes derives from randomly varying initial conditions. In mathematical terms: a Kolmogorovian probability space represents "random variables" as deterministic functions of some hidden variable omega. Richard Gill (talk) 15:51, 20 May 2022 (UTC)

You seem to have forgotten or misunderstood this one, though. He is not doing a deterministic model at all. The randomness of measurement outcomes does not derive from randomly varying initial conditions, there's no explanation whatsoever for randomness, as that is not the point of the paper. What he assumes is local causality, that the probability of an event a only depends on events in its past light cone, $\psi$, and in particular does not depend on spacelike separated events $\gamma$. In symbols, $p(a|\psi ,\gamma )=p(a|\psi )$. It's really that simple. And quantum mechanics clearly violates this assumption, as the probability $p(a|\psi )$ will be 0.5 when $\psi$ is a maximally entangled state and a is the event of Alice obtaining 0 when measuring in the computational basis, but if we let $\gamma$ be the event of Bob obtaining 0 when making his measurement, $p(a|\psi ,\gamma )$ becomes either 0 or 1. Again, this is not about determinism or hidden variables. Tercer (talk) 16:19, 20 May 2022 (UTC)

Indeed, there's no *explanation* for the randomness. Why should there be? He assumes randomness. He assumes randomness exists. In other writings (Bertlmann's socks, 1981) he explains that the randomness could come from the uncontrollable initial values of the myriad fundamental constituents of all the stuff in the source and both measuring devices and the transmission lines joining them shortly before the measurement settings are introduced. And I agree, it is as simple as you say. But I would take 1981 as the definitive mature Bell statement. Richard Gill (talk) 16:34, 20 May 2022 (UTC)

I find it rather ironic that you're quoting Bertlmann's socks. In that paper Bell is complaining about how people refuse to listen, and that determinism is not an assumption in either his theorem or EPR's argument. Instead, local causality is the whole point. It doesn't really matter what Bell said, though. It's still a mathematical theorem what he proved in 1976. That will be true regardless of anyone's opinion, including his. Tercer (talk) 18:47, 20 May 2022 (UTC)

I agree, local causality is the whole point. He proved a theorem. In 1976 and in 1981. Deterministic or random is not the issue. His definition of local causality depends on a concept of probability. The interpretation of “probability” is another question. Richard Gill (talk) 03:55, 21 May 2022 (UTC)

Yes, people come up with elaborate excuses to dismiss the fact that the quantum probabilities violate local causality. Probabilities are subjective, probabilities are not real, probabilities do not exist, only relative frequencies. Take your pick. It's just denial. Quantum probabilities are obviously objective, and they do violate local causality. Tercer (talk) 09:16, 21 May 2022 (UTC)

It’s obvious to you. Fine. Richard Gill (talk) 10:44, 21 May 2022 (UTC)

Where John Bell is all wrong

Discussion obviously going nowhere.

It has been pointed out that Bell's work is based on false concepts of the Copenhagen Interpretation, those false concepts of wave function collapse and superpositions, which do not even exist in a minimalist statistical ensemble interpretation of quantum mechanics, so Bell's work is totally wrong. This should be more clearly pointed out in Wikipedia. — Preceding unsigned comment added by 47.205.188.45 (talk) 19:06, 19 September 2022 (UTC) Can you please provide citations to reliable sources that support your claims? Tayste (edits) 02:42, 20 September 2022 (UTC) Bell clearly used wavefunction collapse, and not all interpretations recognize wavefunction collapse. 47.205.188.45 (talk) 03:17, 20 September 2022 (UTC) This is an article about Bell's Theorem. The theorem has nothing to do with wavefunction collapse. I suggest that you propose your changes here on the Talk page, as I cannot figure out what you are trying to do. Roger (talk) 03:30, 20 September 2022 (UTC) Read the second paragraph of the wikipedia Intro, it is indeed about wavefunction collapse.47.205.188.45 (talk) 04:04, 20 September 2022 (UTC) That is a paragraph about Einstein's beliefs, not Bell's. Roger (talk) 04:17, 20 September 2022 (UTC) Bell test experimenters speak in those terms, superpositions and wavefunction collapse. Neither exists according to many interpretations of quantum mechanics. This belongs in Wikipedia here.47.205.188.45 (talk) 13:14, 20 September 2022 (UTC) So go write an article on the philosophical beliefs of the Bell test experimenters. It doesn't have anything to do with this article. Roger (talk) 19:23, 20 September 2022 (UTC) The Intro leaves one hanging its full implications for the interpretation of quantum mechanics remain unresolved. There is needed an additional sentence to complete the Intro. I suggest Bell used quantum entanglement which entails superpositions and wave function collapse. In some other interpretations of quantum mechanics there is no superposition and no wavefunction collapse. This explanatory note needs be in the Intro to complete the thought of the Intro. — Preceding unsigned comment added by 47.205.188.45 (talk) 13:42, 21 September 2022 (UTC)

Deriving the expectation values

The expectation values for measuring the bell state:

$$\langle A_{0}\otimes B_{0}\rangle ={\frac {1}{\sqrt {2}}},\langle A_{0}\otimes B_{1}\rangle ={\frac {1}{\sqrt {2}}},\langle A_{1}\otimes B_{0}\rangle ={\frac {1}{\sqrt {2}}},\langle A_{1}\otimes B_{1}\rangle =-{\frac {1}{\sqrt {2}}}\,.$$

are simply given without derivation. I'd love to see a calculation of at least one, because the notation is kind of dense: The measured properties are referred to quite abstractly as $A_{0}$ etc. but the Pauli matrices later on then suggest that these are spin states. Thereafter, |01> seems to refer to a two particle system with one spin up and one spin down, but written as is it could also mean a single particle with spin down. I would really appreciate the original author expanding at least one of these given expectation values into a calculation, where you can see which part of the tensor operator acts on which part of the state. — Preceding unsigned comment added by 2A02:8109:B540:7CB:798F:B853:53BD:F07D (talk) 21:37, 8 October 2022 (UTC)

I'm a little unclear as to what you're unclear about. In the first part, the derivation of the inequality, the measured quantities are "referred to quite abstractly" because the point is that they can be anything. Then, in the second part, a particular choice of quantum state and measurement combinations is shown to violate that inequality. I tweaked the sentence flow and the notation a bit just now in a way that might (or might not) be an improvement, but really, the calculations are just matrix algebra. I'm not sure how much more we can say about that without violating the rule that Wikipedia isn't a textbook. XOR'easter (talk) 15:51, 9 October 2022 (UTC)

Many Worlds nonsense

There is a lot of nonsense in the many worlds section. Bell's theorem assumes that experiments have single outcomes. It does not say anything about many worlds. The section has a couple of references, but they don't really support the text. The text is a lot of fringe theorizing with no mainstream support. Roger (talk) 08:33, 9 October 2022 (UTC)

Note that this section is specifically about the Many-Worlds interpretation of Bell's theorem. It's not being presented as the mainstream interpretation. You clearly haven't read the references, because they do support the text. Tercer (talk) 09:54, 9 October 2022 (UTC)

Nobel Prize in lede

Sorry, I don't buy it. I don't see why we need to step outside the ordinary way of organizing articles because an incidental feature of a topic will be in the news for ~2 weeks. For comparison, quantum electrodynamics doesn't mention the 1965 Nobel for Feynman, Schwinger, and Tomonaga in its lede. Radium doesn't mention the 1911 prize to Curie. The lede of photoelectric effect mentions Einstein but not the 1921 prize. Bose–Einstein condensate doesn't get around to mentioning the Prize until the History section. There's no hard-and-fast rule about this, and in some articles the inclusion might make sense, but the intro here is already quite long. What matters more, the actual subject matter or a particular institution's delayed recognition of its importance? XOR'easter (talk) 15:11, 9 October 2022 (UTC)

On a second thought, I think you're right. I might be suffering of short-termism myself. Tercer (talk) 15:15, 9 October 2022 (UTC)

Much ado, I think. I simply thought that noting the Nobel within the context of the paragraph on Bell tests was appropriate. I am not a physicist or a Wikipedia guru. Tachyon (talk) 15:37, 9 October 2022 (UTC)

Recent changes

@HossenfelderS: rather than repeatedly reverting in the changes you would like to make to the article, why not discuss here on the talk page, and say why you think the changes are correct (with sources)? Wikipedia has sourcing requirements that can take some getting used to. Russ Woodroofe (talk) 07:36, 10 October 2022 (UTC)

Just to note the existence of a Twitter thread about this [1]. I'm encouraging them to engage here. Thanks. Mike Peel (talk) 08:18, 10 October 2022 (UTC)

It seems that the Twitter followers have arrived. And of course, they don't want to discuss anything, just edit war. Tercer (talk) 10:32, 10 October 2022 (UTC)

It looks to me like users coming from Twitter are editing in good faith, but may be unaware of the necessity of consensus and sourcing on Wikipedia. Perhaps experienced editor Tarinth will comment? Russ Woodroofe (talk) 12:07, 10 October 2022 (UTC)

I was surprised that sourced material was simply reverted based on an opinion. If there's a dispute on accuracy then it is incumbent on the editor to add a citation of sufficient notability. Alternatively, it would probably be fine to just acknowledge that there is public disagreement amongst physicists on the topic, and cite the contrasting views. This would be more productive than reverting cited material. Tarinth (talk) 12:26, 10 October 2022 (UTC)

As I wrote when originally reverting this edit: that's WP:UNDUE. We definitely shouldn't mention in the first sentence of the article a hypothesis experts considered necessary to even do science. this would be like starting mathematics articles adding the qualifier that the result only holds under the standard axioms of set theory. This is not about sourcing, it's about Hossenfelder's insistence on advertising her pet theory in the first sentence of the article. Tercer (talk) 12:46, 10 October 2022 (UTC)

I am sure that your edit was in good faith and that you are unable to see how the wording "pet theory" suggests bias. Please refer to WP:NPOV. Tarinth (talk) 12:54, 10 October 2022 (UTC)

Is there a citation contradicting that a mere "1 / 15 bits of prior correlation" is sufficient to establish a loophole, or experimental evidence showing that measurements have less correlation than that?

There appear to be multiple peer reviewed, published articles showing that correlation via either superdeterminism or retrocausality are loopholes in the current evidence supporting Bell's theorem. It seems reasonable to cite the contrasting views. LeBleu (talk) 12:51, 10 October 2022 (UTC)

I agree with LeBleu's suggestion of capturing the contrasting views with citations (ideally more than one given the apparent controversial nature). WP:UNDUE as suggested above is intended for fringe theories such as flat-Earthism, not for legitimate contrasting views amongst experts in the field.

Tarinth (talk) 12:59, 10 October 2022 (UTC)

You are missing the point. There's no controversy about the correctness of Hall's results, or that superdeterminism evades Bell's theorem. The dispute is about mentioning this in the first sentence in the article (as it was already mentioned later in the article). It's WP:UNDUE, as it's a position all experts in the field find ridiculous. Tercer (talk) 13:04, 10 October 2022 (UTC)

Hall's paper is not cited at all in the current article. But are you saying that your concern is simply that the citation and the four words it was linked to ("which fulfil measurement independence") ought to appear in the Superdeterminism section as opposed to the first paragraph? Why not move it to the section you feel more appropriate rather than revert an editor's work. That would be less disruptive. I also note that you believe that it is a "position all experts in the field find ridiculous" (emphasis on "all") -- if you believe that's the case, I suggest adding something to that effect to the section with a citation regarding this consensus. Tarinth (talk) 13:22, 10 October 2022 (UTC)

I have no problem with citing Hall's paper in the superdeterminism section. The position of experts in the field is already discussed in superdeterminism. Tercer (talk) 13:39, 10 October 2022 (UTC)

I found the citation for Wiseman & Cavalcanti in superdeterminism that provide their opinion. Extending this opinion to "all experts" is WP:OR, but perhaps you have a cituation for it? Tarinth (talk) 13:56, 10 October 2022 (UTC)

That article also quotes the opinion of Zeilinger. This article notes the opinion of Larsson, that it is necessary to even do science in the first place. What more do you want? A comprehensive survey of all experts on nonlocality? That doesn't exist. What I can assure you is that any expert will have written that it's ridiculous, or won't have wasted their time with this nonsense to start with. In any case, I don't need to provide any citation for this claim, since I'm not trying to include it in the article. Tercer (talk) 14:06, 10 October 2022 (UTC)

Got it -- so there isn't a citation, just WP:OR. Tarinth (talk) 14:13, 10 October 2022 (UTC)

It's not "just WP:OR", it's a reasonable summary of the sources that do exist on a topic of niche interest. XOR'easter (talk) 14:58, 10 October 2022 (UTC)

WP:UNDUE isn't just about outright nonsense like flat-Earthism. It also applies to things that legitimate experts have said. Bell inequalities have been discussed for decades, and plenty of peer-reviewed papers have been published on them that have sunk into obscurity and thus would be WP:UNDUE inclusion here. XOR'easter (talk) 15:08, 10 October 2022 (UTC)

I looked at the most recently reverted edit (not sure if there were other versions), but at first blush the proposed modifications strike me as sensible and broadly proportionate. If I understand the record, there is a possible alternative explanation of the Bell inequality experiments that was acknowledged decades ago, is still the subject of professional discussion, and is already reflected in an existing section of the article (i.e. "Superdeterminism"). There does not appear to be any argument that this alternative interpretation is disproven or impossible, just that there is no particular evidence to support it (and perhaps no clear way to acquire any). So, it appears to be a legitimate consideration among experts in the field, even if perhaps most consider it unlikely. In that context, including a small caveat in the article header, such as was proposed, seems appropriate to me. Of course, we could also entertain alternative ways to express that caveat in the lead, but in a technical article like this, I lean towards the inclusion of such caveats when they are backed (as in this case) by an apparent history of scientific discussion. Dragons flight (talk) 15:08, 10 October 2022 (UTC)