Entanglement and FTL signaling in professional scientific literature

[Demystifier]

For me signal and influence is a synonym.

For you, not for the professional scientific literature.

This is at least standard in all treatments of relativistic QFT in the HEP community.

That's a wrong community for such questions. The relevant community is the community professionally working on quantum foundations (Bell theorem etc). I know that the HEP community thinks that they have nothing to learn from quantum foundations community, but they are wrong.

 

[vanhees71]

Then, please, define what you precisely understand under "influence" vs. "signal" or point to the "professional scientific literature". I always thought, mathematical facts do not depend on the community using this math ;-)).

 

[martinbn]

Sure, I have written it myself, but such papers emphasize that it is not in contradiction with "the mathematical facts of microcausal relativistic QFTs".

They can emphasize whatever they want to, but it is inconsistent. You cannot say that two space-like events are in cause effect relation and that your interpretation doesn't contradict relativity.

The idea of such papers is that relativistic QFT is right but incomplete. Completeness is often an additional tacit assumption in standard QFT literature, but completeness is not a mathematical fact.

Incompleteness will not save you, because the inconsistency is with what is already there. Adding more to the theory to complete it will not help.

 

[martinbn]

That's true, but signal is the key word. Your original statement was wrong because it didn't contain the word signal. Standard QFT forbids FTL propagation of signals, but it doesn't forbid FTL propagation of influences. I know that you don't distinguish signal from influence, but professional scientific literature does.

Different terminology makes no difference. If your theory says that A is the cause of B, and A and B are space-like separated, then you are saying that relativity is way off. Which is by itself ok, the problem is that it is just wishful thinking with no shred of evidence to support it.

 

[PeroK]

If by nonlocal influences you mean faster-than-light causal influences then any interpretation which assumes them is in plain contradiction with the mathematical facts of microcausal relativistic QFTs!

I don't see how you can appeal to mathematics as "facts". All mathematics can be is consistent. There should be nothing mathematically inconsistent about any theory, whether it's right or wrong.

That's a wrong community for such questions. The relevant community is the community professionally working on quantum foundations (Bell theorem etc). I know that the HEP community thinks that they have nothing to learn from quantum foundations community, but they are wrong.

They only have something to learn, IMO, if the relativistic QFT championed by @vanhees71 turns out to be incomplete! I don't share his evangelical confidence in QFT, but neither do I share the philosophical doubts that seem to trouble many.

 

[vanhees71]

My argument simply is that microcausal relativistic QFT excludes by construction faster-than-light influences, i.e., within this theory it's a "mathematical fact" that space-like separated events cannot be causally connected. So far relativistic QFT is utmost successful in describing all known matter and its interactions. This is a good reason to "believe" in it as a good physical theory. This doesn't mean that it's complete, which for sure it is not, but indeed not for philosophical but hard reasons, i.e., it's inability to satisfactorily describe also the gravitational interaction.

 

[Vanadium 50]

A correlation does not necessarily imply a causal connection/signal/influence between two events

This is important. (A fun example is below)

Somehow the combination of the ideas that "QM is fundamentally probabilistic" and "you can have a state consisting of multiple particles" confuses and/or bothers people. They true and shoehorn ithis into the idea that each particle is independent, except for a spooky causal connection that causes a measurement of one to influence the other.

This is the same sort of thinking that is confusing about virtual particles: the quantum world is classical except for these little unseen virtual particles bouncing around.

 

[PeterDonis]

By construction microcausal relativistic QFT exclude causal connections between space-like separated events.

This statement is too strong. The condition you call "microcausality" says that spacelike separated measurements must commute. That is not the same as saying they cannot be causally connected. A causal connection that is consistent with them commuting is still possible.

 

[PeterDonis]

You cannot say that two space-like events are in cause effect relation and that your interpretation doesn't contradict relativity.

Yes, you can, as long as whatever cause effect relation you are postulating is consistent with the spacelike separated measurements commuting. See my post #113 in response to @vanhees71 just now.

 

[vanhees71]

This statement is too strong. The condition you call "microcausality" says that spacelike separated measurements must commute. That is not the same as saying they cannot be causally connected. A causal connection that is consistent with them commuting is still possible.

Can you give an example for such a scenario? The microcausality principle says that all local operators that represent observables commute at space like separation of their arguments, particularly the Hamilton density commutes with such an operator at space-like-separated arguments!

 

[martinbn]

Yes, you can, as long as whatever cause effect relation you are postulating is consistent with the spacelike separated measurements commuting. See my post #113 in response to @vanhees71 just now.

And what kind of a cause effect relation is consistent with space-like separated measurements commuting?

 

[PeterDonis]

Can you give an example for such a scenario?

The very scenario we've been discussing: spacelike separated Bell-type measurements on a pair of entangled particles. The measurements commute, as QFT says they must, but the correlations violate the Bell inequalities, so there is still some kind of connection between them that cannot be explained classically.

The microcausality principle says that all local operators that represent observables commute at space like separation of their arguments, particularly the Hamilton density commutes with such an operator at space-like-separated arguments!

Yes. So what?

what kind of a cause effect relation is consistent with space-like separated measurements commuting?

The very one we've been discussing. See above.

 

[vanhees71]

The very scenario we've been discussing: spacelike separated Bell-type measurements on a pair of entangled particles. The measurements commute, as QFT says they must, but the correlations violate the Bell inequalities, so there is still some kind of connection between them that cannot be explained classically.

Of course, they cannot be explained classically. We talk about relativistic QUANTUM field theory. The correlations are due to the preparation of the photon pair as an entangled quantum state, i.e., for the explanation of the correlations you don't need to invoke causality between the measurement on photon at place A and that on the photon at place B but just to the causal connection due to their common creation, but that's my argument all the time!

Yes. So what?

It rules out that there are causal connections between space-like separated observations.

The very one we've been discussing. See above.

 

[martinbn]

The very scenario we've been discussing: spacelike separated Bell-type measurements on a pair of entangled particles. The measurements commute, as QFT says they must, but the correlations violate the Bell inequalities, so there is still some kind of connection between them that cannot be explained classically.

Don't get it! Why does this imply that there is a cause effect relation?

 

[DrChinese]

Surely this is interpretation dependent and therefore not obvious. Many interpretations would not frame quantum correlations as evidence of nonlocal influences.
https://www.webofstories.com/play/murray.gell-mann/165

I watched the clip (4 minutes) and loved hearing Gell-Mann speak on this. Thanks for that.

However, I certainly don't think his attack (in the clip) on the word "nonlocal" represents the majority view of the community given its near constant use (at least 3188 papers have this in their titles in the past 10 years, for example). Apparently many others find the word "nonlocal" meaningful and representative. Gell-Mann places some of the blame for usage of that word at the feet of John Bell, whom did tend to believe something ("influence", "synchronization", whatever, but not a signal) nonlocal was occurring.

Gell-Mann dismisses Bell's essential point by claiming that it is explained by the "decoherent [sometimes consistent] histories" interpretation. So I guess in that respect, your point about my description being "interpretation dependent and therefore not obvious" has some merit. I just don't see how Alice's selection of a measurement basis *here* - which casts distant Bob's outcome into a precisely synchronized result *there* - should not be described as a (quantum) nonlocal influence. It doesn't happen by coincidence, as Bell showed us...

Bell test results ALWAYS appear to be sharply defined by either Alice's choice of measurement basis, [and/]or of Bob's, along with a random element for an outcome consistent with both measurements per the quantum mechanical prediction (either a perfect correlation, or a statistical one). Since Alice and Bob are spacelike and causally separated, the context is nonlocal - and there is no escape for that (even in decoherent histories). I guess we could agree that the context is nonlocal, but argue that there is no "influence" which is FTL. And that description seems OK to me as well.

 

[PeterDonis]

Why does this imply that there is a cause effect relation?

It implies that there is some kind of connection. Whether you want to call it a "cause effect relation" is a matter of words, not physics.

 

[PeterDonis]

just to the causal connection due to their common creation

It rules out that there are causal connections between space-like separated observations.

Aren't you contradicting yourself here? You are saying the connection due to the entanglement produced by the common preparation process is a causal connection.

 

[martinbn]

It implies that there is some kind of connection. Whether you want to call it a "cause effect relation" is a matter of words, not physics.

But if someone wants to call it cause and effect relation, then he needs to have some reason for that. I don't see any.

 

[martinbn]

...
I just don't see how Alice's selection of a measurement basis *here* - which casts distant Bob's outcome into a precisely synchronized result *there* - should not be described as a (quantum) nonlocal influence.
...

It is not about what you call it. The problem is that you say that what Alice does affects (changing the word to casts doesn't make a difference) what Bob sees. But that is not in the theory nor in the experiment. Think about it this way. Everything is symmetric about Alice and Bob. Your phrasing in English is not. So it cannot be accurate.

 

[PeterDonis]

if someone wants to call it cause and effect relation, then he needs to have some reason for that.

The obvious reason is that the correlations between the measurement results need to have some kind of cause.

 

[PeterDonis]

Everything is symmetric about Alice and Bob. Your phrasing in English is not.

This is a good point, and corresponds to the property I have pointed out, that the measurement events commute.

 

[Demystifier]

They can emphasize whatever they want to, but it is inconsistent. You cannot say that two space-like events are in cause effect relation and that your interpretation doesn't contradict relativity.

Incompleteness will not save you, because the inconsistency is with what is already there. Adding more to the theory to complete it will not help.

Your statements cannot be found in professional scientific literature, mine can. If you are so self-confident why don't you write a paper about it and publish it? I know why, the Dunning-Kruger effect.
https://onlinepethealth.com/the-dunning-kruger-effect/

 

[martinbn]

The obvious reason is that the correlations between the measurement results need to have some kind of cause.

But not necessarily that one of the measurements is the cause for the other. All I am saying is that such statements needs justification or at least acknowledgment that they depend on an interpretation because they don't follow from QM and the experiments. May be I misunderstand @DrChinese, but that is how I read his wording.

 

[vanhees71]

But not necessarily that one of the measurements is the cause for the other. All I am saying is that such statements needs justification or at least acknowledgment that they depend on an interpretation because they don't follow from QM and the experiments. May be I misunderstand @DrChinese, but that is how I read his wording.

Exactly!

I don't understand, why it is so difficult to understand that correlations do not necessarily imply causal influences. One problem is indeed the ambiguous use of the word "local" or "nonlocal", and that's indeed also due to Bell, who used the word in a different meaning than he should have been used to as an expert in relativistic QFT.

The microcausality principle, applied to the entire experiment, excludes the possibility that the observed correlations in Bell experiments are due to a mutual causal influence of the local space-like separated measurements (detection of the two photons, which have been prepared in an entangled state). The only conclusion is that the correlations are due to the preparation in this entangled state, i.e., although the single-photon observables are "maximally uncertain" (in the sense of the maximum-entropy principle) when the preparation is in one of the four possible Bell states and thus as much indetermined as they can be, the strong correlations of all kinds of possible coincidence measurements (including the ones demonstrating the violation of Bell's inequality and thus rejecting the hypothesis of any "local realistic HV theory" in the sense of Bell) are due to the preparation and not due to causal influences of one of the measurements on the other.

 

[vanhees71]

The obvious reason is that the correlations between the measurement results need to have some kind of cause.

Yes, indeed. The cause is the preparation in an entangled state.

Aren't you contradicting yourself here? You are saying the connection due to the entanglement produced by the common preparation process is a causal connection.

The measurements are in the forward lightcone of the "preparation event". So there is no contradiction to causality.

The measurements on the far-distant photons with space-like separated registration of the photons cannot have a causal influence on each other. This makes the standard microcausal local relativistic QFT consistent with the causal structure implied by Minkowski space as the special-relativistic space-time model.

 

[gentzen]

Your statements cannot be found in professional scientific literature, mine can.

There is an interesting substantive discussion between vanhees71 and other experts, and you have to go into your petty fights with martinbn. I find it fine if he wants to support vanhees71 in this discussion. martinbn is not the one to annoy "me" (and maybe others too) by endlessly insisting on "non-established" ways to use certain words.

Having joined that discussion now, here is my opinion: In this specific discussion, Lord Jestocost's approach to cite specific usages and clarification of words in the existing literature makes most sense to me.

From “Quantum entanglement” by Ryszard Horodecki, ... Published 17 June 2009):

“Nature on its fundamental level offers us a new kind of statistical non-message-bearing correlations, which are encoded in the quantum description of states of compound systems via entanglement. They are ‘nonlocal’²² in the sense, ...

Generally speaking, quantum compound systems can reveal holistic nonsignaling effects even if their subsystems are spatially separated by macroscopic distances. ...

²² The term nonlocality is somewhat misleading. In fact ...”

This both acknowledges that 'nonlocality' is the accepted word to talk about that stuff in the literature, offers constructive criticism against the usage of that word, and also hints at an explanation ("holistic nonsignaling effects") what is going on.

So who could I cite, to offer something similar? Being a fan of Mermin, I go with:

How clearly and convincingly to exorcise nonlocality from the foundations of physics in spite of the violations of Bell inequalities. Nonlocality has been egregiously oversold. On the other hand, those who briskly dismiss it as a naive error are evading a direct confrontation with one of the central peculiarities of quantum physics. I would put the issue like this: what can one legitimately require of an explanation of correlations between the outcomes of independently selected tests performed on systems that no longer interact?

I watched the clip (4 minutes) and loved hearing Gell-Mann speak on this. Thanks for that.

However, I certainly don't think his attack (in the clip) on the word "nonlocal" represents the majority view of the community ... Gell-Mann places some of the blame for usage of that word at the feet of John Bell, whom did tend to believe something ("influence", "synchronization", whatever, but not a signal) nonlocal was occurring.

My overall impression is that Gell-Mann's insights often share the fate of Heisenberg's insights and are being ignored or dismissed without sufficient reason. (Gell-Mann had a certain kind of arrogance and direct openness which didn't seem to go down too well for many people.) Even so Griffiths version of "locality" in the sense of the consistent histories interpretation is too weak for me, and I know that Gell-Mann bases his critisicm of the word "nonlocality" on that interpretation here, the fact that Gell-Mann dislikes the word "nonlocality" gives me more pause than anything else whether it is really appropriate.

 

[martinbn]

...and you have to go into your petty fights with martinbn. I find it fine if he wants to support vanhees71 in this discussion. ...

I don't think he meant it the way it looks if you just read what he wrote.

ps Are you suggesting we often get into petty fights? That is probably my fault. I will tone it down.

 

[PeroK]

I know that Gell-Mann bases his critisicm of the word "nonlocality" on that interpretation here, the fact that Gell-Mann dislikes the word "nonlocality" gives me more pause than anything else whether it is really appropriate.

We could call it the QC factor instead. What would that change?

Talk about the tyranny of words!

 

[vanhees71]

Well, but Heisenberg is one of the main culprits adding to the confusion about interpretations of QM from the very first moment on. Bohr corrected a lot, but unfortunately he's also not known for clear writing either.

It is also a bit ridiculous to claim that my view were not present in the "professional scientific literature", to which I count both peer-reviewed articles in scientific (and not philosophical) journals as well as textbooks written by renowned scientists. I think the best sources about microcausality and its implications are Weinberg's and Haag's QFT books. I'd count them to the "professional scientific literature"...

 

[gentzen]

The only conclusion is that the correlations are due to the preparation in this entangled state, i.e., although the single-photon observables are "maximally uncertain" (in the sense of the maximum-entropy principle) when the preparation is in one of the four possible Bell states and thus as much indetermined as they can be, the strong correlations of all kinds of possible coincidence measurements (including the ones demonstrating the violation of Bell's inequality and thus rejecting the hypothesis of any "local realistic HV theory" in the sense of Bell) are due to the preparation and not due to causal influences of one of the measurements on the other.

Wow, that is a really long sentence. I am not so sure whether "The only conclusion ..." is really without reasonable alternatives. Depending on the protocol, especially if there is communication going on after the actual measurements, the possible stories where the correlations came from can also shift to a later point in time. You get this typical quantum complementarity that it is hard to tell which part of the influence happened due to the preparation before the measurements, and which part due to the communication after the measurements. But there is one thing we can be sure of: there is never any FTL signaling going on. And every influence that happens in the end has some non-FTL communication channel for which one can construct a suitable story how it enabled the information exchange.

 

[vanhees71]

Of course, to get the correlations you have to compare the measurement protocols with the accurate time stamps to ensure to investigate the correlations between the photons in each entangled photon pair. I'm not an expert about all kinds of "loopholes" being discussed concerning such experiments, but I thought that there is some consensus in the quantum-optics community that these loopholes all have been ruled out.

 

[Demystifier]

and you have to go into your petty fights with martinbn

I think I found a way how to avoid this in the future. We shall see if this will work.

 

[gentzen]

It is also a bit ridiculous to claim that my view were not present in the "professional scientific literature",

Agreed, it was a bit ridiculous already before it "escalated" ...

Well, but Heisenberg is one of the main culprits adding to the confusion ...

Heisenberg was never afraid to make mistakes. I see him as a communicator, and communication can't even start if you insist that every thing you say always has to be perfectly right.

Bohr corrected a lot, but unfortunately he's also not known for clear writing either.

Bohr is nearly unreadable, precisely because he tried to never say anything wrong. He could correct Heisenberg, because Heisenberg had the courage to simply say what he thought, and be willing to be corrected later. It is true that Heisenberg's initial misunderstanding got amplified and perpetrated forever, but I have no idea who is to blame for that.

Of course, to get the correlations you have to compare the measurement protocols ... I'm not an expert about all kinds of "loopholes" being discussed concerning such experiments, ...

I am not talking about "loopholes" here, and also not about "theories". I am talking about things like how Alice and Bob use their freedom to make different measurements. If they agree on some sequence of measurements in advance, then all the signaling happens due to the preparation. If they agree to do independent totally random measurements, then the signaling happens due to their communication later which measurements they actually did. And if they ... then ... and so on ...

It is never the individual run in such an experiment that can be interpreted and gets a story, but only collections of runs of the experiment as a whole including their followed protocol.

 

[vanhees71]

With "preparation" I meant the production of the entangled photon pair only. The choice of which observable A and B measure on their photons is part of the "measurement" (that's of course just semantics for clarity in a way). I thought one of the loopholes is exactly the one you are bringing up here, i.e., that the correlations are due to some causation in the choice of the measured observables. AFAIK this loophole has been ruled out by performing experiments, where the choice of observables was made locally and random in some way such that one can exclude this type of causal connection between the measurements also by making these local random choices space-like separated, and still the expected correlations and violations of Bell's inequality have been observed in accordance with the predictions of QT. So I think this kind of loophole can be considered to ruled out (?).

I also consider the measurements as finished and irreversible when the results of the local measurements including an accurate timestamp are stored on some storage device. Then these measurement protocols can be evaluated at any time later without changing the empirical facts stored in these protocols. Of course you can "postselect" subensembles by just using the measurement protocols like, e.g., in the "quantum eraser experiments", where you can erase which-way information by simply postselcting appropriate subensembles, which then show double-slit interferences fringes.

 

[PeterDonis]

But not necessarily that one of the measurements is the cause for the other.

If the only way to have a causal connection is to have it be one-way, then yes, there is no way for the two measurements to be causally connected, since the relationship between the measurements is symmetric but a one-way causal connection is not. But is it really necessary that every causal connectio must be one-way? The causal connections we are used to are one-way, but is that really a hard requirement for all causal connections?