Does entanglement violate special relativity?

[Dmitry67]

Sorry, I meant some specific examples of entropy increase.

I should say
'Deterministic single-history theory can not, in principle, break a symmetry, for example, convert an empty space (empty everywhere) into space where somehwere there are stras, planets, while in the other parts space is void'

 

[lalbatros]

Any way to FTL with entangled system?

I believe the answer is no. But I have some doubts.

Assume we create every microsecond an entangled spin pair, for a total of 1000000 pairs.
Assume we will measure the entangled pairs when they are separated by 1 lightyear!
Assume the measuring device on side A will measure 0.1 microseconds before the measuring device on the other side B.
Assume the spin orientation on side A can be chosen freely for each individual measurement.

I have the intuiton that it would be possible to sent a few bits of information (say one bit) from A to B at a Faster Than Light speed. But this transfer would not be totally reliable. The reliability would increase with the number of pairs analysed as well as with the ingeniosity of the experimenter.

Am I dreaming?

 

[alxm]

I have the intuiton that it would be possible to sent a few bits of information (say one bit) from A to B at a Faster Than Light speed.

How did your particles get separated at FTL speed?
Or were they separated at the start? If they were, how did they become entangled? That requires them to interact with each other. How did they interact at FTL speed?

 

[DrChinese]

It does not rule out Bohmian/dBB type theories because such theories are naturally interpreted as superdeterministic (provided that the notion of "superdeterministic" is understood appropriately).

In superdeterminism, the choice of an experimenter's measurement setting is controlled by initial conditions in just such a way as to allow Bell's Inequality to be violated. In strict determinism, the experimenter's choice of measurement setting is not "free", but Bell's Inequality is not violated as a postulate of the theory. Instead, there is some other mechanism (in your case non-locality) which is responsible.

 

[lalbatros]

How did your particles get separated at FTL speed?
Or were they separated at the start? If they were, how did they become entangled? That requires them to interact with each other. How did they interact at FTL speed?

Just as in the Alain Aspect experiment, just on a longer scale for the story.
Two entangled photons traveling in opposite ways, and the experimenter waiting long enough.

 

[alxm]

Just as in the Alain Aspect experiment, just on a longer scale for the story.
Two entangled photons traveling in opposite ways, and the experimenter waiting long enough.

And how does this transmit information FTL?

 

[Demystifier]

In superdeterminism, the choice of an experimenter's measurement setting is controlled by initial conditions in just such a way as to allow Bell's Inequality to be violated. In strict determinism, the experimenter's choice of measurement setting is not "free", but Bell's Inequality is not violated as a postulate of the theory. Instead, there is some other mechanism (in your case non-locality) which is responsible.

I think I still don't understand the difference between superdeterminism and strict determinism. Can you explain the GENERAL difference between these two concepts, without referring to QM and nonlocality?

Anyway, in Bohmian mechanics both the initial conditions and nonlocal forces are responsible for the violation of Bell inequalities as predicted by standard QM. The initial conditions are important because the initial particle positions in the statistical ensemble must be distributed according to |psi|^2, which corresponds to the so-called quantum equilibrium hypothesis. If this initial condition has not been satisfied, then the violation of Bell inequalities could be even stronger than predicted by standard QM (see the papers of Valentini).

 

[DrChinese]

I think I still don't understand the difference between superdeterminism and strict determinism. Can you explain the GENERAL difference between these two concepts, without referring to QM and nonlocality?

Anyway, in Bohmian mechanics both the initial conditions and nonlocal forces are responsible for the violation of Bell inequalities as predicted by standard QM. The initial conditions are important because the initial particle positions in the statistical ensemble must be distributed according to |psi|^2, which corresponds to the so-called quantum equilibrium hypothesis. If this initial condition has not been satisfied, then the violation of Bell inequalities could be even stronger than predicted by standard QM (see the papers of Valentini).

In ordinary determinism: [initial conditions] + [physical laws] -> [observed results]. However, there is no reason that this would ever lead to violation of a Bell Inequality. Why would it? There would be no reason that a measurement setting at Alice would be somehow correlated with a result at Bob - at least certainly no more likely than a result involving unentangled particles elsewhere (let's say Chris).

The only purpose of superdeterminism (rather than determinism) is to provide a way to have Bell's Inequality violated in a local realistic manner. This has been postulated by a few authors (such as t' Hooft). From the Wikipedia article on Bell loopholes:

Superdeterminism

"Even if all experimental loopholes are closed, there is still a theoretical loophole that may allow the construction of a local realist theory that agrees with experiment. Bell's Theorem assumes that the polarizer settings can be chosen independently of any local hidden variable that determines the detection probabilities. But if both the polarizer settings and the experimental outcome are determined by a variable in their common past, the observed detection rates could be produced without information traveling faster than light (Bell, 1987a). Bell has referred to this possibility as "superdeterminism" (Bell, 1985)."

There is no other reason for superdeterminism to exist OTHER than to explain Bell's Theorem or any other portion of physics you disagree with for which there is physical proof of the side opposite to your view. There is, by definition, no possibility of disproof of this view. And further, there is no possibility - by definition - that it could ever lead to any advance in science. That is because it is strictly an AD HOC theory. It should be discussed in philosophical or religious forums rather than those devoted to physics.

 

[Demystifier]

Thank you for the clarifications, Dr. Chinese.
Anyway, I still claim that the free-will theorem, which in its refined version essentially says:
"If experimenters have free will than observed particles also have free will"
does not rule out BM, simply because in BM experimenters do not have free will.
Do you still disagree?

 

[Ilja]

That seems a weird reference for you to agree with... they claim it rules out all deterministic theories (including Bohmian/dBB type) other than superdeterministic ones. Nice paper, by the way, covers a lot of interesting ground.

It is irrelevant for dBB. They postulate some maximal speed of information transfer, which is violated on the hidden variable level in dBB.

 

[ueit]

I fully agree! Indeed, there is a theorem that confirms this (in QM):
http://xxx.lanl.gov/abs/quant-ph/0604079
http://xxx.lanl.gov/abs/0807.3286

I think that this "free-will theorem" is a good example of how untenable the position against the existence of hidden variable theories has become.

The "theorem" is nothing but a tautology (if the particles in the experimenter's brain are not described by a deterministic theory it follows that all other particles share this property). Given that there is no evidence that a brain consists of a different type of particles than the rest of the world there is nothing more to prove, no reason to appeal to EPR or even to QM. The "proof" follows from logic alone.

 

[lalbatros]

alxm,

And how does this transmit information FTL?

I don't know. I just tried to imagine some attempt. Below is a naïve idea.

I also wonder if there is a general proof that FTL is impossible by entanglement.
On the simplest communication protocol, this is obvious.
Probably this can be proved also for any sophisticated protocol and FTL would be as illusory as free energy.

Naïve idea
Let's assume side A wants to send one bit to side B.
Creating one unique entangled pair as carrying media will never allow FTL.
Any action on side A, like orientation of a polarizer (polarisation), will imply a statistical "response" on side B.
But there is no way to transmit information since the noise is maximal.

It is then natural to try doing something using many well separated entangled pairs.
Naïvely speaking, the number or repetition could improve the statistics or reduce the noise.
There are at least two (desparate) ways to try FTL then.

1) One way "communication". Side A sends a pattern of polarisations. Surely this is hopeless because the probabilities are independent in each "broadcast" from A.

2) Two way "communication". Side A and B will alternatively be emitter or receiver and adjust their polarization based on their observations. The goal will be to beat FTL from A to B. Would it be possible when B and A chose their new polarization based on their previous observations?

 

[DrChinese]

Thank you for the clarifications, Dr. Chinese.
Anyway, I still claim that the free-will theorem, which in its refined version essentially says:
"If experimenters have free will than observed particles also have free will"
does not rule out BM, simply because in BM experimenters do not have free will.
Do you still disagree?

I definitely believe an experimenter has free will to choose a measurement setting, for all intents and purposes. I realize that at some level, there could be complete determinism but that doesn't change my view as a good and useful hypothesis.

There have been a lot of papers in the past 10 years that purport to rule out BM; or alternately prove the universe is non-local; or prove contextuality; or prove MWI; etc. I like the ambition and direction of some of these papers. But no, I do not believe that BM/dBB type theories are ruled out yet.

I do believe that the BM/dBB school needs to quit asserting that there is *no* possibility of a testible difference between QM and BM. If there isn't, what is the value of BM as a theory? It simply becomes an ad hoc explanation for existing facts. The sQM school is taking the possibility of verifying or rejecting BM seriously; and I would like to see more of that attitude.

But all this is just my opinion.

 

[DrChinese]

I think that this "free-will theorem" is a good example of how untenable the position against the existence of hidden variable theories has become.

I think this is an exaggerated statement. The no-go work on hidden variables is quite extensive, although it is not yet what I would call persuasive. On the other hand, no amount of experimental effort has been able to uncover even a hint of a hidden variable anywhere (of course to a Bohmian, the hidden variables are literally everywhere). That is strongly suggestive, but still not conclusive. So I think it would be manifestly unfair to characterize a single approach (such as Conway & Kochen) as representative of the entire interpretation.

 

[nfelddav]

alxm,
I don't know. I just tried to imagine some attempt. Below is a naïve idea.

A lovely attempt, but not likely to get anywhere... on the positive side, if it did, you'd probably get a Nobel prize.
But yes, there is a theorem that says you can't use entanglement to send information: there is no way to learn anything about the polarization of the measurement.
http://en.wikipedia.org/wiki/No_communication_theorem

 

[Isaac_Newton]

If one entangled particle goes on a high speed journey and is then younger by 2 days than its entangled partner they still maintain perfectly correlated states. Although the answer is easy for me, why do you think that is? (more universes?, more crafty pilot waves?, backwards in time fairies?, superdeterminism knew it would happen? a new cat theory?)

 

[nfelddav]

I don't see why the age difference in the particles should prevent them being correlated. why should there be any change in entanglement?

 

[DrChinese]

I don't see why the age difference in the particles should prevent them being correlated. why should there be any change in entanglement?

There is in fact an interesting nuance on this particular item. No actual test - that I am aware of, at least - has been done to confirm what we assume the result will be (which is continued correlation). Now why would this matter?

If gravity is a quantum force - an open question in physics - then any particle which is "younger" must have been accelerated or otherwise been subject to a very strong gravitational field (assuming I haven't got younger and older reversed). It therefore interacted with the gravitational field by absorbing and/or emitting gravitons. Presumably, that implies the possibility of a measurement involving the associated graviton(s). Would that affect the spin of the entangled particle? If it did, it would end the correlation with the entangled twin. If it didn't, the implication would be that gravity is not a quantum force.

From Sean Carroll's blog at Discover magazine:

"Brad DeLong, in re Quantum Hyperion, wonders whether photons are really responsible for the decoherence of Saturn’s moon:

'But gravity works–presumably, at some level–by massive objects constantly bombarding each other with gravitons, so we are also averaging over all the possible states of gravitons that we are not keeping track of, aren’t we? That should cause decoherence too, shouldn’t it?'

"This is an annoyingly good question. In fact, I’m probably not giving anything away if I reveal that my esteemed co-blogger Daniel and I once tried to figure out whether or not dark matter, if it truly interacts with ordinary matter only through gravity, would be in a coherent quantum state..."

 

[nfelddav]

Ah, very interesting...
Is it necessary that if it is a quantum interaction it will impact the spin?

 

[DrChinese]

Ah, very interesting...
Is it necessary that if it is a quantum interaction it will impact the spin?

Spin, I would think so, though I have no idea how that would work. Else momentum, frequency, etc.

 

[Isaac_Newton]

I don't see why the age difference in the particles should prevent them being correlated. why should there be any change in entanglement?

Its up to you to say why entangled particles of different ages (brought about by a high speed relativity trip) remain in the exactly the same correlation even though one is two days older than the other.
What is it about entanglement that did not change?

 

[Demystifier]

The "theorem" is nothing but a tautology (if the particles in the experimenter's brain are not described by a deterministic theory it follows that all other particles share this property). Given that there is no evidence that a brain consists of a different type of particles than the rest of the world there is nothing more to prove, no reason to appeal to EPR or even to QM. The "proof" follows from logic alone.

I almost agree:
https://www.physicsforums.com/showthread.php?t=144925

In addition, I think that free will implies incompleteness of QM:
https://www.physicsforums.com/showthread.php?t=174544

 

[DrChinese]

Its up to you to say why entangled particles of different ages (brought about by a high speed relativity trip) remain in the exactly the same correlation even though one is two days older than the other.
What is it about entanglement that did not change?

Well, the obvious point is: if Alice is observed after .001 second, and Bob is observed at any time thereafter (so that Bob is older), the correlation remains (as predicted by QM). So the only issue relates to the effect, if any, of having Bob interact with a gravitational field. Does that cause decoherence and therefore end entanglement? As of today, this is entirely an open question and there is no evidence either way. Feel free to speculate...

 

[Ilja]

I do believe that the BM/dBB school needs to quit asserting that there is *no* possibility of a testible difference between QM and BM. If there isn't, what is the value of BM as a theory? It simply becomes an ad hoc explanation for existing facts. The sQM school is taking the possibility of verifying or rejecting BM seriously; and I would like to see more of that attitude.

First, that is not a question about asserting - or there is equivalence, or there is not. Once there is a theorem about equivalence in quantum equilibrium, there is no possibility, if we are trapped in equilibrium. Very sorry, but there is no choice.

If there isn't, the value of BM as an empirical theory is the same as of QM. Once you do not seem to propose the throw away QM, I see no reason to throw away BM.

There are important metaphysical differences, in particular there is no measurement problem, no nonunitary evolution of the wave function, and, very important, the whole measurement theory postulated in QM can be derived from something much more simple. Given the simplicity of this derivation, BM should be preferred by the usual standards of preference for the simpler theory.

My personal interest is to understand how the universe works. It is not to find some theory which differs in some prediction from the standard one - if it happens that such a modification allows for a much simpler understanding of our universe - fine, I will embrace it. (For example, my theory of gravity contains some additional terms, see ilja-schmelzer.de/glet/, but there was a good reason for introducing them.) But I see absolutely no reason to change successful theories.

 

[Demystifier]

There are important metaphysical differences, in particular there is no measurement problem, no nonunitary evolution of the wave function, and, very important, the whole measurement theory postulated in QM can be derived from something much more simple. Given the simplicity of this derivation, BM should be preferred by the usual standards of preference for the simpler theory.

My personal interest is to understand how the universe works. It is not to find some theory which differs in some prediction from the standard one - if it happens that such a modification allows for a much simpler understanding of our universe - fine, I will embrace it.

I cannot agree more.

 

[debra]

I cannot agree more.

I cannot agree less! We are discussing "Does entanglement violate special relativity" not pushing Bohm - you don't seem interested.

 

[Demystifier]

I cannot agree less! We are discussing "Does entanglement violate special relativity" not pushing Bohm - you don't seem interested.

That's fair. However, I think that the question "Does entanglement violate special relativity?" cannot be answered irrespective of the interpretation one works with. Thus, it is interesting to see how different interpretation answer this question. One of the interpretations is the Bohmian one, and that's how we arrived at it.

 

[Ilja]

I cannot agree less! We are discussing "Does entanglement violate special relativity" not pushing Bohm - you don't seem interested.

The questions are closely related. First, it seems unjustified to take some part (entanglement) out of quantum theory and to ask separately if it "violates relativity". It is quantum theory as a whole which violates (or not) relativity.

But this question cannot be discussed without specifying particular interpretations. For relativity, you have to distinguish two interpretations: One where relativistic symmetry is about observables, one where it is about reality. And for quantum theory, you need such differentiation as well.

In the pilot wave interpretation special relativity (its fundamental interpretation) is violated, its observable variant not. In general, the observable variant of relativity is not violated by quantum theories. There are non-realistic interpretations of quantum theory, and for these interpretations the fundamental, realistic interpretation of relativity simply makes no sense. For realistic interpretations, realistic relativity is violated.

The exception is the many words interpretation (no typo), which claims to be realistic and claims to be in agreement with relativity. All this, in the modern variants, without formulas defining their basic objects or probabilities or whatever else you would like to compute.

 

[Dmitry67]

It is not SR but CAUSALITY which is violated
But causality is somehow volated in the closed timelike loops, so GR also violates causality.

 

[Ilja]

It is not SR but CAUSALITY which is violated
But causality is somehow volated in the closed timelike loops, so GR also violates causality.

Sorry, in the quantum context there is no violation of causality. Again, pilot wave interpretation is the explicit counterexample. One with formulas and theorems.

Handwaving I leave to MWI.

 

[Dmitry67]

I mean the topic subject is incorrect.
When people say 'entanglement violates SR' they mean FTL -> causality problems.
But in fact (I agree with you) there is no such violation, even the entanglement is discussed in that context over and over again

P.S.
I understand that you hate MWI, but just curious, what interpretation do you like?

 

[debra]

Lets set a simple scenario:
Lets first of all make an empty 3D space about the size of the Universe. Into which we will place two particles. We want to study SR and entanglement using real physical classical objects.
Is that OK or is it illogical in some way?

 

[nikman]

Lets set a simple scenario:
Lets first of all make an empty 3D space about the size of the Universe. Into which we will place two particles. We want to study SR and entanglement using real physical classical objects.
Is that OK or is it illogical in some way?

Someone like (say) Max Tegmark might say it's not coherent. He maintains that physical existence began with (1) a specific, defined set of initial conditions and (2) something like the Schrödinger equation to determine future evolution (within the probabilities associated with both the SE and classical complexity).

So, a thought-experiment of your kind posits a kind of scale separation and may be a simulation fallacy. It's a computationalist abstraction. Can you really separate those two particles and their behavior from the complete history of the universe? And how would you incorporate that into the model?

 

[nikman]

Also, maybe a more practical consideration: GR takes the Universe to be in part a function of the total matter-energy it contains. Could you have "an empty 3D space about the size of the Universe" if the Universe is only two particles?

So you scale down. Make it a Universe of a size appropriate to a Universe of only two particles. Would that be equivalent?

 

[DrChinese]

He maintains that physical existence began with (1) a specific, defined set of initial conditions and (2) something like the Schrödinger equation to determine future evolution...

...Can you really separate those two particles and their behavior from the complete history of the universe?

Max Tegmark is a Many Worlder, I think. He has written articles on parallel universes. Addressing your comments:

1. From a factual viewpoint, this does not appear to be accurate based on what we currently know. There are parts of the universe that have never been in causal contact with us. And the part that has is constantly changing - i.e. is not static.

2. Further, there is a powerful virtual particle field that we are immersed in. Clearly, these particles are affecting us. Are virtual particles evolving according to these same deterministic laws?

I don't think you are saying that you are advocating a deterministic viewpoint, but I just wanted to get clarification.