I Question about an entanglement paper

[exmarine]

I am re-reading paper(s) by Huw Price and Ken Wharton that present an interesting explanation of entanglement. I will see if I can paste a link, otherwise one can find them on arXiv: one is for example, “Disentangling the Quantum World”. The idea is that there could be retro-casual communication between the particles, an idea first suggested by Costa de Beauregard in the late 1940s (a student of de Broglie!). Basically, a quantum particle could “know” in advance what measurement will be made in its future.

They claim that explanation is compatible with the prohibition of faster-than-light communication in Special Relativity. My question is very specifically limited to that claim. Wouldn’t such a retro-casual explanation require super-luminal retro-communication? If a particle such as a photon knows in advance what type of measurement it will eventually encounter, then that seems to mean that information had to travel back in time faster-than-light. Or else that information had to start the trip even before the photon came into existence? It does not seem to me that their explanation is compatible with Special Relativity.

 

[Dale]

They claim that explanation is compatible with the prohibition of faster-than-light communication in Special Relativity. My question is very specifically limited to that claim. Wouldn’t such a retro-casual explanation require super-luminal retro-communication?

I am sure they are aware of that, so what was their justification in the paper?

It might be good to link to the specific paper and identify the paragraph in question.

 

[exmarine]

I don’t see where they explicitly address the speed that would be required for their retro-causal communication. How about I cut and paste a few paragraphs from one of their papers? 1510.06712.pdf (5 Nov 2015)

A live alternative to quantum spooks

Bottom page 1, top page 2:

Hanson’s experiment does, as claimed, make a convincing case for closing
some of the best-known loopholes in the case for action-at-a-distance.
However, it is much further from settling the case against Einstein than
these responses suggest. There’s a large and promising loophole that simply
doesn’t show up on most commentators’ radar – or most physicists’ radar,
for that matter.3 When it gets noticed at all, it tends to get confused for
something else. So the fact that there’s still a viable alternative to actionat-
a-distance – arguably, a much more attractive alternative, and certainly
one that is untouched by the results from Hanson and his team – remains a
well-hidden secret.
This invisible loophole falls into a well-known category. The argument
for action-at-a-distance assumes that quantum particles don’t know what
measurements they are going to encounter in the future. A little more
technically, it assumes that the state of a particle before a measurement
is independent of the particular setting chosen for that measurement (the
choice whether to measure position or momentum, say). This sounds innocuous
enough. How could the particle know about that, before it reaches
the measurement device? But innocuous or not, it is crucial. Without this
independence assumption, the argument for action-at-a-distance just doesn’t
go through.
The invisible loophole rejects this independence assumption, but it is
confused for and obscured by another proposal for doing the same thing.
This better-known cousin is a well-recognised but deservedly unpopular little
loophole called superdeterminism. To see how superdeterminism proposes to
reject the independence assumption – and how there’s a much more attractive
way of doing the same thing – let’s take a detour via medieval theology.
(Superdeterminism has ancient ancestors.)

Bottom page 5:

As Costa de Beauregard also pointed out, an attractive feature of his
proposal is that because both arms of the zigzag lie in or on the lightcones,
it is immediately congenial to special relativity, in a way in which direct
action-at-a-distance is not. This also distinguishes retrocausality from other
explanations that hope to fill the gap between entangled particles, perhaps
with continuous processes traveling faster than light.

Top page 6:

Many commentators think of the tension between action-at-a-distance
and special relativity as the source of some of deepest puzzles about quantum
theory. As David Albert and Rivka Galchen put it, in a piece in Scien-
tific American: “Quantum mechanics has upended many an intuition, but
none deeper than [locality]. And this particular upending carries with it a
threat, as yet unresolved, to special relativity—a foundation stone of our
21st-century physics.”8 But if we keep our eyes on the fact the zigzag retrocausal
proposal avoids this tension completely, it is easy to see that it isn’t
just another form of spooky action-at-a-distance.

 

[DarMM]

The retrocausal signals are confined to the past light cone of the emission event, thus they never traverse spacelike distances.

 

[Dale]

because both arms of the zigzag lie in or on the lightcones,
it is immediately congenial to special relativity, in a way in which direct
action-at-a-distance is not. This also distinguishes retrocausality from other
explanations that hope to fill the gap between entangled particles, perhaps
with continuous processes traveling faster than light.

So it seems like they are still forbidding any causal connection outside of the light cone, but they are allowing effects to precede causes as long as they are still within the light cone. It is an unusual approach, but it does seem to respect the causal structure of SR. Sort of.

 

[PeterDonis]

they are allowing effects to precede causes as long as they are still within the light cone. It is an unusual approach, but it does seem to respect the causal structure of SR. Sort of.

There is actually no issue with SR as long as the "effect" and the "cause" commute--in other words, that what happens at each of those events does not depend on the order in which they occur. Since this is true for any pair of measurements on entangled particles in QM, there's no problem with having such a pair of measurements be timelike separated and having the one you call the "effect" precede the one you call the "cause".

 

[DarMM]

So it seems like they are still forbidding any causal connection outside of the light cone, but they are allowing effects to precede causes as long as they are still within the light cone. It is an unusual approach, but it does seem to respect the causal structure of SR. Sort of.

Yeah retrocausality is one of the five classes of mechanical explanations of entanglement permitted by Bell's theorem. They're all pretty odd.

 

[Dale]

Yeah retrocausality is one of the five classes of mechanical explanations of entanglement permitted by Bell's theorem. They're all pretty odd.

What are the other four?

 

[exmarine]

Actually, I don’t think I understand reverse time light cones. I guess it has to be like a movie going exactly in reverse, one frame at a time, back down the world lines to the apex of the forward light cone from the emission event. Does the retro-signal have to get to the emission event before the emission? Yes, since “we” cannot actually go back in time, the detection event boundary conditions from each branch are “already in” the other’s branch, and that is super-luminal. Or if the detection boundary condition message does not start until the first detection event, and then travels back to the emission and forward along the other branch, that is also super-luminal.

It seems that retro-time boundary condition information movement requires super-luminal movement regardless of how one imagines it happens. The only way I can see where it might not is if both of the final detection boundary conditions start their journeys long before even the emission event. Spooky^2.

But wait: there is no proper time lapse for a photon? Maybe that plays a role?

 

[DarMM]

What are the other four?

They are:

1. Multiple Worlds
2. Acausal. There is simply no dynamic account of what occurs. The 4D history as a whole obeys a certain constraint, but that constraint results in a set of events that can't be broken down in a 3+1D way, i.e. as initial events/conditions on a 3D surface which then evolves under a PDE or similar
3. Nonlocal interaction, i.e. objects affecting each other at spacelike distances. Ultimately this would mean Relativity is wrong
4. Spacetime is highly topologically nontrivial, e.g. there are microscopic wormholes everywhere. This essentially invalidates treating any small system as if Special Relativity were true, you must always use General Relativity for objects beneath a certain size due to the topologically nontrivial nature of spacetime

Note that some don't think any of the above or retrocausality is correct and would say entanglement does not have a mathematical explanation.

 

[kurt101]

Nonlocal interaction, i.e. objects affecting each other at spacelike distances. Ultimately this would mean Relativity is wrong

Can you explain why non-local interaction means relativity is wrong? I have seen this said by various moderators and I don't understand the conflict with relativity. For example in another thread PeterDonis said:

#1 is incorrect because of relativity: if the two spin measurements are spacelike separated, their time ordering is not invariant, so there is no invariant fact of which one was first. The only invariant is that the measurement results are independent of which one is first. Which means there is no valid way to interpret either one as triggering a change in the other.

If the measurement of one entangled particle instantly affects (in a universal clock sense) the other entangled particle, how would we even know it broke a rule of relativity since the states of each particle are effectively hidden until the measurement?

 

[PeterDonis]

Nonlocal interaction, i.e. objects affecting each other at spacelike distances. Ultimately this would mean Relativity is wrong

That's not quite right. Relativity (more precisely, quantum field theory, i.e., quantum mechanics + relativity) does not say that there cannot be any interaction between spacelike separated events. It only says that any such interaction must commute, i.e., what happens at the events cannot depend on the order in which they happen.

 

[PeterDonis]

instantly affects (in a universal clock sense)

There is no universal clock. That's what relativity says.

Also see my response to @DarMM just now.

 

[kurt101]

There is no universal clock. That's what relativity says.

Relativity applies when events are described relative to the speed of light, but this interpretation requires instant action (faster than light). How else can you describe that measuring one entangled particle instantly affects the state of the other entangled particle without invoking the concept of a universal clock?

Relativity (more precisely, quantum field theory, i.e., quantum mechanics + relativity) does not say that there cannot be any interaction between spacelike separated events. It only says that any such interaction must commute, i.e., what happens at the events cannot depend on the order in which they happen.

I agree that this interpretation violates the statement "interaction must commute, i.e., what happens at the events cannot depend on the order in which they happen". Assuming this is an accurate law of relativity then I am forced to say that DarMM is correct in saying "
Ultimately this would mean Relativity is wrong". That is all fine, but this violation of relativity does not affect the math for normal events that we calculate or measure; and so there is no good reason to discount this interpretation on the basis that it violates relativity.

And ultimately what I am really trying to understand is why would anyone discount this interpretation? And so far the answer because it violates relativity just seems like a cop-out to me.

 

[Dale]

Relativity applies when events are described relative to the speed of light

As far as we know relativity always applies. Sometimes other non-relativistic theories are good approximations to relativity, but relativity always applies.

 

[weirdoguy]

Relativity applies when events are described relative to the speed of light

I'm sorry, but that does not make any sense.

 

[kurt101]

I'm sorry, but that does not make any sense.

The special theory of relativity is based on the idea that the speed of light is constant in all reference frames and based on this you get proper time. If you are discussing events that are caused by faster than the speed of light action, then this theory and this definition of time can not apply. That is what I am trying to convey in this sentence.

 

[PAllen]

The special theory of relativity is based on the idea that the speed of light is constant in all reference frames and based on this you get proper time. If you are discussing events that are caused by faster than the speed of light action, then this theory and this definition of time can not apply. That is what I am trying to convey in this sentence.

This is not true. There are several models extending conventional SR to include FTL phenomena that maintain frame variance of light speed and the same definition of proper time.

 

[kurt101]

This is not true. There are several models extending conventional SR to include FTL phenomena that maintain frame variance of light speed and the same definition of proper time.

By FTL, I mean instant, which is why I used the term universal clock. How you measure one entangled photon impacts the other entangled photon instantly. I am trying to understand how the theory of relativity discredits this interpretation.

 

[PAllen]

By FTL, I mean instant, which is why I used the term universal clock. How you measure one entangled photon impacts the other entangled photon instantly. I am trying to understand how the theory of relativity discredits this interpretation.

‘Instant’ raises no separate issues, because any ftl influence is instant or retrocausal for some observers. Note that instant is really not meaningful. An entanglement situation will look retrocausal in a different frame. None of this need affect invariance of c or or the definition of proper time. Just look at something as simple as tachyons, a simple extension of conventional SR. A tachyon message trivially may be instant or retrocausal depending on the frame viewing it.

 

[PeterDonis]

Relativity applies when events are described relative to the speed of light

No, relativity applies period. More precisely, quantum field theory (QM + relativity) applies period. There is no way to avoid relativity by not "describing events relative to the speed of light".

How else can you describe that measuring one entangled particle instantly affects the state of the other entangled particle without invoking the concept of a universal clock?

By realizing that the statement "measuring one entangled particle instantly affects the state of the other entangled particle" is simply wrong. Neither particle even has a state before the measurement. Only the entangled two-particle system does.

this violation of relativity does not affect the math for normal events that we calculate or measure

Huh? Why do you think that? Quantum field theory has extensive experimental confirmation, so there are lots of events that have been calculated and measured that confirm relativity in this regime.

Assuming this is an accurate law of relativity

It's an accurate law of quantum field theory, yes.

so far the answer because it violates relativity just seems like a cop-out to me.

It shouldn't. See above.

 

[kurt101]

No, relativity applies period. More precisely, quantum field theory (QM + relativity) applies period. There is no way to avoid relativity by not "describing events relative to the speed of light".

I accept and agree that all events must be described relative to the speed of light. In the interpretation that I am discussing, the entanglement collapse happens instantly compared to the speed of light. Given that this instant collapse event can't actually be observed, only inferred, I don't see how it can violate relativity.

By realizing that the statement "measuring one entangled particle instantly affects the state of the other entangled particle" is simply wrong. Neither particle even has a state before the measurement. Only the entangled two-particle system does.

I don't see how the description of entanglement having shared versus individual states are effectively different. If I had to write a program to model the instance behavior of entanglement collapse, how I describe the states prior to measurement does not matter; what matters is the output to the program, i.e. the behavior at measurement.

Maybe you are not being clear about what you are really trying to say and what you are really trying to say is there is currently no instance model that can reproduce the statistics that QFT can. That I can readily accept as criticism to Bell's non-local action idea, but if this is what you are really trying to say than why not just say that? Why say the non-local action that Bell describes is wrong unless you know for certain that it can't be a possibility?

Huh? Why do you think that? Quantum field theory has extensive experimental confirmation, so there are lots of events that have been calculated and measured that confirm relativity in this regime.

I am not disputing QFT in anyway that I am aware of. I am trying to understand why Bell's non-local action idea is in conflict with QFT. You are the one who is saying Bell's non-local action possibility is wrong. I am taking your word of "wrong" to mean that Bell's non-local action is not a valid possibility for the universe. If this is truly what you are saying, than I would like to understand why.

The only invariant is that the measurement results are independent of which one is first. Which means there is no valid way to interpret either one as triggering a change in the other.

How can you even know this? You can't replay history and try to measure the system both ways to see if you get the same or differing results.

 

[PeterDonis]

I accept and agree that all events must be described relative to the speed of light.

That's not what I said. In fact I don't even understand what you mean by "described relative to the speed of light".

In the interpretation that I am discussing, the entanglement collapse happens instantly compared to the speed of light.

Which doesn't even make sense. See above.

Given that this instant collapse event can't actually be observed, only inferred, I don't see how it can violate relativity.

In whatever this interpretation is that you are discussing, do the results of spacelike separated measurements depend on the order in which they are made? If they don't, the interpretation does not violate relativity (more precisely, it violates [Edit: does not violate] QFT). If they do, it does.

Notice that I posed that question without ever saying anything about "collapse", much less "instantaneous relative to the speed of light".

Why say the non-local action that Bell describes is wrong

I have said no such thing. You are not reading carefully. See above.

How can you even know this?

Know what? That the results of spacelike separated measurements cannot depend on the order in which they are made? That is a requirement of QFT, which has extensive experimental confirmation.

 

[PAllen]

Know what? That the results of spacelike separated measurements cannot depend on the order in which they are made? That is a requirement of QFT, which has extensive experimental confirmation.

Also, if the order mattered you could send an FTL message using entanglement. Any proposed scheme for this is clearly testable.

 

[Dale]

By FTL, I mean instant, which is why I used the term universal clock.

There is no universal clock because there is no frame-invariant definition of what "instant" means. That is an essential part of relativity regardless of if you have FTL or not.

In the interpretation that I am discussing, the entanglement collapse happens instantly compared to the speed of light.

You need to use quantum field theory, not non-relativistic quantum mechanics.

 

[Buzz Bloom]

There is no universal clock because there is no frame-invariant definition of what "instant" means. That is an essential part of relativity regardless of if you have FTL or not.

Hi Dale:

I am not sure I am understanding this discussion, but I would like to ask a question.

Assumptions:
Imagine a train of carriages, each with a technologically identical clock. Imagine this train is moving at a constant speed on a circular track. Assume the track itself is stationary. Imagine in the exact center of the track is another same kind of clock. The center clock periodically broadcasts a signal identifying the current time on the center's clock. When each carriage clock receives the broadcast, it makes an immediate acknowledgment response. The center clock checks that all of the acknowledgment returned signals arrive simultaneously, and also notes the time it took for the round trip of signals. The center clock includes with the time signal a message (1) confirming all of the round-trip times of the carriage returned signals are identical, and (2) that the message also communicates the time(on the center clock) the signal was sent, and (3) the amount of time the round trip signal took. Each of the carriage clocks for itself confirms for each received message, that the time on the carriage clock is exactly equal to the time given in message from the center clock plus one-half of he round trip message travel time.

Question:
Is it correct, or is it not-correct, to say that all of he clocks involved are synchronized?

Regards,
Buzz

 

[Dale]

Is it correct, or is it not-correct, to say that all of he clocks involved are synchronized?

In which frame?

 

[Buzz Bloom]

In which frame?

Hi Dale:

I apologize for posting my question in this thread. I thought (senior moment) I was in the thread

https://www.physicsforums.com/threa...chronized-clocks-in-a-rotating-system.973180/ .​

I also just checked, and that thread was closed to further posts, so maybe I am fortunate to have posted in this thread after all, since you were kind enough to respond.

I am puzzled by the concept of "frame", but if I am required to specify a frame, it would be the frame of the stationary central clock.

Regards,
Buzz

 

[PeterDonis]

Is it correct, or is it not-correct, to say that all of he clocks involved are synchronized?

It depends on what you mean by "synchronized"--or, to put it another way, it depends on your choice of simultaneity convention, i.e., your choice of which events happen at the same time. There is no physical fact of the matter about that; it's a choice we humans make for convenience in modeling.

 

[Dale]

I am puzzled by the concept of "frame", but if I am required to specify a frame, it would be the frame of the stationary central clock.

In that frame all of the carriage clocks would be synchronized. They would not remain synchronized with the central clock except momentarily.

 

[DanielMB]

I am re-reading paper(s) by Huw Price and Ken Wharton that present an interesting explanation of entanglement. I will see if I can paste a link, otherwise one can find them on arXiv: one is for example, “Disentangling the Quantum World”. The idea is that there could be retro-casual communication between the particles, an idea first suggested by Costa de Beauregard in the late 1940s (a student of de Broglie!). Basically, a quantum particle could “know” in advance what measurement will be made in its future.

They claim that explanation is compatible with the prohibition of faster-than-light communication in Special Relativity. My question is very specifically limited to that claim. Wouldn’t such a retro-casual explanation require super-luminal retro-communication? If a particle such as a photon knows in advance what type of measurement it will eventually encounter, then that seems to mean that information had to travel back in time faster-than-light. Or else that information had to start the trip even before the photon came into existence? It does not seem to me that their explanation is compatible with Special Relativity.

I've found this : https://arxiv.org/abs/1508.01140

 

[kurt101]

In whatever this interpretation is that you are discussing, do the results of spacelike separated measurements depend on the order in which they are made? If they don't, the interpretation does not violate relativity (more precisely, it violates QFT). If they do, it does.

Is that a typo? Did you mean to say (more precisely, it does not violate QFT)? If not, I am very confused.

I have said no such thing. You are not reading carefully. See above.

In response to my statement "measuring one entangled particle instantly affects the state of the other entangled particle" you said "is simply wrong". How is my statement not a description of Bell's non-local action? Is it because I added the word "instantly"? Because that is the only difference I can see between the two.

The reason I use the term "instant" is because in experiments like the Aspect's 1983 EPR experiment, the measurement device is modified right before the entangled photon is measured, yet the non-local action is still present in the statistical results. I realize this is with entangled photons, not entangled particles, but that is the experiment that I have in my head when I am using the term instant and invoke terms like universal clock.

Know what? That the results of spacelike separated measurements cannot depend on the order in which they are made? That is a requirement of QFT, which has extensive experimental confirmation.

I know that the order of measurement does not affect the statistical result of QFT or the actual experiments. In fact that is what I was trying to convey earlier. However this is not the same as saying that the order of a measurement affects the result for a single pair of entangled photons. So when you say it is a requirement of QFT, are you referring to the statistical result or the result of a single pair?

 

[PeterDonis]

Is that a typo? Did you mean to say (more precisely, it does not violate QFT)?

Oops, yes, it was a typo, I meant "does not violate QFT". I have fixed the original post.

How is my statement not a description of Bell's non-local action? Is it because I added the word "instantly"?

No, it's because your statement requires each individual entangled particle to have a state before measurement, and that is false. I have repeatedly explained why.

when you say it is a requirement of QFT, are you referring to the statistical result or the result of a single pair?

Both. QFT requires field operators at spacelike separated events to commute. That means single pairs must commute, and it also implies that the statistical results will be independent of ordering.

 

[kurt101]

Also, if the order mattered you could send an FTL message using entanglement. Any proposed scheme for this is clearly testable.

I think the order matters to the interpretation because it preserves causality, but unless you can some how predict the state of the entangled particle before measuring it, I don't see how you could use this knowledge to send a message FTL.

 

[PeterDonis]

I think the order matters to the interpretation because it preserves causality

Not in QFT; in QFT the ordering of spacelike separated measurements is irrelevant to causality. That's what the condition that spacelike separated measurements must commute means.

 

[kurt101]

Oops, yes, it was a typo, I meant "does not violate QFT". I have fixed the original post.

The interpretation of non-local action that I am using as applied to the Alain Aspect 1983 experiment is that measuring one of the entangled photons affects the other. For this interpretation, statistically the order does not matter and for all practical purposes you would not be able to detect this order. To the best of my knowledge QFT applies to the practical things we can actually measure. So based on the test you outlined, I would not rule out this non-local interpretation of the Aspect experiment as a possibility.

Both. QFT requires field operators at spacelike separated events to commute. That means single pairs must commute, and it also implies that the statistical results will be independent of ordering.

Single pairs must commute for the math to work out, but I am thinking that this is different than the actual mechanism having to commute. If the actual mechanism does not commute, but the state it acts upon is random, you would get the same statistical result.

 

[PeterDonis]

Single pairs must commute for the math to work out, but I am thinking that this is different than the actual mechanism having to commute.

In QFT, the field operators are the "actual mechanism".

 

[PeterDonis]

The interpretation of non-local action that I am using

Are you trying to describe the interpretation used in the paper described in the OP? Or is this an interpretation that appears in some other reference? Or is it just something you made up?

 

[PAllen]

I think the order matters to the interpretation because it preserves causality, but unless you can some how predict the state of the entangled particle before measuring it, I don't see how you could use this knowledge to send a message FTL.

If the order matters, you have causality outside of the light cone. It is precisely the commuting property that guarantees there is no ftl causality, because you can’t distinguish whether A caused B or B caused A.

 

[Dale]

I think the order matters

The order can’t matter because the order is not the same in all frames.

 

[Buzz Bloom]

They would not remain synchronized with the central clock except momentarily.

Hi Dale:

I think I may be understanding this. Each carriage experiences a radial acceleration away from the center. This acceleration changes the relative time rate in each carriage as, compared with the time rate at the center. Is this correct? Is it also correct that the the rate of time is the same in all of the carriages? If I am correct about these guesses, I would appreiate seeing an equation that compares the rate of time differences between the carriages and the center point in terems of the accelleation and/or the radius and the tangential velocity. I tried (and failed) to find this by an online search, was probably chose the search keys badly.

Regards,
Buzz

 

[Ibix]

This acceleration changes the relative time rate in each carriage as, compared with the time rate at the center. Is this correct?

No. It's just that the circling clocks are moving at constant speed $v$ compared to the central clock, so have the usual time dilation factor of $1/\sqrt{1-v^2/c^2}$. GPS satellite clocks tick at a very slightly modified rate to account for a (slightly more complex because of gravity) version of this.

 

[kurt101]

Are you trying to describe the interpretation used in the paper described in the OP? Or is this an interpretation that appears in some other reference? Or is it just something you made up?

No, nothing from the paper, but what I thought Bell meant by non-local action as applied to an actual experiment.

 

[PeterDonis]

what I thought Bell meant by non-local action as applied to an actual experiment.

Please give a reference for whatever Bell paper you are basing this on.

 

[kurt101]

The order can’t matter because the order is not the same in all frames.

If the order matters, you have causality outside of the light cone. It is precisely the commuting property that guarantees there is no ftl causality, because you can’t distinguish whether A caused B or B caused A.

The order matters to the *interpretation* of non-local action; the interpretation is not something we can directly measure, but only something we can infer from statistics and the experiments.

With the non-local action interpretation there is no way to directly distinguish whether A caused B or B caused A and so your comment about causality outside of the light cone does not apply.

Take the EPR experiment done by Alain Aspect in 1983. Call the entangled photons A and B. You can only measure either A or B first. Say you decide to always measure A first. At the end of the experiment, after many iterations, you still get the correlation that A's polarizer had an effect on B. You get that correlation even if you change the orientation of A's polarizer so it changes faster than any signal that could be transmitted at the speed of light. You can also say that B's polarizer had an effect on A, but this would violate causality, because you measured B after you measured A. You might say there were some observers that did not see it that way, and to that comment I would say steer B back around to where A was measured by using gravity, so that there is no question to any observer that B was measured second.

So if you are in the non-local action camp and subscribe to causality, based on this experiment it would be logical to conclude that there is a definite order even though you can not directly observe it.

 

[PeterDonis]

The order matters to the *interpretation* of non-local action

This seems to contradict:

With the non-local action interpretation there is no way to directly distinguish whether A caused B or B caused A

"No way to directly distinguish whether A caused B or B caused A" sounds like "order doesn't matter".

You can only measure either A or B first.

No, you set up the measurements to be spacelike separated so there is no frame-independent fact of the matter about which one you measure first.

Say you decide to always measure A first.

As the experiment was set up, there was no attempt to control the exact measurement times in this way. The only thing that was controlled was that the measurements were spacelike separated (meaning that the time windows for the measurements had to be small compared to the light travel time over the path lengths involved).

So I assume you are proposing an alternate version of the experiment, where the ordering in some particular chosen frame (the "lab" frame) is controlled explicitly?

f you are in the non-local action camp and subscribe to causality, based on this experiment it would be logical to conclude that there is a definite order even though you can not directly observe it.

I don't see how, since you could just as easily do the experiment enforcing the opposite order, and the results would be indistinguishable. In other words, experimentally the order doesn't matter. And relativity explains why: because the order is not invariant, and only invariants can affect the actual physics.

 

[kurt101]

Please give a reference for whatever Bell paper you are basing this on.

Bell's paper Bertlemann's socks: https://hal.archives-ouvertes.fr/jpa-00220688/document
The third explanation:

Thirdly, it may be that we have to admit that causal influences
- do go faster than light. The role of Lorentz invariance in the completed theory would then be very problematic. An "ether" would be the cheapest solution /22/. But the unobservability of this ether would be disturbing. So would the impossibility of "messages' faster than light,
which follows from ordinary relativistic quantum mechanics in so far as
it is unambiguous and adequate for procedures we can actually perform.
The exact elucidation of concepts like 'message' and 'we' would be a
formidable challenge.

 

[kurt101]

So I assume you are proposing an alternate version of the experiment, where the ordering in some particular chosen frame (the "lab" frame) is controlled explicitly?

Yes, it is a slightly modified version of the experiment.

I don't see how, since you could just as easily do the experiment enforcing the opposite order, and the results would be indistinguishable. In other words, experimentally the order doesn't matter. And relativity explains why: because the order is not invariant, and only invariants can affect the actual physics.

Exactly, the results would be indistinguishable! This is what I have been saying all along and is why I call it an interpretation.

 

[Dale]

The order matters to the *interpretation* of non-local action

I am OK with that. The interpretation can change from frame to frame.

the interpretation is not something we can directly measure, but only something we can infer from statistics and the experiments

Not really. Interpretations are simply assumed for convenience, aesthetics, philosophical, or personal reasons. I am not really interested in interpretations.

 

[PeterDonis]

the results would be indistinguishable! This is what I have been saying all along and is why I call it an interpretation

I don't see the point of an interpretation that tries to distinguish indistinguishable results.

Bell's paper Bertlemann's socks: https://hal.archives-ouvertes.fr/jpa-00220688/document
The third explanation:

This explanation does not say that spacelike separated events that are causally connected have to have a definite order. It just says that spacelike separated events can be causally connected. Note that the word "order" appears nowhere in Bell's description of this interpretation.

Bell does propose an "ether" (which would seem to suggest some sort of preferred frame and therefore a preferred simultaneity convention) as one possible way of implementing this solution, but it's not the only possible one (and Bell does not claim it is--he only says ether would be the "cheapest" solution). Another would simply be to develop a concept of "causality" that did not require one to distinguish between the "cause" and "effect", at least not in the sense that one has to come before the other. Which is still a challenge, but as Bell says, any explanation you pick is going to be a challenge.