Is action at a distance possible as envisaged by the EPR Paradox.

[unusualname]

If the events are space-like related, even 1 m/s faster than c, there is a frame in which those events are simultaneous.

If you assume that the signal doesn't travel in classical space then its journey is not related to a SR reference frame, but you're right that there is a privileged frame in which it would appear (to the observer) that the (entangled) events had zero time between them (but the observer wouldn't be able to observe the "signal" traveling between the particles in any classical manner)

For 3 or more particle entanglements (eg quantum computer) you wouldn't have such a privileged reference frame such that all entanglement events were simultaneous.

I need to read that preprint linked to above carefully to understand the importance of the privileged reference frame they mention (apparently it's crucial to bohm-hiley's pilot wave constructions)

I don't think it's a big "cheat" to suggest that the signal travels in non-classical space, not with the ridiculous plethora of calabi-yau manifolds and the like thrown up by string theory

 

[unusualname]

The test of entanglement speed (they call it "speed of quantum information") from the preprint Testing spooky action at a distance assumes a causal mechanism with a signal traveling in classical space, so I'm not sure if their measurement in the conclusion of a lower bound is even correct if we instead assume the signal travels in non-classical space (either other dimensions or something even weirder) (although there obviously is a lower bound, however it's the upper bound I'm interested in)

A simplistic model might be like this:

   ##      ##      ##      ##      ##
  #  #    #  #    #  #    #  #    #  #
 #    #  #    #  #    #  #    #  #    # 
#      ##      ##      ##      ##

 ------>  particle traveling at speed <= c in classical spacetime

    #
   # #    signal travels around extra dimensional
  #   #   ("kaluza-klein") coils at speed c^k

For "circular" coils we'd require the signal to travel at ~c^2 to keep up with the particle, but we can expect even faster speeds for efficient "communication" between the particles.

Now, it may be that these "spooky" signals can only travel "near" a path traced out by particles in classical space-time, and they should influence events restricted to those observed in entanglement (so the signals are responsible for the "magic" entanglement correlations we see, which can happen faster than the speed of light would allow with a classical signal)

In particular, there is no mechanism for transmitting FTL information through classical space (the entangled particles communicate with each other FTL but this only achieves entanglement correlations of their quantum properties, which we can't deterministically influence) I can see how this may be difficult to check for in current Bell Test experiments since you need to fine-tune an experiment to measure what may be a very tiny delay between the entangled particles switching quantum states.

So perhaps the better way forward would be to detect delays in multi-particle entanglements such as sufficiently complex quantum computers (what are we up to so far ~100 bits yet?)

On a side note, I wonder why Bohm, Hiley et al didn't consider a signal traveling in non-classical space, perhaps physics models based on exotic topology weren't the vogue and they were scared of additional ridicule?

And on the speculative subject of entanglement & consciousness I might also note that our brains don't seem to allow us to think at infinite speed, there seems a (fairly cumbersome) delay involved, but of course that may be due to the requirement for complex chemical and biological mechanisms relating to memory and the like (In fact there is a well known 40hz effect observed in human brains, eg see this paper). It's interesting that autistic savants and very young children seem able to process certain information quicker than normal adults, that may be due to their brains lacking certain biological mechanisms and hence they are not slowed down so much.

I would think the determination of an upper bound on entanglement speed would be at least as important as finding the Higgs Boson, so it's surprising that there doesn't seem much experimental effort in this direction.

 

[zonde]

Where in the paper does it say ANYTHING remotely similar to the idea that the Product State statistics are approached?

You where talking about evidence not reference. You said: "That isn't so. There is absolutely no evidence (cite it if you think I am wrong) ..."
Besides this experiment wasn't about violation of BI so the calibration of entangled state was only part of preparations for main experiment.

By way of example: at 0 degrees, the Product State is 25.0% and the stated observation was apparently 1.9%. Does not seem too close. At 45 degrees, the Product State value should be 50.0% and the actual was apparently 42.6%.

visibility is defined as follows:
V=(max-min)/(max+min)
In case of visibility for minimum we can use this formula (but it's the same as above):
V=100%-2*min/(max+min)

Taking into account these formulas visibility for product state at +45deg/-45deg (minimum) would be 0 for 100% efficiency. Maximum in this case is coincidence rate at +45deg/+45deg.
Visibility for product state at H/V (minimum) would be 1 (maximum is coincidence rate at H/H).
So for product state with diminishing interference term as efficiency approaches 100% H/V visibility should stay the same but visibility for +45deg/-45deg should tend to 0 with increasing efficiency.

QM prediction used by Bell was that theoretical visibility for 100% efficiency is 1 for any angle α and β=α+Pi/2.

So according to QM visibility should be the same for +45deg/-45deg and H/V (of course not exactly 1 when considering effect of noise).
This is not so in this experiment by quite significant amount.

Don't you think the authors would be raising flags if the stats deviated from QM predictions by a significant amount?

No, I don't think so. Because QM says that "decoherence happens". Another thing is that it might be assumed that interference term can be diminished due to some imperfections of setup that you somehow can't track down. And you generally do not rise flags because of some unknown imperfection or noise or whatever (that as you would think you are too lame to track down). You try to eliminate it and if you can't you just ignore it (maximum include it in some statistical error estimates) hoping that it will not affect your experiment too much.

By the way, the 25% detection stat is a bit deceiving. That is because the value is net. Net meaning for both detectors jointly. Obviously, there are a lot of unmatched hits too. I would estimate the gross efficiency at close to 50% (since 50%^2 = 25%).

No, it is not. There is no reason to square efficiency if you have identical rate in both arms.
Let's say we have 25% efficiency in first arm - 80k/s rate, but 100% efficiency in second arm - 320k/s rate that we divide into two parts 80k/s and 240k/s.
Now we have probability that 1 of each 4 clicks from first arm will make coincidence with that 80k/s part from second arm and 3 of each 4 clicks will make coincidence with that 240k/s part. Together of course every click from first arm makes coincidence with click at second arm (as it has 100% efficiency).

 

[RUTA]

For 3 or more particle entanglements (eg quantum computer) you wouldn't have such a privileged reference frame such that all entanglement events were simultaneous.

But, any two are simultaneous in SOME frame and their temporal order switches in other frames, so how do you argue for an unambiguous causal ordering without resorting to a preferred frame?

I need to read that preprint linked to above carefully to understand the importance of the privileged reference frame they mention (apparently it's crucial to bohm-hiley's pilot wave constructions)

I don't think it's a big "cheat" to suggest that the signal travels in non-classical space, not with the ridiculous plethora of calabi-yau manifolds and the like thrown up by string theory

My understanding of BM is the wave function is updated in configuration space, so configuration space is "real" for them. But, I don't see how the precise mechanism avoids FTL comm or a preferred frame for an unambiguous causal ordering. The events (measurement outcomes) occur in spacetime and their temporal order is relative if they're space-like related, regardless of what is going on "behind the scenes." Therefore, if you want to explain them via "causal relations" you either need a preferred frame or FTL comm. [There is another way out -- you can say the future has causal influence on the past, but that's another discussion.]

 

[unusualname]

But, any two are simultaneous in SOME frame and their temporal order switches in other frames, so how do you argue for an unambiguous causal ordering without resorting to a preferred frame?

You argue that a classical observer can't observe the sequence of causal events being caused (by the non-classical FTL signal), they only observe the final states.

The fact that an observer might see simultaneous and backward events isn't really a problem, except that it might confuse the observer :) (and remember I'm arguing that the entanglement events can't transmit classical information FTL)

Incidentally, I stated above that we couldn't deterministically influence the quantum state, I think that's true unless, er, you actually are the particle! ie consciousness allows us to choose certain quantum states, which then propagate to macroscopic events.

So in my (very speculative) theory, if we had a brain the size of a galaxy we probably could communicate FTL information within it provided the "information" was restricted to thoughts.

Ignoring the wild speculations, for a significant breakthrough in quantum theory a reasonably simple result is required on an upper bound for "the speed of quantum information"

That would be paradigm changing :)

 

[unusualname]

ah, hold your horses, this has been theoretically postulated:

fundamental quantum limit on the rate of operation of any information-processing system

In a paper published in the journal Physical Review Letters, Levitin and Toffoli present an equation for the minimum sliver of time it takes for an elementary quantum operation to occur. This establishes the speed limit for all possible computers. Using their equation, Levitin and Toffoli calculated that, for every unit of energy, a perfect quantum computer spits out ten quadrillion more operations each second than today's fastest processors.

http://arxiv.org/abs/0905.3417

 

[unusualname]

so given the above result (which is basically derived from the energy-time uncertainty relationship, there are other papers which discuss this, eg see Ultimate physical limits to computation) we already have an upper bound for the speed of quantum computers, hence it would be difficult to distinguish the time taken for the entanglement effects to propagate as opposed to this limit on qubit switching time.

So we'd need to devise our experiment more carefully to distinguish the time taken for propagation of entanglement events.

If we can assume that a classical space distance is quantitatively related to the length of the non-classical path followed by FTL entanglement signals (so if we increase the usual classical space between our entangled pairs we can assume the non-classical path I'm hypothesising also increases) then we could demonstrate that the "speed of entanglement" is finite by carrying out identical experiments at different distances and recording an increase in the time for the entangled pairs to correlate.

Admittedly, this is looking trickier and trickier...

 

[RUTA]

You argue that a classical observer can't observe the sequence of causal events being caused (by the non-classical FTL signal), they only observe the final states.

They can observe the sequence. The problem is other people will observe different sequences.

The fact that an observer might see simultaneous and backward events isn't really a problem, except that it might confuse the observer :) (and remember I'm arguing that the entanglement events can't transmit classical information FTL)

They will be confused if (1) the events are space-like separated, (2) they believe there is a causal connection b/w the events, (3) there is no preferrred frame, (4) the future doesn't causally influence the past. They will be confused b/c this a self-inconsistent set of assumptions.

 

[unusualname]

They can observe the sequence. The problem is other people will observe different sequences.

No they can't, they can only observe the final quantum states, which had a causal sequence determined by the journey taken by the FTL non-classical signal that's responsible for the entanglement correlations. The observer will just see the final (classically observable) quantum states pop up in some arbitrary order determined by their classical reference frame.

It doesn't matter that other observers might claim the states appeared in a different order, no rule of Special Relativity is broken if no classical information was transmitted.

They will be confused if (1) the events are space-like separated, (2) they believe there is a causal connection b/w the events, (3) there is no preferrred frame, (4) the future doesn't causally influence the past. They will be confused b/c this a self-inconsistent set of assumptions.

Well QM is confusing :) They might believe all they like there is a causal connection between the events, but there's no classical causal connection between them, that's for sure.

I don't believe in fuzzy (or philosophically devious) interpretations of entanglement which assumes some magic instantaneous effect, I'd rather be more scientific and accept that our reality has some additional (but mathematically constructible) components which when taken into consideration give us a way to construct new physics models which append to the old and don't contradict long proven observations of theories like SR

 

[DevilsAvocado]

... For 3 or more particle entanglements (eg quantum computer) you wouldn't have such a privileged reference frame such that all entanglement events were simultaneous.

Found one multiqubit (four-qubit) entangled experiment by Zeilinger et al. that cannot be described by local realism:

http://homepage.univie.ac.at/philip.walther/paper/ClusterBelll_PRL05_95_020403.pdf"
Cluster states are a new type of multiqubit entangled states with entanglement properties exceptionally well suited for quantum computation. In the present work, we experimentally demonstrate that correlations in a four-qubit linear cluster state cannot be described by local realism. This exploration is based on a recently derived Bell-type inequality [V. Scarani et al., Phys. Rev. A 71, 042325 (2005)] which is tailored, by using a combination of three- and four-particle correlations, to be maximally violated by cluster states but not violated at all by GHZ states. We observe a cluster-state Bell parameter of 2.59 ± 0.08, which is more than 7 σ larger than the threshold of 2 imposed by local realism.

 

[unusualname]

Found one multiqubit (four-qubit) entangled experiment by Zeilinger et al. that cannot be described by local realism:

Over the last few decades I think local realism has clearly been shown to be untenable, and not even desirable ( what an uninteresting world it would be : ) )

I'm puzzled why there has been resistance, since we all possesses one thing that can't be modeled by local realist physics, our consciousness

I don't think my suggestions are even that adventurous, I mean, just asking for an extra dimension or two to bypass SR restrictions on FTL signalling is hardly less bizarre than the Copenhagen Interpretation, many-worlds or assuming instantaneous multiple correlations by magic.

You can ignore my speculations about consciousness and entanglement, but I think it's a good bet that the two are related, and I must again emphasise that a really simple process (evolution) created conscious beings.

 

[DrChinese]

Over the last few decades I think local realism has clearly been shown to be untenable, and not even desirable ( what an uninteresting world it would be : ) )

I'm puzzled why there has been resistance, since we all possesses one thing that can't be modeled by local realist physics, our consciousness.

So true! While some would like to return to local realism, it seems to me that the exciting things to ponder are in the other direction entirely. There could be some weird dimensions which exist in supersets of existing theory. Perhaps quantum non-locality is local in those dimensions. Or maybe there are strange beings there. (I mean: stranger than the strange beings here.)

 

[DevilsAvocado]

... (I mean: stranger than the strange beings here.)

Please forgive me! But I can’t help it! Moooooaaahhhhaaaa LOL!

 

[DevilsAvocado]

... or assuming instantaneous multiple correlations by magic.

I guess this will qualify me for the "severe strangeness classification" , but I can’t get this out of my head. I assume you are proposing a finite limited upper "speed" of the quantum non-local connection, right?

What 'mechanism' would handle the 'negotiation' between entangled particles, since there is only one particle (or 'end' of the entangled WF) who can 'settle/decide' the correlated outcome?

I know there are interpretations like RBW or MWI that makes this an non-issue, but these interpretations brings other 'stuff' that are more complex than this 'entangled negotiation' – and I always been a big fan of Occam's razor.

Any thoughts?

 

[DevilsAvocado]

But, any two are simultaneous in SOME frame and their temporal order switches in other frames, so how do you argue for an unambiguous causal ordering without resorting to a preferred frame?

How do we handle this view in respect of my question above? (i.e. the 'negotiation')

 

[my_wan]

So true! While some would like to return to local realism, it seems to me that the exciting things to ponder are in the other direction entirely. There could be some weird dimensions which exist in supersets of existing theory. Perhaps quantum non-locality is local in those dimensions. Or maybe there are strange beings there. (I mean: stranger than the strange beings here.)

Here is a paper that shows a violation of Bell's inequalities in classical statistics, among other things.

Abstract: http://arxiv.org/abs/0906.4919"

http://arxiv.org/abs/0906.4919]Quantum[/PLAIN] mechanics can emerge from classical statistics. A typical quantum system describes an isolated subsystem of a classical statistical ensemble with infinitely many classical states. The state of this subsystem can be characterized by only a few probabilistic observables. Their expectation values define a density matrix if they obey a "purity constraint". Then all the usual laws of quantum mechanics follow, including Heisenberg's uncertainty relation, entanglement and a violation of Bell's inequalities. No concepts beyond classical statistics are needed for quantum physics - the differences are only apparent and result from the particularities of those classical statistical systems which admit a quantum mechanical description. Born's rule for quantum mechanical probabilities follows from the probability concept for a classical statistical ensemble. In particular, we show how the non-commuting properties of quantum operators are associated to the use of conditional probabilities within the classical system, and how a unitary time evolution reflects the isolation of the subsystem. As an illustration, we discuss a classical statistical implementation of a quantum computer.

Section VIII covers Bell's inequalities. I can't honestly disparage considerations that local realism may in fact be a dead horse. What I can disparage is the claim that it is certainly so. It's not entirely honest on either side of the fence.

 

[DevilsAvocado]

... local realism may in fact be a dead horse. What I can disparage is the claim that it is certainly so. It's not entirely honest on either side of the fence.

I agree, and to be real honest we must also consider that fact that local realism may be a stone dead parrot... with the Norwegian Blue plumage...

Sorry, just a really bad joke. I agree in what you are saying – "It ain't over 'til the fat lady sings."

 

[my_wan]

I agree, and to be real honest we must also consider that fact that local realism may be a stone dead parrot... with the Norwegian Blue plumage...

Sorry, just a really bad joke. I agree in what you are saying – "It ain't over 'til the fat lady sings."

Actually I thought it was a pretty good Monty joke. [PLAIN]https://www.physicsforums.com/images/smilies/smile.gif

Given the weakness of the opposing arguments here, it appeared to me that many would get a false impression of excessive certainty of a particular view. I felt that needed corrected, as the actual physics involved don't presently allow a significant degree of certainty. Ironic [PLAIN]https://www.physicsforums.com/images/smilies/laughing.gif

 

[Galap]

But... even if we cannot use entanglement to send usable information FTL, the particles must clearly be 'communicating' in some way to present the opposite random property, right? And Bell showed there are no local hidden variables involved... or did I miss something?

MWI is the only 'way out' of this is, as I understand...?

The particles do not need to communicate. One explanation is that they are two aspects of the same thing that aren't in the same place. We have pretty much determined that space and time aren't fundamentals. They are more emergent properties of the universe. Why is it so surprising then that not all phenomena work that way. Something can be in two places at once. In this case, it's the wavefunction. No communication, just evidence that GR isn't a complete theory.

 

[unusualname]

The particles do not need to communicate. One explanation is that they are two aspects of the same thing that aren't in the same place. We have pretty much determined that space and time aren't fundamentals. They are more emergent properties of the universe. Why is it so surprising then that not all phenomena work that way. Something can be in two places at once. In this case, it's the wavefunction. No communication, just evidence that GR isn't a complete theory.

The problem with that argument is that it's a struggle to explain multiparticle entanglement in quantum computers (and perhaps the conscious brain), since all the particles would have to occupy the same "thing" (or have some component of their positions in multidimensional space fixed) undisturbed for long periods of time.

I think it's easier to imagine that signalling between the particles occurs in some non-classical space.

DrChinese posted a link to a recent experiment which established entanglement in photons from different sources ( http://arxiv.org/abs/1005.1426 ) , and several similar results have been published. These type of results make it difficult for me to see how entanglement can be due to particles being aspects of the same "thing" (or having the same component in multidimensional space)

But it's true that we can't rule it out, so we need to make some progress on this impasse in quantum understanding, and I suggest that testing for signalling is not beyond the bounds of current experiments.

It will be difficult if the signalling speed (the "speed of entanglement") is so fast as to be comparable with the minimum qubit switching time predicted by the energy-time uncertainty relation (see links above to limits on computing), since any experiment will have to distinguish the entanglement propagation speed from the quibit switching time.

In fact, if god's being a devil she might have made the signalling time smaller than we can measure classically, then we won't be able to distinguish the effect from an instantaneous one.

But for historical record, in case these forums are archived for future generations, I want to state I believe that signalling occurs in some not overly exotic non-classical space (so not some freaky fractal dimension topology or weird discrete construct).

I'm sure reality can't be completely bizarre, because if it was, evolution would have created beings to take advantage of the bizarreness. That's essentially why most paranormal stuff doesn't work, if it did it would have emerged in an obvious way from evolution, and we would have telepathic cats and levitating birds ;)

 

[DevilsAvocado]

Actually I thought it was a pretty good Monty joke. [PLAIN]https://www.physicsforums.com/images/smilies/smile.gif[/QUOTE]

(Thanks! but) "Look, matey," I do know there are 'customers' who "https://www.physicsforums.com/showpost.php?p=2709968&postcount=253"" about this very silly joke!

Given the weakness of the opposing arguments here, it appeared to me that many would get a false impression of excessive certainty of a particular view.

If you suggest: Running down the whole 18th century for an "optical solution", or disqualifying Einstein, Bohr, Bell & Aspect as bummers "who doesn’t care" – I’m with you all the way bro! (severe irony)

 

[DevilsAvocado]

The problem with that argument is that it's a struggle to explain multiparticle entanglement in quantum computers (and perhaps the conscious brain), since all the particles would have to occupy the same "thing" (or have some component of their positions in multidimensional space fixed) undisturbed for long periods of time.

I think it's easier to imagine that signalling between the particles occurs in some non-classical space.

...

But for historical record, in case these forums are archived for future generations, I want to state I believe that signalling occurs in some not overly exotic non-classical space (so not some freaky fractal dimension topology or weird discrete construct).

I'm sure reality can't be completely bizarre, because if it was, evolution would have created beings to take advantage of the bizarreness. That's essentially why most paranormal stuff doesn't work, if it did it would have emerged in an obvious way from evolution, and we would have telepathic cats and levitating birds ;)

Many BIG THANKS for that! It's one of the best post I’ve seen so far! Awesome!

But you still have to explain to me – if only one of these entangled photons can decide spin up/down – WHO decides!?

 

[DevilsAvocado]

I'm sure reality can't be completely bizarre, because if it was, evolution would have created beings to take advantage of the bizarreness.

Or put it this way:

Either our brains are incomplete and the reality of QM is fooling us all the time and every day, or QM (and/or GR) is incomplete?

But would incomplete brains discover a complete theory??

 

[DevilsAvocado]

... One explanation is that they are two aspects of the same thing that aren't in the same place.

As UN mentioned, you run into trouble with the complete madness that DrChinese has enforced into our heads – https://www.physicsforums.com/showthread.php?t=376225" ??

And in this case; there has to be four "aspects of the same thing", and this of course is going to get 'worse' in the future:

 

[DrChinese]

...Section VIII covers Bell's inequalities. I can't honestly disparage considerations that local realism may in fact be a dead horse. What I can disparage is the claim that it is certainly so. It's not entirely honest on either side of the fence.

Not sure I would agree with that assessment. Granted, that may not have been true in 1965, but we have had time to consider Bell since. I am somewhat familiar with this paper (I keep links on a lot of local realists for easy reference, and this was one). Not the first time violation of Bell-like inequalities have been alleged in classical situations. If you are interested in discussing this specific paper, I would be happy to. However, I don't sense that is the point you are making.

I think you are saying that the matter is not decided. And I think it quite is. Bell Inequalities are violated experimentally in agreement with the predictions of QM. EPR local realism is untenable. Now, keep in mind that in the intervening years since Bell, all kinds of entanglement phenomena has been discovered. With a green light from Bell, QM has made prediction after prediction which can be verified - none of which remotely smack of local realism and in fact get farther and farther away.

For example: in another thread, I presented evidence that particles outside each others' light cones can be entangled. It's going to be a cold day before that one can be explained classically.

 

[DevilsAvocado]

... I think you are saying that the matter is not decided. And I think it quite is.

DrC, would you say that "the fat lady has sung", even though we haven’t yet fully understood the complete 'mechanism' behind entanglement?

 

[DevilsAvocado]

What the he*k!? Frame Dragger is permanently banned?

 

[IcedEcliptic]

Opinions put aside, do the Bell Tests and ones that Dr. Chinese is speaking of mean that this is a settled issue? I am not in this community, so I do not know. It seems to me, that there is no classical means for this to occur, that separate sources entangle, game over, yes?

I am left with: Reality is not what I think it is from my daily life, and action in time and space is not ordered the way I expect it to be. Spatial separation may be meaningless when it does not change causality?

 

[DrChinese]

Opinions put aside, do the Bell Tests and ones that Dr. Chinese is speaking of mean that this is a settled issue? I am not in this community, so I do not know. It seems to me, that there is no classical means for this to occur, that separate sources entangle, game over, yes?

It is settled to the mainstream community. There are a few doubters. Of course, there are also doubters of a spherical Earth, General Relativity, the Big Bang, etc. If we found out everything we know is wrong, then this could be too.

 

[my_wan]

Not sure I would agree with that assessment. Granted, that may not have been true in 1965, but we have had time to consider Bell since. I am somewhat familiar with this paper (I keep links on a lot of local realists for easy reference, and this was one). Not the first time violation of Bell-like inequalities have been alleged in classical situations. If you are interested in discussing this specific paper, I would be happy to. However, I don't sense that is the point you are making.

I appreciate the offer, but your right, that's wasn't my intention. I tend to lean on the realist side, as a personal preference, and I'd do well just to hear out the objections. I have a lot of issues with the classical models suggested so far, but for reasons unrelated to EPR. Of the attempts at these models 't Hooft seem to be the most torturous. They're generally like trying to force fit a car motor on a moped, and explaining away the extra parts. Yet I still haven't found a fundamental reason why EPR must be defined in terms of a physical switch activated by a FTL mechanism. In fact it seems that not only does this FTL interpretation require assuming a realistic mechanism, contrary to the standard interpretation, but also assumes a particular type of physical character of this mechanism. Vector spaces and statistical ensembles both by their very nature allows an arbitrary number of parameters to be summed up in just a few variables. Ensembles only provide correlation, not causation, classical or otherwise.

I think you are saying that the matter is not decided. And I think it quite is. Bell Inequalities are violated experimentally in agreement with the predictions of QM. EPR local realism is untenable. Now, keep in mind that in the intervening years since Bell, all kinds of entanglement phenomena has been discovered. With a green light from Bell, QM has made prediction after prediction which can be verified - none of which remotely smack of local realism and in fact get farther and farther away.

Nobody can seriously question QM or the limits provided under Bell's Theorem. I'm aware of all sorts experiments from EPR where both detectors did the measurement first in their own frame, delayed choice, frame dependent correlations, Afshar, single photon pictures, pictures taken using photons that never seen the object being photographed, metamaterials, etc. I might have missed something, but I hope not. I'm presently considering a sort of delayed choice/Afshar hybrid. Still I have yet to see an empirically backed argument to rule out all class of models as generally defined by Relational QM.

For example: in another thread, I presented evidence that particles outside each others' light cones can be entangled. It's going to be a cold day before that one can be explained classically.

Yet therein lies the weakness of your case. Essentially 'entangled' means correlated. Thus your case holds under the assumption that correlation equals causation. Even wise tales warn of that one. The assumption that these properties are absolute real properties may turn out to be akin to assuming velocity is an absolute. Nobody is surprised that the relative velocity of distant objects instantly change with a local boost. The fact that entanglement can be manipulated, is frame dependent, actually lends some support to a purely relational interpretation.
http://arxiv.org/abs/quant-ph/0302095"
http://arxiv.org/abs/quant-ph/0205179"

So the objections I would like to hear is how to empirically rule out such models. I'm not asking any given interpretation to be proved one way or the other, merely that EPR correlations rule out this relational model class the same way it rules out a local real signal switching actual mechanisms. This would be trivial if an actual FTL message could be sent, otherwise it hinges on a correlation equals causation claim.