CFD - Counterfactual Definiteness

[Simon Phoenix]

You can have locality and realism (CFD) with retrocausality

Yes, that's a good point - I think that makes intuitive sense.

 

[Simon Phoenix]

≠

At least one of the instruments KNOWS THE VALUE of both the setting and the outcome of the other instrument.

I am saying that this statement is true no matter what. Otherwise, the quantum correlations cannot happen as they do.

You are saying that when we employ 'realistic variables' in the description, we are forced to accept such a statement. But when we don't do that, the matter is left open and it depends upon interpretation.

... Did I get that right?

Sort of - but I think there's always going to be a struggle pinning everything down with 'precise enough' words. In my opinion we do have to be a bit careful using words like 'know' and even 'information'. I've tried to be careful about this in my posts above so that I would say that in a realistic variable treatment the information about what's happening at B has to be in some sense available to A. And here I'd hope that the use of 'in some sense' would indicate that I'm not being overly precise but rather attempting to convey an intuition.

If we had some probability distribution P(X,Y) where X and Y are 2 random variables then we could determine from this the various marginal and conditional probabilities, like P(X | Y), for example. If X and Y are completely uncorrelated then we have that
P(X) = P(X | Y). In information terms we would say that information about Y gives us no information about X. If they were correlated so that
P(X | Y) ≠ P(X)
then knowledge of Y gives us some information about X.

In the Bell inequality set-up we have a slightly more complicated conditional probability - but it can be understood in exactly the same way we'd approach the more simple distribution for X and Y above.

So, for example, in the Bell case we might want to examine a conditional probability distribution
P(A | a, b)
and this means something like the "probability that the result A is some value given that the settings of the measuring instruments are a and b". It's certainly one that we can measure from actually performing the experiment.

So here we're assuming that there is some functional dependence on a and b. In a loose sense we might then say that the outcome at A 'knows' that the settings are a and b. More technically we might make the statement that knowledge of the settings at a and b gives us additional information about the probabilities of A.

If there were no such functional dependence on b, say, then we would write
P(A | a, b) = P(A | a)
This latter identification is, of course, a critical component of the Bell proof - it's the imposition of the locality condition. It's an assumption that says that the probability of the outcomes at A is not conditioned upon the settings at b. Expressed another way, we could say that knowledge of the settings at b confers no additional information pertinent to the results at A.

With this locality condition and the assumption of CFD then we can show that the probability functions we measure are constrained by the Bell inequality.

If we don't want them to be so constrained (i.e. have the possibility to violate the inequality) then at least one of these assumptions has to go. If we choose to retain CFD then we have to lose locality, if we want to have the possibility to break those constraints.

 

[Eye_in_the_Sky]

Difficulty is related to independence of spacetime events. I mean that standard philosophical basis takes spacetime events as fundamental to reality and they are arranged in some system (spacetime) that does not allow arbitrary connections between them. And scientific method is developed on top of that standard philosophical basis.

As I see nonseparability is in conflict with the idea of spacetime events as fundamental.

Well, at the core of the "scientific method", I think, is the idea of 'cause' and 'effect'.

On top of that are the constructs of 'space' and 'time', and within that framework the various 'cause' and 'effect' relationships are to be organized.

Now ... what does it mean to say "spacetime events are fundamental"? Does it mean
this?

Given all of the 'effects' in spacetime, the joint-state of all of their 'causes' is 'separable'.
(This joint-state can be regarded as a description of a 'system' that 'resides' within spacetime.)

 

[Eye_in_the_Sky]

... then we are forced to say

The 'quantum state' is PHYSICAL.

That depends on your interpretation of QM. You're certainly not forced to say that.

That is VERY interpretation dependent.

Ok ... Ok. I will have to recast my claim.

______

IF

there is no violation of 'causal locality'

AND

the states of their instruments bear a relation of 'separability' throughout the spacetime regions A and B

THEN:

If the "correlation phenomenon" is an 'effect' in spacetime, then there is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime
______
______

What is the explanation for the non classical correlations that violate Bell inequalities that do not include any superluminal signaling in any aspect of experiment ?

According to what I have claimed above, the answer to your question would be:

If you wish to construe the "non classical correlations" as an 'effect' in spacetime, then the 'cause' of them is a 'nonlocal beable'; it 'acts' within spacetime, but does not itself 'reside' within spacetime.

 

[Eye_in_the_Sky]

*There are classical phemonena with no explanation as well.

Zafa, please give some examples if you can. I can't get clear what you mean.

 

[RUTA]

______

IF

there is no violation of 'causal locality'

AND

the states of their instruments bear a relation of 'separability' throughout the spacetime regions A and B

THEN:

If the "correlation phenomenon" is an 'effect' in spacetime, then there is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime
______
______

The existence of a beable is not necessary. Bohr, Ulfbeck and Mottelson's genuine fortuitousness claims the events are uncaused. Here is a quote from them lifted from http://arxiv.org/pdf/quant-ph/0202148.pdf:

the individual fortuitous event, which quantum mechanics deals with, comes by itself, without any cause, and is entirely beyond theoretical analysis.

 

[bhobba]

If the "correlation phenomenon" is an 'effect' in spacetime, then there is a 'nonlocal beable' that acts within spacetime as its 'cause'.

The correlation is simply the result of how you set things and usually a consequence of conservation laws.

Now via Noethers theorem there is a deep connection between conservation laws and space-time symmetries but that is whole new story.

There is nothing non local about it unless you insist on properties regardless of observation. The nature of the non locality is very interpretation dependent. BM is the best known one - but there are others:
http://arxiv.org/pdf/quant-ph/9508021.pdf

Thanks
Bill

 

[morrobay]

Would the correlations of entangled photons described above
produce outcomes sufficient to violate an Aspect type Bell inequality experiment ?
A(aλ) = ± 1
B(bλ) = ± 1
S ≤ 2
S = E(a,b) + E(a,b') + E(a'b) - E(a'b')

 

[zonde]

Would the correlations of entangled photons described above
produce outcomes sufficient to violate an Aspect type Bell inequality experiment ?

If you consider only measurements that produce perfect correlations you can't get Bell inequality violations.

 

[zonde]

There is nothing non local about it unless you insist on properties regardless of observation.

This is upside down. If you do not consider common past as possible explanation for correlations then you have to invoke non-locality as an explanation.

 

[stevendaryl]

You can have locality and realism (CFD) with retrocausality. Of course the CFD is irrelevant given the outcomes and settings are part of the ontological explanation.

I would say that with retrocausality, you can have realism without CFD. CFD implies that there is a meaningful answer to the question: What result would I have obtained, if I had chosen to measure something else? But in a retrocausal model, I wouldn't think that such questions would be meaningful, in general.

 

[RUTA]

I would say that with retrocausality, you can have realism without CFD. CFD implies that there is a meaningful answer to the question: What result would I have obtained, if I had chosen to measure something else? But in a retrocausal model, I wouldn't think that such questions would be meaningful, in general.

Exactly, that's what I meant when I said CFD is "irrelevant" in retrocausality. It turns out many of the problems facing dynamical explanation per the Newtonian Schema in the mechanical universe are irrelevant when using adynamical explanation per the Lagrangian Schema in the block universe. That was the leitmotif of my Insights series https://www.physicsforums.com/insig...ions-part-1-time-dilation-length-contraction/ . I'm very glad to see someone is understanding the 4D perspective You may not like it, but if you at least understand it, you can appreciate its explanatory power.

 

[Eye_in_the_Sky]

If the "correlation phenomenon" is an 'effect' in spacetime, then there is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime

The existence of a beable is not necessary. Bohr, Ulfbeck and Mottelson's genuine fortuitousness claims the events are uncaused.

I have asked you whether or not you agree with the "If/then"-statement, and, instead of answering YES or NO, you have answered me that the "If"-part is not necessarily true.

So, I will ask again:

IF

there is no violation of 'causal locality' ,

and

the states of their instruments bear a relation of 'separability' throughout the spacetime regions A and B ,

and

the "correlation phenomenon" is an 'effect' in spacetime ;

THEN

There is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime

 

[Eye_in_the_Sky]

The correlation is simply the result of how you set things and usually a consequence of conservation laws.

Bhobba, from these words of yours it is unclear whether or not you are saying the "correlation phenomenon" is CAUSED or UNCAUSED.

... Which INTERPRETATION are you espousing?

 

[Zafa Pi]

Zafa, please give some examples if you can. I can't get clear what you mean.

Why is the speed of light the value we've measured it to be?
Why do masses attract?, or in GR terms, why does mass affect the geometry of space in the first place?

 

[RUTA]

I have asked you whether or not you agree with the "If/then"-statement, and, instead of answering YES or NO, you have answered me that the "If"-part is not necessarily true.

So, I will ask again:

IF

there is no violation of 'causal locality' ,

and

the states of their instruments bear a relation of 'separability' throughout the spacetime regions A and B ,

and

the "correlation phenomenon" is an 'effect' in spacetime ;

THEN

There is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime

My answer stands.

 

[Zafa Pi]

I have asked you whether or not you agree with the "If/then"-statement, and, instead of answering YES or NO, you have answered me that the "If"-part is not necessarily true.

So, I will ask again:

IF

there is no violation of 'causal locality' ,

and

the states of their instruments bear a relation of 'separability' throughout the spacetime regions A and B ,

and

the "correlation phenomenon" is an 'effect' in spacetime ;

THEN

There is a 'nonlocal beable' that acts within spacetime as its 'cause'.

... Can we agree on that?
______

beable: that which can bring about an 'effect' in spacetime

Is it true that: If 1 + 1 = 3, then I'm the Pope? The answer is yes, but a better answer is RUTA's, i.e. the If is false.
See: http://cds.cern.ch/record/980036/files/197508125.pdf

 

[bhobba]

Bhobba, from these words of yours it is unclear whether or not you are saying the "correlation phenomenon" is CAUSED or UNCAUSED.

They are correlated because they involve bell states. Cause is not the usual word for initial conditions.

I am not espousing any interpretation. I am simply asking you to delve into the detail of Bells theorem. Its all there.

Thanks
Bill

 

[bhobba]

This is upside down. If you do not consider common past as possible explanation for correlations then you have to invoke non-locality as an explanation.

They are correlated because they are bell states. How the bell states are created is an experimental matter - but usually involves some kind of conservation law.

Common past - beats me what you even mean.

Thanks
Bill

 

[Eye_in_the_Sky]

Is it true that: If 1 + 1 = 3, then I'm the Pope? The answer is yes, ...

Oh, come on, Zafa Pi. The "If" of your statement is demonstrably false. But to say:

the "correlation phenomenon" has a 'cause'

– is that demonstrably false?

I think not.

As far as I can tell, the contention that the "phenomenon" is 'caused' belongs the realm of 'possibility'. So, all I am doing is raising the question:

What does that 'possibility', if true, entail?

 

[RUTA]

Oh, come on, Zafa Pi. The "If" of your statement is demonstrably false. But to say:

the "correlation phenomenon" has a 'cause'

– is that demonstrably false?

I think not.

As far as I can tell, the contention that the "phenomenon" is 'caused' belongs the realm of 'possibility'. So, all I am doing is raising the question:

What does that 'possibility', if true, entail?

Turning on the source "causes" the correlated detector outcomes. However, that does not entail the existence of a beable. That was my point.

 

[morrobay]

Oh, come on, Zafa Pi. The "If" of your statement is demonstrably false. But to say:

the "correlation phenomenon" has a 'cause'

– is that demonstrably false?

I think not.

As far as I can tell, the contention that the "phenomenon" is 'caused' belongs the realm of 'possibility'. So, all I am doing is raising the question:

What does that 'possibility', if true, entail?

As bhobba said in # 127 " The correlation is ... usually a consequence of conservation laws "

http://arxiv.org/pdf/quant-ph/0407041.pdf

 

[morrobay]

If you consider only measurements that produce perfect correlations you can't get Bell inequality violations.

https://www.physicsforums.com/threads/murray-gell-mann-on-entanglement.884743/page-5
See post #87. My understanding of perfect correlations is that they only apply when detectors A and B are aligned,
I see no reference on this in post #87.
Also it is stated that A only considers the 50% of photons that are polarized in Φ direction.
My question would apply to all photons detected.

* I would have replied earlier but for unknown reason you and two others were incorrectly put on ignore list.
:

 

[zonde]

See post #87. My understanding of perfect correlations is that they only apply when detectors A and B are aligned,
I see no reference on this in post #87.

This sentence:
"As it turns out, if you consider only those B photons for which A found polarization in direction $|\phi \rangle$, then B will always find polarization in direction $\phi+\pi/2$."
Polarizers at relative angle $\pi/2$ will give perfect correlations.

Also it is stated that A only considers the 50% of photons that are polarized in Φ direction.
My question would apply to all photons detected.

If light is unpolarized then 50% of photons will pass polarizer at any angle. If you want to consider all photons you have to use polarization beam splitter (PBS). It has two outputs: one where you get H-polarized photons and the other one where you get V-polarized photons.

I am not sure I understand your question. Do you ask if we can violate Bell inqeuality with polarization entangled photons? If this is the question then the answer is certainly yes as most of the Bell test experiments are using these.

 

[morrobay]

Am aware that entangled photons violate the inequality. It is the correlations/outcomes without superluminal signaling - explanation
I am interested in. As vanhees71 is describing in post 87.
If you or anyone else can reference another post by vanhees71 in that topic that makes a better example of this I would like to see it.

 

[zonde]

It is the correlations/outcomes without superluminal signaling - explanation I am interested in.
As vanhees71 is describing in post 87. If you or anyone else can reference another post by vanhees71 in that topic that makes a better example. of this I would like to see it.

Ok, then I am not the right person to answer your question. I can only provide arguments why general explanations without superluminal signaling can't work.
Maybe @vanhees71 himself can help you?

 

[Eye_in_the_Sky]

As bhobba said in # 127 " The correlation is ... usually a consequence of conservation laws "

http://arxiv.org/pdf/quant-ph/0407041.pdf

Let's see, here. We are putting these two in juxtaposition:

"the correlation" | "conservation laws" .

On top of that, we are saying that the 'phenomenon' (to the left) and the 'principles' (to the right) bear some kind of a relationship. At the very least, we can say they have a 'logical' relationship of 'implication':

conservation laws → the correlation .

But beyond this, it does not seem correct to me to say that the two can bear a relationship of 'causation'.

Besides, here we are asking about matters of 'causation' between entities (like the instruments of Alice and Bob, or the instrument of preparation) that 'act' in spacetime.

 

[morrobay]

http://arxiv.org/pdf/quant-ph/0407041.pdf

Did you read the paper ? In particular page 4 and 5. If so then what exactly are you questioning ?

 

[Eye_in_the_Sky]

I am not espousing any interpretation. I am simply asking you to delve into the detail of Bells theorem. Its all there.

Ok. I will try.

 

[Eye_in_the_Sky]

If you do not consider common past as possible explanation for correlations then you have to invoke non-locality as an explanation.

Common past - beats me what you even mean.

Zonde means the overlap of the backward light-cones of the spacetime regions A and B.