No drama quantum electrodynamics? (was: Local realism ruled out?)

[Bryan Sanctuar]

Bell's view of breakdown of local causality

Where have you seen this stated about Bell's position on non-locality? As far as I have read, he felt Bohmian type theories were good candidates.

You can read this point of view in a paper by T. Norsen Local causality and completeness: Bell vs. Jarrett

http://arxiv.org/pdf/0808.2178.pdf

 

[akhmeteli]

That looks very interesting, as the experimental results to date suggest to me that perhaps no experiment can be done that proves or disproves "quantum non-locality" - a bit like the original formulation of the PoR. Thanks!

Thank you for your interest.

 

[akhmeteli]

OK that clarifies it a little to me. However, I remain puzzled (surely not in the least because spinors and Fock space go over my head!):

You don't need to now anything about spinors to understand the model based on scalar electrodynamics. The model based on spinor electrodynamics is more realistic, but more complex. However, the logic is pretty much the same for both models. The Fock space... Well, you do need to know something about the Fock space to understand some aspects of my work. However, you need to know something about it to understand quantum field theory anyway.

What does that mean in practice, concerning expected predictions with your model? Similarly, following DevilsAvocado simplification that "classic says 1+1=2 and QM says 1+1=3"(sic), where will the predictions of your model fit in, for a typical "non-ideal" Bell experiment? You seem to suggest in your latest paper that you expect results that are close to that of standard QM.

Well, you see, the model of my work is based on spinor electrodynamics. This is a complex nonlinear theory, so it is not easy to derive specific predictions for specific experimental setups. However, this model is reasonably realistic, as it includes spinor electrodynamics, which is a decent theory, so it is indeed reasonable to expect that predictions of this model will be reasonably close to those of quantum electrodynamics. For this reason one can hope that the model's predictions would be close to the experimental results of typical "non-ideal" Bell experiments. This possibility is not eliminated by the Bell theorem, although the model of my work is local realistic, as those experiments are not loophole-free. Let me also add that the model of my work may require some modification to provide predictions that are closer to those of quantum electrodynamics.

Maybe the above is overcomplicated, so let me rephrase it "in DevilsAvocado's terms": QM says 1+1=3 only if you start with the two mutually contradictory assumptions of QM, and that is not a good way to start anything anyway.

Let me clarify in this context my comparison with the PoR.
The null result of MMX to verify wave theory would have been interpreted to prove ballistic light theory if such had not been at odds with earlier results from other types of experiments. Then a "loophole" was found: length contraction. And later as a follow-up, the null result of KTX could again be interpreted to prove ballistic light theory, as length contraction is insufficient to compensate the expected effect. However, another "loophole" had already been found: time dilation. Is that a conspiracy of Nature to save the relativity principle? Only if one considers the conservation laws a "conspiracy".

Something like that (although I had difficulties with your abbreviations: I guess MMX is Michelson-Morley experiment, PoR is Principle of relativity, and KTX is Kennedy-Thorndike experiment).

PS: I would appreciate it if you could elaborate on the following statement:

if you use the "EPR definition" of local realism, then classical electrodynamics is not local realistic. The models of my articles are similar to classical electrodynamics and are not "EPR realistic" either - the statement "there are definite values for spin components at all times" is not correct for them, but they are no less realistic than classical electrodynamics.

How is classical electrodynamics not local realistic in that sense?

I thought I explained that, but let me try to rephrase that. As far as I understand, under EPR definition, realism assumes that observables have definite values irrespective of any measurement, however, classical electrodynamics does not seem realistic in this sense, as electromagnetic field is typically distributed, so it does not have, say, definite coordinates, independent of any measurement, but, depending on your instrument, you can observe the field in some point, which will depend on the instrument, so classical electrodynamics is "contextual" in this respect.

 

[DrChinese]

You can read this point of view in a paper by T. Norsen Local causality and completeness: Bell vs. Jarrett

http://arxiv.org/pdf/0808.2178.pdf

This paper states the exact opposite of your statement. Bell felt non-local theories to be viable. In case you weren't aware, Norsen is a Bohmian.

 

[harrylin]

[..] Well, you see, the model of my work is based on spinor electrodynamics. This is a complex nonlinear theory, so it is not easy to derive specific predictions for specific experimental setups. However, this model is reasonably realistic, as it includes spinor electrodynamics, which is a decent theory, so it is indeed reasonable to expect that predictions of this model will be reasonably close to those of quantum electrodynamics. For this reason one can hope that the model's predictions would be close to the experimental results of typical "non-ideal" Bell experiments. This possibility is not eliminated by the Bell theorem, although the model of my work is local realistic, as those experiments are not loophole-free. Let me also add that the model of my work may require some modification to provide predictions that are closer to those of quantum electrodynamics. [..]

OK

[..] I had difficulties with your abbreviations: I guess MMX is Michelson-Morley experiment, PoR is Principle of relativity, and KTX is Kennedy-Thorndike experiment).

Yes indeed - sorry for that!

I thought I explained that, but let me try to rephrase that. As far as I understand, under EPR definition, realism assumes that observables have definite values irrespective of any measurement, however, classical electrodynamics does not seem realistic in this sense, as electromagnetic field is typically distributed, so it does not have, say, definite coordinates, independent of any measurement, but, depending on your instrument, you can observe the field in some point, which will depend on the instrument, so classical electrodynamics is "contextual" in this respect.

OK, meanwhile it's also getting clearer to me from other papers - apparently, Bell's interpretation of EPR is that "realism" means that all properties are pre-existing and fixed, independent from measurement. Indeed, EM fields do not adhere to such a concept.

 

[Ilja]

As far as I understand, under EPR definition, realism assumes that observables have definite values irrespective of any measurement,

Completely wrong.

It follows from EPR realism and from locality (more accurate, from Einstein causality) and from the 100% correlation predicted from quantum theory that those special observables which have such 100% correlations between nonlocal observations have to have definite values before measurements.

however, classical electrodynamics does not seem realistic in this sense, as electromagnetic field is typically distributed, so it does not have, say, definite coordinates, independent of any measurement

Also completely wrong. Of course, there is no such observable as "the coordinate of the EM field", thus, there is also no such value of the classical EM field.

But, independent of what defines the classical EM field - or the vector fields E_i(x,t), H_i(x,t), or the corresponding potentials A_μ(x,t), these fields at a given moment t define every observable of classical EM theory which can be measured at a given point x.

 

[akhmeteli]

Completely wrong.

It follows from EPR realism and from locality (more accurate, from Einstein causality) and from the 100% correlation predicted from quantum theory that those special observables which have such 100% correlations between nonlocal observations have to have definite values before measurements.


Also completely wrong. Of course, there is no such observable as "the coordinate of the EM field", thus, there is also no such value of the classical EM field.

But, independent of what defines the classical EM field - or the vector fields E_i(x,t), H_i(x,t), or the corresponding potentials A_μ(x,t), these fields at a given moment t define every observable of classical EM theory which can be measured at a given point x.

Thank you. I will try to reply in a couple of days.

 

[akhmeteli]

As far as I understand, under EPR definition, realism assumes that observables have definite values irrespective of any measurement

Completely wrong.

It follows from EPR realism and from locality (more accurate, from Einstein causality) and from the 100% correlation predicted from quantum theory that those special observables which have such 100% correlations between nonlocal observations have to have definite values before measurements.

I stand corrected. And sorry about "a couple of days" turning into a year:-(

What I should have said is the models of my article do not predict definite values before measurement for the conditions of the EPR (or EPR-B) experiment. For example, when we measure a spin projection for one particle of the singlet in the EPR-B, then, according to standard quantum theory, the spin projection of the other particle of the singlet becomes determinate with 100% probability immediately, no matter what separation between the particles of the singlet. This is not true for the models of my articles. However, this also contradicts unitary evolution of quantum theory, as unitary evolution cannot turn a pure state into a mixture.

classical electrodynamics does not seem realistic in this sense, as electromagnetic field is typically distributed, so it does not have, say, definite coordinates, independent of any measurement, but, depending on your instrument, you can observe the field in some point, which will depend on the instrument, so classical electrodynamics is "contextual" in this respect.

Also completely wrong. Of course, there is no such observable as "the coordinate of the EM field", thus, there is also no such value of the classical EM field.

But, independent of what defines the classical EM field - or the vector fields E_i(x,t), H_i(x,t), or the corresponding potentials A_μ(x,t), these fields at a given moment t define every observable of classical EM theory which can be measured at a given point x.

Again, I stand corrected, and this mistake is a direct consequence of the first one - the wrong idea of what is EPR reality.

So what should I have said instead? The models of my articles describe independent evolution of electromagnetic field. On the other hand, they describe electrons (interacting with the electromagnetic field) in the same time, as, say, the Klein-Gordon equation or Dirac equation is satisfied within the models, however typical observables of quantum theory, such as electron coordinates or spin projections, are typically not defined precisely before measurement (so the uncertainty principle is not violated). Whatever is the electron wave function in quantum theory, is a complex function of electromagnetic potential in the models of my articles.