Jack,
John found out about Dopfer's dissertation from a paper by Raymond Jensen that I sent him several years ago. I think what he's saying is that the purpose of the coincidence counting circuit in the experiment proposed (and done by Dopfer) is to allow you to discriminate entangled photons from those that are not entangled. So, it would seem, if you could produce a source that generates only entangled photons (or one with a sufficiently high proportion of entangled photons), the coincidence counting circuit would not be necessary to separate entangled photons from the "noise" of unentangled photons. In those circumstances it would seem that retrocausal signaling should be possible. This, I think, was Srikanth's view too. Given my iconoclast tendencies, I'm rooting for John. :-) If it works, it would substantiate his transactional interpretation.
Regards,
Jim
Subject: Anton Zeilinger's Rev Mod Phys paper on the Dopfer experiment: John Cramer's & Yakir Aharonov's back from the future physics? (Dr. Quantum)
In the early 1979 editions of the best-seller Dancing Wu Li Masters, later removed by Gary Zukav, I had double slits for both photons 1 and 2 with a Heisenberg microscope encoder/modulator at the sender to create and destroy fringes at the receiver similar to the scheme below. Of course I did not have the coincidence circuit and the standard unitarity-linearity no-signaling theorem says that there will be no local fringes observed without the coincidence post-selecting the sub-ensemble as Zeilinger shows in Fig 3 below. However, now John Cramer argues that the smearing-out effect of not post-selecting coincident photons 1 & 2 is only a 15% noise effect so that local fringes should be seen when the Heisenberg microscope encoder is switched off (i.e., Heisenberg detector in the focal plane of the lens in Fig 3) - at least for that particular parametric down-conversion pair source is concerned. However, I think actual experiments with double slits have been done without seeing local fringes when the coincidence circuit is switched off, but perhaps without that particular allegedly 85% efficient source?
The coincidence issue is automatic in the nano-scale solid state device envisioned in Lawry Chickering's 1982 letter to Richard de Lauer UnderSecty DOD Research & Engineering. That is, it is a non-issue because we automatically have pre-post selected sub-ensembles in Aharonov's sense.
Basically I suggested back then 30 years ago nanoscale systems like quantum dots today that would be clearly labeled like marbles in boxes.
We have N = N'n pairs of entangled quantum dots (ai,bi) i = 1 to N arranged in N' bins each of n pairs. "n" is large enough (a few dozen) to show the statistical difference in the patterns.
The N bi go on the Nuclear Submarine (Receiver), the N ai stay at Naval Command (Sender).
My point was that since the entangled individual pairs are mesoscopic and individualized and obviously protected against decoherence the sender can decide to do incompatible measurements P or Q where [P,Q] =/= 0 that will collapse the states of the n matched twins in each bin at the receiver either into a "coherent" or an "incoherent" pattern functionally analogous to the Dopfur experiment though quite different in implementation of course.
http://faculty.washington.edu/jcramer/PowerPoint/AAAS_20060621.ppt.
References:
Nonlocality:
A. Einstein, B. Podolsky, and N. Rosen, Phys. Rev. 47, 777-780 (1935).
The Dopfer Experiment:
B. Dopfer, PhD Thesis, Univ. Innsbruck (1998); A. Zeilinger, Rev. Mod. Physics, 71, S288-S297 (1999).