From:
c.h.thompson
To:
I.C.Percival
Cc: "Wolfgang
Tittel";"Nicolas Gisin"; "Valerio Scarani"; "Hugo
Zbinden"
Subject: Is
this going too far? quant-ph/0007009,
0007008 and 0009055
Date:
Thursday, September 14, 2000 9:45 AM
Dear Ian
I was
looking last night for examples for you of cases in which the experimenter did
an EPR-type experiment but said that the loopholes didn't matter because they
were only using the Bell test statistic as a figure of merit. I remembered the pair of papers I'd read
recently re work in Geneva. Do look
them up (quant-ph/0007009 and 0007008).
I haven't seen the most recent, of course, but 0007009 is really, I
think, going too far. I didn't realise
when I sent off my letter to New Scientist
quite how serious were its problems! (I
sent copies of the letter to the experimenters concerned.)
The phrase
I was looking for was on page 11: "Note however, that hidden variables are
no issue in this work". I now see
why. What they are testing is the
totally ridiculous hypothesis that the "stop" output channels of the
interferometers somehow affect the results!
The idea of EPR correlations
pales into insignificance compared to this assumed blatant infringement of
causality.
So far as I
am concerned, the experiment includes at least 4 other major absurdities:
(a) It
depends on a nonsensical application of SR
(b) It
depends on some rather dubious theory about chromatic dispersion, presumably
the same that features in an article by Gisin et al (quant-ph/9901043) that I
meant to write to the authors about.
(The truth is, I suspect, that each pair of pulses is purely one
frequency so dispersion affects both equally.
The important thing is that we are dealing with the degenerate case of
PDC and the formula concerning conjugate frequencies is irrelevant.)
(c) Re
validity of any Bell test: one interferometer is kept fixed, so the experiment
is open to the possibility of rotational invariance failure (see my paper
9912082)
(d) They
also subtract accidentals (see 9903066)
Gisin's
abstract to 0009055 is:
"Past, present and future experimental
tests of quantum nonlocality are
discussed. Consequences of assuming that the
state-vector collapse is a real
physical phenomenon in space-time are
developed. These lead to experiments
feasible with today's technology."
But the
hypothesis is illogical! This is worse,
much worse, than throwing away money investigating a perpetual motion
machine. "Science" is surely
not in the business of investigating predicted magical effects before they've
even happened? Yes, I suppose they had
to investigate the EPR correlations, but to carry on from there, before they
have been proved to happen, and investigate the consequences of theories that
depend on the combination of EPR effects and yet more magic is ridiculous. There is no limit to the number of magical
effects that do not happen.