Tim Bradshaw wrote:
> Not at all. For the results of observations in QM the standard story
> is `it is random' -- or rather `all you can know in advance are the
> probabilities of the various outcomes'. And you can investigate this
> a whole lot -- for instance you can hypothesise that no, it's not
> random, but there is some `hidden variable' which determines the
> outcome of the experiment. And when you look at that hypothesis you
> find it doesn't make sense because the hidden variables must be
> `non-local' which basically means `causality violating'. And you can
> do experiments & theoretical work which explore this whole area, like
> the experiments done by Alain Aspect in the 80s and the theory stuff
> by Bell somewhat earlier (60s?).
Bell's inequation is related to `realistic local theories', which limits
the number of local theories quite a bit (the others are dismissed as
being `unrealistic'). The experiments of Aspect et al did rule out all
realistic local theories, but left a few `unrealistic' in place, because
of a small flaw in the final experiment.
There is the possibility, that the information of the polarizers' state
move to the source of the entangled state by some ordinary mean (thus at
light speed), and influences the creation of the entangled particles in
such a way that Bell's inequation is violated.
The only way to rule this theory out is a dynamic experiment, which
changes the settings of the polarizers while the entangled particles are
under the way. Aspect et al did that, but unfortunately, they published
a result with 6m distance source-polarizer and 50MHz switch time. Bummer
- this is exact the position where this slightly `unrealistic' local
theory still works (it's not a `realistic local theory', because it
violates Bell's inequation, but it's still a local theory, since all
interactions travel with light speed or less).
Anton Zeilinger, one of the most active experimentators in this field,
pointed out that flaw some time ago, and promised (in private
communication) to fix this flaw and redo the experiment with unrelated
switch/distance, but I haven't heard anything. There seem to be a lot of
more interesting experiments than redoing a flawed experiment, where
most people think this issue is closed.
"If you want it done right, you have to do it yourself"