🚀 go-pugleaf

In article <1992Aug11.220546.5574@sei.cmu.edu> firth@sei.cmu.edu (Robert Firth) writes:
>In article <mcirvin.713566546@husc8> mcirvin@husc8.harvard.edu (Mcirvin) writes:
>
>>The commuting of observables at spacelike separation prevents
>>spacelike communication via quantum entanglement precisely
>>*because* the measurements act like eigenstate projectors.  An
>>eigenstate of one such observable is always in an eigenspace of
>>the other, so the order in which the projections occur doesn't
>>matter, and there's no way one measurement can affect the result 
>>of the other.
>
>But that's simply not true.  In the straightforward Aspect type of
>experiment, the observables commute, but the results of the
>measurement are correlated, precisely because either measurement
>projects the total wavefunction into an eigenstate of BOTH
>observables.

The point in the no-signalling theorems is that while QM does certainly
violate "outcome-independence" locality, in that the outcome of a measurement
at one detector may depend "nonlocally" on the outcome of a measurement at
the other detector, it SATISFIES "parameter-independence," i.e., no
expectation values at one detector depend nonlocally on the parameters
(that is Bell-speak for "the knobs I have to play with") at the other
detector.  A communication scheme requires that what you receive depends
on what I choose to send, not on what random bit I happen to receive.
The exception to this is the recent proposals for quantum public key
distribution (not the bizarre radio-wave idea on the net, which I've been
ignoring, but one which appearing in a PRL by Artur Ekert, I think), in
which the entire goal is to send two different people the same random
number.  Even in his scheme (although only for the purposes of avoiding
any wiretaps), the process is not complete until the two guys get together
(I should say "two beings," as they always seemed to be named "Alice and
Bob") on some local communication channel and discuss matters.  Sort of
like a quantum version of bridge.

>This seems to me to be the flaw in most criticisms of Mr Sarfatti's
>experiment: they assume that the wave functions are decoupled, or,
>alternatively, that a measurement of one observable cannot affect
>an observable of the "other" particle.  This is circular reasoning:
>it is precisely that assumption that leads to Bell's inequality.

It's not only not circular, it's right!
(My opinion, anyway.)

-- 
Aephraim M. Steinberg                    | "WHY must I treat the measuring
UCB Physics                              | device classically??  What will
aephraim@physics.berkeley.edu            | happen to me if I don't??"
                                         |       -- Eugene Wigner