🚀 go-pugleaf

In article <1992Oct01.193934.129902@watson.ibm.com>, platt@watson.ibm.com (Daniel E. Platt) writes:
> |> Bell's result depends only on the macroscopic experimental
> |> predictions of QM. ...
> 
> It isn't 'action at a distance.'  It is correlated information (while
> not deterministic or complete) about what two isolated observers could
> observe.  It doesn't matter whether the observation is space-like or
> time-like, and is not a causal connection.

I am afraid that most physicists who have not looked into
the details of this believe something like you just stated.
I know you have spent some time looking at the details so I am a little
surprised by your remark. Lets examine what is happening.

We have an experimental something like the following:

    Location A                                     Location B
    Detector  Polarizer        Photon Source       Polarizer  Detector
       |--------\------------------O------------------/----------|

Quantum mechanics predicts that the probability of a joint detection
is a function of the angle between the polarizers at A and B. It is the
nature of this function that is the critical element. One cannot reproduce
the predictions of quantum mechanics without this difference in angles
*affecting* the results at one of the two locations. That is one cannot
write down a mathematical expression using hidden variables
or anything else that produces these predictions of quantum mechanics without
including the angle difference in the equation that describes the 
probability of a joint detection. This means you cannot model the
predictions of quantum mechanics unless information about one of the
polarizer angle influences results at the other location. By the conventional
classical definitions the following are true.

Either special relativity or causality is violated.

Action at a distance is involved i. e. by changing a polarizer angle at
one site you can change the results at a distant site. 

Information is transferred instantly. The information about the angle
at one of the sites must influence the detection at the other site.

Knowledge is transferred instantly. You know something about the
combination of polarizer angles and detections at the remote site.

It is true that the probability distribution of the individual detections
is not changed by distant manipulations. It is only the joint probability
distribution that is affected. If you define causality, as John Baez advocated,
as being based only on these local distributions you escape violating causality.
This seems fraudulent to me. Quantum mechanics not only predicts individual
probability distributions it also predicts joint distributions. If you are
talking about causality you must take into account all predictions.
It is true that the state vector evolves in a causal fashion. The predictions
of the quantum mechanics are either not causal or not Lorentz invariant.

>...
> There's lots of physicists who don't subscribe to the postulate as stated.
> What is inappropriate is assuming every physicist accepts non-unitary collapse
> as a mechanism, separate from they dynamic Hamilton/Schroedinger equation,
> that can modify the state vector.

I did not say that and I do not think that.

>...
> Here, you'll find that I agree with the predictions.  The point we don't
> seem to agree on is whether the observation of correlated spins over
> a space-like interval requires a transmission of information between
> the observers.  I claim it doesn't.  So does any physicist who does not
> subscribe to collapse.

Observing correlations over space-like intervals does not necessarily involve
instantaneous information transfer. In those cases in which Bell's inequality
is violated it must. This depends only on the predictions of quantum
mechanics, not on the interpretation or the mathematical model used to
compute the predictions. There is another paper by Eberhard (besides
the one I sent you a long time ago) that I urge you to read on this point:
P. H. Eberhard, Il Nuvo Cimento 46 B, 392, (1978). If you have any reference
that makes the case against action at a distance I would be happy to check
it out.

>...
> The only issue is collapse, and the objective reality of collapse.

Nonsense! There are several issues. The main one here is the observable
predictions of quantum mechanics.

>...
> This
> seems to remain an open question.  In a way, you're in a strange position
> trying to convince me that quantum mechanics requires collapse so that
> you can shoot it down, while I don't believe in collapse (at least, as
> described by Dirac or by Cohen-Tanoudji or by Sakurai).

I am not trying to convince you that quantum mechanics require collapse.
I happen to believe in a local version of collapse, but that is not the
issue here. It is the experimental predictions of quantum mechanics that I am
calling preposterous.

>...
> |> I mean objective reality in the sense that Einstein used the term. The
> |> current formulation of quantum mechanics is an incomplete description
> |> of an underlying objective reality. I do not think all physicists reject
> |> Einsteins approach. Most seem to. 
> 
> How substantial is something that eludes our ability to achieve consistent
> description?  This is a question of epistomology.

Maybe the problem is with your `ability to achieve consistent description'
and not with the existence of objective reality. I think a radically different
class of models is called for and it is going to take hard work
by many people at both the experimental and theoretical end
to achieve such an explanation. That would make it a hard physics problem
not a problem in philosophy.

> ... But I still have
> one question for you:
> 
>     What is the Hamiltonian for measurement?  Or, what is your dynamic
>     description of measurement?  Or, how do you interpret the results
>     of QM if you believe that most of the results of QM will be preserved?

My recent answer to Daryl's question about the multiparticle state vector
provides a partial answer to these questions. I do not pretend to have
a complete theory. What I claim is that quantum mechanics makes preposterous
predictions that cannot be justified. Physicsits need to be clearer about what
they do understand and what they do not understand. They do not understand
the nonlinear changes that occur when an observation is made.

Paul Budnik