🚀 go-pugleaf

In article <717712154.20@mailbox.bison.mb>, Paul.Budnik@mailbox.bison.mb.ca (Paul Budnik) writes:
|> Organization: Mountain Math Software, P. O. Box 2124, Saratoga. CA 95070
|> 
|> Look up at the sky this evening. Photons from stars and galaxies
|> as distant as millions and even billions of light years will strike
|> your eye. Many of these will, according to quantum mechanics, be in
|> a singlet state, paired with another photon. These two were created
|> in a single decay event and have been journeying through space in 
|> opposite directions for eons. Yet quantum mechanics maintains that
|> there are still connected. Put on a pair of polarizing sun glasses
|> (not ordinarily recommended for star gazing). Now as you turn your head
|> and change the angle of the polarizer you will instantaneously influence
|> all the singlet state photons that are paired with the ones
|> you are observing. This is true even though these paired photons may
|> be billions of light years distant. At least that's what quantum mechanics
|> implies.
|> 
|> Suppose there is an alien star gazer on a distant galaxy. By chance she
|> has on polarizing sunglasses and glances at a photon paired with one
|> you see. (An unlikely event, but this is a thought experiment.) The relative
|> angle of the polarizers of these two distant star gazers will affect
|> in an experimentally detectable way the probability that they will both
|> detect the two photons in a singlet state. 

This is not entirely accurate.

If the polarizers were aligned, then no events would be detected that would
violate conservation of angular momentum (which is what one would hope to
observe).  If the polarizers were not aligned, then it would be impossible
to see if the measurements agreed or not.

The problem is that, if the state vector is a real physical thing instead
of a descriptive construct, then it implies that superpositions of spin states
must be selected by collapse of the state vector upon measurement -- over
space-like intervals.  However, if you don't even collapse the state vector
and compute cross-sections for mixed measurements (those that violate the
correlations imposed by the production process), you will find that the
cross-section is zero... even if you don't 'collapse' the state vector.


|>                                            This would not be surprising
|> if it were because of a polarization in the photons that was determined when
|> they separated in the distant past. However this is not the case. Quantum
|> mechanics predicts that the polarization for the singlet state
|> is determined at the instant one photon is observed. 

Depends on whose development of QM you read.  In part, the above argument
implicitly assumes that the state vector must ultimately convey the information
of what was measured by each observer to the other observers.  Actually,
the only information it is carrying is what cannot be observed by both
observers -- even over space-like intervals.

<stuff deleted>

|> Why would serious physicists believe this seemingly preposterous prediction?

It depends on who you talk to.  Obviously, not everyone agrees with an 
interpretation that leads to the problems you are challenging.


|> The theoretical principle on which this prediction is made is ... (sigh, I
|> cannot even give it a name without creating controversy). You see this
|> principle is so ill defined that it would be difficult to get two physicists
|> to agree on exactly what it is. Let us call it the observation principle.

Here, you are correct.  There's lots of noise in the signal surrounding this
point.  Different people have treated it in different ways.  One of the more
common ways is stated in Dirac; he does not cite background for the argument.
Cohen-Tanoudji et al followed suite in their quantum text.  It is not quite
the same way that Feynman or Schiff have discussed the same material.  Then,
there's intermediate positions, which is sort of where I'd put Merzbacher.
My tastes tend to lean towards Feynman and/or Schiff's descriptions.

|> It states that the wave function we use to make predictions is different
|> after an observation is made. Physicists cannot agree on whether the
|> wave function changes or we just select a different wave function that
|> was there all the time. 

What do you mean a wave function (state vector?) 'was there all the time'?
Here, again, the assumption is made that the state vector must ultimately
carry the information of what measurement was ultimately made to all observers.

|>                          But in any event we must change the model we
|> use to make predictions after an observation and this model changes
|> instantaneously when the observation is made throughout all space
|> that is relevant to an experiment. This raises some interresting questions
|> regarding special relativity on which there are about as many opinions as
|> there are physicists that have thought about these issues, but all
|> physicists agree on what the predictions are. They just have
|> no consistent theoretical basis for making them.

That's not entirely true.  There's several levels of premises upon which
this problem is based.  Some of these include: 
	o	How 'real' is the state vector?  
	o	What happens to a state vector upon measurement (this
		question is strongly influenced by how the first 
		question was answered)?
	o	Does this involve the actual transfer of information
		between the two photons?

|> 
|> Well, if there is no experimental evidence in support of these preposterous
|> predictions and there is no well defined theoretical principle on which
|> they are based, then *surely* there is no way to change existing theory
|> so that all the thoroughly verified predictions of quantum mechanics are
|> intact but these preposterous ones are gone. Not exactly.

"Preposterous?"  Sounds a tad emotional... the word keeps coming up...

One of the elements in this line of argument is that it fails to
recognize that there is a spectrum of responses that have been
taken to the assumptions listed above.  Some of them are better
than others.  I find it a bit surprising given that it was
acknowledged that physicists don't seem to agree on all the points
of interpretation.  Particularly, you can have two physicists see
the same results, say they agree on what the results were, and then
disagree on whether it means the photon state vectors collapsed.

<stuff that's starting to degenerate into repetativeness deleted>

|> 
|> Well if there is absolutely no reasonable experimental or theoretical
|> foundation for these preposterous predictions why, in heaven's name,
|> do physicists make them?

Physics, for all that it is an attempt to describe and work with
objective observation, is after all a human endeavor since it is
humans doing the observing and analysing in the end.  Its a hard
problem; physicists have argued about it for 70 years.  Even
debates involving some of the most respected physicists in the field
at Copenhagen failed to close the question (which obviously still
persists).  Some of it involves the appeal to authority (Dirac has
makes such a nice logical presentation; not only had he obtained a
very important position on the basis of his work, but his text on
quantum mechanics was very well presented and written; it is reasonable
that the position he adapted would be communicated and accepted; however,
I'd been warned not to read it until I'd looked at several developments
particularly because of this point).

|> 
|> I suggest two answers. This first is best understood through
|> the Hans Christian Anderson fairy tale *The Emperor's New Clothes*.

This would be fair if everyone agreed, and nobody argued.  As it stands,
most people argue, and not everyone agrees.  Consensus has not been
achieved -- for what ever reasons.  This implies that most people are
aware of the problem with the clothes; however, most people have 
acquired some degree of comfort or acceptance of the situation.

|> 
|> You see it was Einstein's (and his coauthors') paper, EPR, that started
|> the line of inquiry that ultimately led to Bell's result. Poor
|> befuddled Einstein thought there must be such a thing is objective
|> reality. He thought than quantum mechanics was an incomplete theory.
|> He really never understood `modern physics'. Unfortunately for it, the
|> non preposterous (local Lorentz invariant collapse) version of quantum
|> mechanics is an explicitly incomplete theory. It implies that the wave
|> function has an objective physical reality.

Does it?  That's one interpretation... not one that's universally held.

|>                                              No self respecting `modern'
|> physicist would want to be accused of believing in objective reality or
|> failing to understand the new deep philosophical principles that quantum
|> mechanics has revealed. (Fortunately there are notable exceptions to this,
|> but I think the lesson from this fairy tale is clear.)

Wouldn't they? I expect there are quite a few who accept objective reality.
There was a survey taken via the net recently that seemed to indicate most
physicists believed in objective reality.  (Even then, there was lots of
argument over what that 'meant'.)  On the other hand, there was a fair
amount of argument over whether state vectors were real, or whether their
description was substantially related to the object they were describing.

|> 
|> The work of Kuhn provides the second answer. Thomas S. Kuhn was educated
|> as a physicist, but has spent his life as a historian of science. In
|> his investigations he concluded that far from seeking revolutionary new
|> ideas scientists abhor them. For if you are professional physicist, a
|> major revolution in physics is likely to be disruptive to your career and
|> thus your personal life. Being a human being first and physicist after that
|> major revolutions in your field are real personal threats. The more stature
|> and prominence you have obtained the greater the threat.

Kuhn looked at the quantum and relativistic revolutions.  The resistance was
substantial at that time... for several reasons.  When Einstein was in school,
he was frustrated because Maxwell's equations were not being taught.  There
was also a sort of Victorian mentality that the foundations of everything
had been more or less worked out.  While many of the old school didn't go
along for the ride, there were quite a few of the 'old school' who did
advance and change.  The "revolution" took 30 years (starting at
1890, and running through to the 20's -- with lots of activity lasting longer
than that).  As I grow older, I become painfully aware that that is the
length of a productive career.  The impact of that is that surely, the people who
started the revolution will probably be done with their careers by the time
the revolution is over.

On the other hand, the above is not a dismissal of Kuhn's ideas... just pointing
out some of the historical context.  Besides, there's been other examples...
concommitant with the revolution in phyiscs has been the revolution in biology
and genetics.  There's also the revolution in geology (plate tektonics).

To put this in perspective, in some sense, many of the questions that emerged in the
last revolution in physics persist -- people are still arguing about collapse.
However, even pre-revolution questions have been hot topics -- such as questions
involving statistical mechanics and ergodic theory.  These questions become
more and more important as molecular dynamics and monte-carlo simulations are
done.

|> 
|> Just think about what will follow when they finally do the experiment
|> correctly and observe those pesky little time delays that preserve locality. 
|> Much of leading edge theoretical physics of today will be out the window.
|> If the wave function is an objective physical entity that is consistent with
|> locality, an entirely new approach will be needed. High energy physics
|> will still be of some interest, but the focal point of leading edge research
|> will shift to understanding the space-time structure of quantum collapse.
|> The priority of high energy physics will decline dramatically. Billions
|> will no longer be available for building superconducting-supertoys. For
|> there will now be an experimental window opened to an entirely new class
|> of physical phenomena. Unlike the events observed in particle accelerators
|> that have little or no practical significance, quantum collapse is
|> a universal process. Every time every transistor in a computer chip changes
|> state, quantum collapse is involved. Understanding this process will be
|> of the highest theoretical and the highest practical importance.
|> 

Assuming collapse occurs as you describe it.  State flips in computer chips
don't really involve that question, however.  That's an interesting point in
that most of the research involving QM does not explicitly involve collapse one
way or the other.  The problems studied are much more practical, and have
been very relevent without appeal to collapse (including most of the physics
of semiconductors).  I suspect that this is the reason the question has 
persisted so long without the emergence of some consensus -- it isn't easily
measured; most of the time it is irrelevent; it is likely not going to change
the utility of the results of QM as currently developed.

Dan