cjh@petsd.UUCP (Chris Henrich) (10/27/84)
[]
1. Replay to Howard Hull:
Howard Hull's letter on photons is interesting but does
not touch on quite the same puzzle as I was trying to. I
think he has jumped from my suggestion of "bilocality" to
photons. If a correlation is bilocal, then it need not have
anything to do with photons or other particles of zero rest
mass. Indeed, the simplified EPR experiment can use two
electrons, whose combined state has spin 0; then they are
correlated, and the correlation becomes a bilocal entity if
they are separated.
Generally, I don't believe we know "what it is like" to
be a photon, except that the (ahem) subjective experience of a
photon has to be quite different from that of a particle with
rest mass. Indeed, there is a story that when Einstein was
about sixteen, it occurred to him that an observer who moved
apace with an electromagnetic wave would (assuming Newtonian
kinematics) see a stationary wave, which supposedly isn't
possible. From here to relativity is not so far.
"Orthogonal to its [i.e. the photon's] path..."??
The subspace orthogonal to the photon's path is a
three-dimensional space which *includes* the path of the
photon. Therefore it, plus the photon's path, do not span the
entire four-dimensional space-time. I didn't make this up; it
is a consequence of special relativity. The sense in which the
photon sees the rest of the universe remains obscure to me.
2. Some more grabs at the strangeness of QM
The simplified EPR experiment is, among other things, a
model of the process of measurement in QM. One of the
electrons can be regarded as a probe measuring the other one.
In any measurement process, the "sample" being measured and
the "probe" doing the measurement are brought into
interaction, then separated. After the interaction, we may
assume that the state of the sample&probe combination is pure,
but not a combination of a single pure state of each part.
Thus neither part, considered by itself, is in a pure state,
but rather in a mixed state.
Now, if you "throw away" the information in the probe,
then the sample remains in a mixed state, which may be
analyzed as being either UP or DOWN with probability 1/2.
But if you "look at" the data in the probe, and infer from it
the state of the sample, then you will, with probability 1/2,
find it to be the pure state UP, and with probability 1/2,
find it to be the pure state DOWN. It appears that the
physical condition of the sample is arcanely affected by the
action of the experimenter. Note, however, that there is no
way for this arcane effect to be observed without some mundane
data path from the observer back to the sample.
The mixed state is indistinguishable from a probabilistic
mixture of the two pure states.
3. A way around the philosophical puzzle.
There is an idea going around that quantum mechanics
brings the mind of the experimenter into the experimental
reality. It is commonly said that the "state" of a quantum
mechanical system is inseparable from the state of our
knowledge of the system. Somehow, our knowledge of it
constitutes the reality of it. I know that I am not clever
enough to make electrons, so I have never been happy with this
formulation. But it becomes much more acceptable if a level of
indirection is added:
What we know about a quantum-mechanical system
cannot be separated from what we know about our
knowledge of the system.
Of course, no-one will say "gee whiz!" to that...
4. A question addressed to physicists
Martin Gardner, in his interesting book _The_Whys_of_a_
Philosophical_Scrivener_, discusses the philosophical puzzles
of QM, and in passing makes a statement that very much
surprises me. It is that physicists, for the most part,
believe that the universe develops randomly; for instance, if
an electron is approaching a screen where it can be observed,
then its wave packet is reduced, and the choice of this
reduction is probabilistic. Hey physicists:
- DO you believe this?
- How do you reconcile it with the assumption that the
universe itself, as a Thing, develops deterministically,
with its wave function developing through time according
to the Schroedinger equation?
Regards,
Chris
--
Full-Name: Christopher J. Henrich
UUCP: ..!(cornell | ariel | ukc | houxz)!vax135!petsd!cjh
US Mail: MS 313; Perkin-Elmer; 106 Apple St; Tinton Falls, NJ 07724
Phone: (201) 870-5853hull@hao.UUCP (Howard Hull) (10/31/84)
In a recent posting, Chris Henrich replied to an earlier subjective posting
of mine. In order to advance the EPR (Aspect) discussion efficiently, I
wish to "tie off some of the loose ends" I seem to have produced.
I plead guilty to diverting the discussion from "bilocality" to photons.
Sorry folks, I just happen to like photons and use a lot of them in my
favorite recipes. I have to admit that even to my own definition, they
possess something more akin to "multilocality" than bilocality. I will
agree with Chris; if you don't need photons to discuss bilocality, then
don't use them. So I will select the option of losing this argument.
However, I wish to register a subjective complaint. If the form of the
wave equation you are dealing with allows linear superimposition, (and I
believe that it does), then I claim that the property of "correlation" is
a property of the observer [and the observer's processing of the observed
information] and not fairly a property of the observed phenomenon. To my
notion, two entities themselves posess the property of correlation only if
they are bound together by the elements of a non-separable expression (i.e.
a non-linear expression). Maybe I shouldn't use the word "correlation",
but if that isn't the word I should use, then I do not know what the correct
word is. The only thing I can think of that might serve as a "correlation
binder" in this case is the Pauli exclusion requirement - and that applies
only if the test objects are co-located particles of the same description,
but differing by one quantum number, and only for the space-time in which
they are co-located. Perhaps what you need to contain this problem is a
"dual test particle" with a single non-separable QM descriptor.
At least, the notion you propose - that one of the objects is a test for the
other - has a greater propensity to survive my particular conception of what
is going on than does a sense of linear equality for the objects. However, I
would be very careful concerning the way in which this notion enters into the
mathematical description. The remaining comments of your issue seem to be
very constructive, and will (fortunately for the net) keep me busy for some
time.
Howard Hull
{ihnp4!stcvax | decvax!stcvax | seismo} !hao!hull