home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.physics:19278 alt.sci.physics.new-theories:2362 sci.optics:1179
- Newsgroups: sci.physics,alt.sci.physics.new-theories,sci.optics
- Path: sparky!uunet!well!sarfatti
- From: sarfatti@well.sf.ca.us (Jack Sarfatti)
- Subject: "Elementary, my dear Ramsay!"
- Message-ID: <By01uA.Mtv@well.sf.ca.us>
- Sender: news@well.sf.ca.us
- Organization: Whole Earth 'Lectronic Link
- Date: Fri, 20 Nov 1992 05:27:46 GMT
- Lines: 114
-
-
- Ramsay writes:
- From: ramsay@unixg.ubc.ca (Keith Ramsay)
- Subject: Sarfatti and would-be FTL
- Date: 19 Nov 1992 05:09:44 GMT
- Organization: UBC, Vancouver, B.C., Canada, "Loudmouths' Anonymous"
-
- In order to talk about the superpositions of two states, the states
- themselves must be already defined. We can, of course, recycle our
- variables and use |a,e,+> and |a,o,+> to refer to analogous states
- within another experimental design-- but each joint reference to
- |a,e,+> and |a,o,+> has to refer to a particular one.
-
- [Ramsay, the states by themselves are rays inb a projective space so that
- there aboslute phase is undefined. In fact they are defined modulo any
- complex number not necessarily unimodular. Therefore, the relative phase
- can very easily be dependent on coefficients in superposition.]
-
- What the principle of superposition means is that we can vary our
- choice of superposition, e.g. z1 |a,e,+> + z2 |a,o,+>, *without* also
- varying the experimental design,
-
- [I don't think that is true. Why is that true?]
-
- and hence also without varying such
- relationships among |a,e,+> and |a,o,+> as the value of <a,e,+|a,o,+>.
- Each design will have its own value of <a,e,+|a,o,+>, not varying
- depending upon z1 and z2.
-
- [Prove this! It's not obvious. Is it in a standard quantum text book. If
- so, which one? What page?]
-
- I assumed that you implicitly fixed the experimental design. That is
- why I "automatically assumed" the invariance of <a,e,+|a,o,+>. If you
- never fix the experimental design, then it is easy to evade questions
- about unitarity as you have just described: by positing a different
- experimental design for each state!
-
- *Fix*, then, any *one* experimental design such as you've described,
- so that it is possible to speak of *the* fixed value of <a,e,+|a,o,+>,
- and so on. Consider the family of possible initial states of the pair
- of photons:
-
- |a,b> = |a,e,+> [ z1 |b,e,+> + z2 |b,o,-> ]
- +|a,o,-> [ z3 |b,e,+> + z4 |b,o,-> ],
-
- where |z1|^2+|z2|^2+|z3|^2+|z4|^2=1. If |a,e,+> evolves in the
- experiment to e^{i*phi} |a,e,+> and |a,o,-> to |a,o,+>, then we get a
- state
-
- |a,b> = e^{i*phi} |a,e,+> [ z1 |b,e,+> + z2 |b,o,-> ]
- + |a,o,+> [ z3 |b,e,+> + z4 |b,o,-> ].
-
- In order for this to be a state, its bracket with itself has to be 1.
- This has to hold true for *all* possible choices of z1,z2,z3,z4. We
- already know that |b,e,+> and |b,o,-> are orthogonal and normal. How
- can you possibly arrange all of this, without <a,e,+|a,o,+>=0? It just
- doesn't work.
-
- [Actually, this is all really beside the point and a nonproblem. Start from
-
- |a,b>=|a,e,+>[ z1 |b,e,+> + z2|b,o,-> ]+|a,o,-> [ z3 |b,e,+> + z4|b,o,-> ],
-
- |a,e,+> = |a,e>|a,+>
-
- |a,o,-> = |a,o>|a,->
-
- The half-wave plate on |a,o,-> causes
-
- U(1/2)|a,-> = |a,+>
-
- so that,
-
- |a,o,-> = |a,o>|a,+>
-
- OK assume for the space kets that
-
- <a,e|a,o> = 0
-
- Nevertheless, recombining both e and o paths at detector whose ket is |1>
- is described by the unitary transformation
-
- |a,b> -> |a,b>'=|a,+>|1>{[ (z1+z3)|b,e,+> + (z2+z4)|b,o,->]
-
- Note, Ramsay, that your condition
-
- |z1|^2+|z2|^2+|z3|^2+|z4|^2 = 1.
-
- from <a,b|a,b> = 1
-
- does not determine relative phases of the z's.
-
- Unitarity of |a,b> -> |a,b>' requires
-
- |z1|^2+|z2|^2+|z3|^2+|z4|^2 + z1*z3 + z2*z4 + c.c.
-
- so that
-
- z1*z3 + z2*z4 + c.c. = 0
-
- In the actual state required by the physics z2 = z3 = 0, so that this
- extra condition for unitarity is automatically satisfied by conservation of
- parity and angular mommentym in the J = 0->1->0 photon pair atomic cascade.
-
- Thus, my dear Ramsay, (as I puff on my pipe and play my violin at 10 Baker
- St.) there is nothing in the unitarity requirement of the standard quantum
- formalism that precludes both spin and space disentanglement of the initial
- entangled pair state |a,b> into a disentangled one |a>|b> in which the
- transmitter (a) phase (phi) information as transfered, on the connection
- communication channel beyond space-time, to the receiver (b) along with @
- the realtive angle between the detector calcites at the moments of the
- respective scatterings of each photon in the same pair. QED!]
-
- Thus writes Sarfatti.
-