home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.physics
- Path: sparky!uunet!darwin.sura.net!mips!mips!smsc.sony.com!galois!riesz!jbaez
- From: jbaez@riesz.mit.edu (John C. Baez)
- Subject: Re: ***********MIND-BLOWING GEDANKENEXPERIMENT**********
- Message-ID: <1992Jul27.214111.11030@galois.mit.edu>
- Sender: news@galois.mit.edu
- Nntp-Posting-Host: riesz
- Organization: MIT Department of Mathematics, Cambridge, MA
- References: <1992Jul24.170144.20607@bnr.ca> <92206.164128MRG3@psuvm.psu.edu> <2387@nic.cerf.net>
- Date: Mon, 27 Jul 92 21:41:11 GMT
- Lines: 41
-
- NOTE: John C. Baez is the author of this article. I am only using my
- account to post this article for John as a favor - Mark Corscadden
-
- In article <2387@nic.cerf.net> mitsu@nic.cerf.net (Mitsuharu Hadeishi) writes:
-
- >Given [|v1>|v2> + |h1>|h2>]/sqrt2 , if by local unitary evolution the
- >phase plate sends |v2> -> e^iphi|v2> and the have wave plate sends
- >|h2> -> |v2> (half wave plates rotate plane of linear polarization by
- >90 degrees says every optics text book) And I am using exactly the same
- >method that Wooters and Zurek used to prove that you cannot clone a photon
- >for arbitrary linear polarization in a laser amplifier which disproved Nick
- >Herberts FLASH FTL communicator - then evidently by the linearity, unitarity
- >and superposition principles and by elementary algebra of the distributive
- >law for kets
- >The initial entangled state [v1>|v2> + |h1>|h2>]/sqrt2 ->
- >[|v1> + e^iphi|h1>]|v2>/sqrt2 probability is conserved! p(v1,v2) = 1/2 and
- >p(h1,v2) = 1/2 { p(v1,h2) = 0, p(h1,h2) = 0, p(v1) = 1/2, p(h1) = 1/2,
- >p(v2) = 1, p(h2) = 0} so what is this garbage that probability is not
- >conserved. What false nonsense.
-
- Calm down please. There is no unitary operator such that for all h1,h2,
- v1,v2, the vector
-
- (|v1>|v2> + h1>h2>)/sqrt(2)
-
- is mapped to
-
- (|v1> + e^{i phi}|h1>)|v2>/sqrt(2).
-
- This is easily seen as follows. Unitary operators are one-to-one. The
- map you write down sends vectors with different h2's to the same thing
- (since h2 doesn't even appear in the second formula). So it is not
- one-to-one hence not unitary.
-
- In QM one often argues for unitarity on the basis of conservation of
- probability so I think the earlier remark about nonconservation of
- probability was just a way of saying what I just said: there's no
- unitary operator that does what you suggest, so it doesn't seem like a
- physically possible process.
-
- John Baez
-