1: \documentstyle[aps,prl,psfig]{revtex}
2:
3: \newcommand{\iy}{\infty}
4: \newcommand{\be}{\begin{equation}}
5: \newcommand{\ee}{\end{equation}}
6: \newcommand{\x}{{\rm x}}
7: \newcommand{\y}{{\rm y}}
8: \newcommand{\bex}{\begin{eqnarray}}
9: \newcommand{\eex}{\end{eqnarray}}
10: \begin{document}
11:
12: \title{Classical Communication via the Einstein-Podolsky-Rosen Channel Alone: A
13: Proposed Experimental Test}
14: \author{R. Srikanth\thanks{e-mail: srik@iiap.ernet.in}}
15: \address{Indian Institute of Astrophysics,
16: Koramangala, Bangalore- 34, Karnataka, India.}
17: \maketitle \date{}
18:
19: \pacs{03.67.-a, 03.65.Ud, 03.65.Ta, 03.30.+p}
20:
21: \begin{abstract}
22: We uncover an apparent instance of classical information transfer via only the
23: Einstein-Podolsky-Rosen channel in a quantum optical protocol between Alice and
24: Bob, involving two-photon maximal path entanglement and based on a recent
25: Innsbruck experiment. The signal is traced to the appearance of coherent
26: reduction due to the onset of
27: spatial degeneracy in the eigenvalue spectrum for Alice's measurement. We
28: present our result primarily as an issue for experimental testing rather than
29: as a definitive prediction at this stage.
30: \end{abstract}
31:
32: \parskip 3pt
33:
34: ~~~~~~~~
35:
36: \section{Introduction}
37: Does quantum nonlocality \cite{epr}, now attested by a vast body of
38: experimental evidence \cite{int,exp,sca00}, consist
39: only in the local inaccesibility of information
40: or, in addition, a nonlocal transfer of information \cite{hen97}? The answer
41: seems to lie subtly hidden in the quantum formalism.
42: The evidence for spacelike transfer of
43: information remains circumstantial, as in quantum teleportation
44: \cite{ben93} and remote state
45: preparation \cite{ben00}-- where it can be shown {\em a posteriori} that the
46: (in principle, infinite) classical information
47: about a quantum state is nonlocally transferred at the cost of only
48: 2 bits \cite{sri0104081}-- or controversial, as in the violation of the
49: Bell inequalities \cite{bel64,exp}, which tells us
50: that quantum mechanics (QM) cannot be local-realistic.
51: However, it is usually agreed that quantum nonlocality is causal \cite{nosig}.
52: Here we present a test of the question raised at the beginning.
53:
54: The article is arranged as follows.
55: The experimental set-up for and the basic idea behind
56: the test are presented in the next section.
57: The main result is derived more rigorously in Section \ref{main}.
58: This result in physically interpreted
59: in Section \ref{degen}, where we show that it owes
60: its origin to the onset of eigenvalue spectrum degeneracy in quantum
61: measurement. We then conclude with a final brief section.
62: In view of the absence of a standard
63: method to handle quantum measurement with degenerate outcomes,
64: experimental tests would be valuable in confirming or refuting the
65: effect predicted here to be possible.
66: Further careful scrutiny of the problem is needed also
67: in view of the far-reaching implications of the question considered.
68:
69: \section{Experiment}\label{xper}
70: The proposed experimental test, based on ideas involved in a recent
71: interesting quantum
72: optical experiment performed by Zeilinger's group at Innsbruck \cite{zei00},
73: involves two observers,
74: usually called Alice and Bob, sharing path-entangled biphotons (pairs of
75: entangled photons) from a suitable Einstein-Podolsky-Rosen source at point $o$.
76: One suggested candidate for the source is a nonlinear crystal (e.g., BBO) with
77: appropriate optical
78: pre-processing, enclosed by the dashed box in Figure \ref{yi}, which is a
79: `folded-out' schematic of the proposed experiment.
80: The crystal produces polarization-entangled biphotons
81: via spontaneous parametric
82: down-conversion (SPDC) in type II phase matching \cite{kwi95,bit01}, pumped by
83: a suitable laser beam (eg., Argon laser with $\lambda$ = 395 nm).
84: A filter restricts outgoing photons to a small bandwidth
85: about the downconverted frequency.
86:
87: The output from the crystal beyond the dashed box
88: is the maximally entangled state
89: \be
90: \label{output}
91: |\psi\rangle_{AB} = \frac{1}{\sqrt{2}}(|HV\rangle_{AB} - |VH\rangle_{AB}),
92: \ee
93: where $\{|V\rangle$,
94: $|H\rangle\}$ represent vertical and horizontal polarization states
95: and subscripts $A$ and $B$ represent Alice's and Bob's photon.
96: On both sides of the crystal, there is: (1) a polarizing
97: beam-splitter (PBS1 or PBS2) to seperate the beams in such a way that
98: eventually the $H$
99: beam is parallel to the $V$ beam; (2) a half-wave-plate (HWP1 or HWP2) in the
100: $V$ paths to rotate them to $H$ in order to permit interference
101: (though the labels retain
102: their original polarization values for the sake of uniformity of notation).
103: In this way, maximal polarization entanglement is converted to maximal
104: path entanglement.
105:
106: Alice is equipped with a lens of focal length $f$, and a movable detector
107: system that can be positioned at distance $f$ or $f - g$ from the lens.
108: Bob is equipped with a Young's double-slit interferometer, located at distance
109: $d$ from the plane containing the crystals.
110: By classical optics, parallel rays entering
111: Alice's lens converge to a single point on the focal plane. Because of
112: path entanglement, a
113: detection by Alice at some point on the focal plane
114: will leave Bob's photon in a superposition of parallel rays, i.e,
115: a definite `momentum state',
116: but with its point of origin in the source indeterminate, as expected
117: on basis of the Uncertainty principle. This has been confirmed
118: by the interference pattern seen in Bob's photons detected in
119: coincidence with Alice's measurement at her focal plane \cite{zei00}.
120: On the other hand, if Alice advances her detector system to a distance
121: closer to the lens, distinguishability is restored and no interference is
122: possible.
123:
124: In the present protocol, Alice will choose to position her detector
125: at distance $f$ or $f - g$ from the lens (Figure \ref{yi}).
126: Alice's detector must be wide enough to intercept all $A$
127: photons. In practice, one detector element (the scanning tip of a fiber optic
128: element) is sufficient, to be positioned at $k$, for the
129: focal plane measurement, or at either $l$ or $m$, for the off-focal-plane
130: measurement (since a non-detection is also a measurement). Suppose she
131: positions her detector on the plane at distance $f - g$. She will in effect
132: detect $A$ at point $l$ or $m$. Since her two possibile outcomes
133: are distinguishable, and because of entanglement, she correspondingly leaves
134: Bob's photon in one of the states $|H\rangle$ or $|V\rangle$, respectively.
135: Neither detection by her produces an
136: interference in the coincidence counts at Bob's interferometer because
137: she has acquired path information for Bob's photon.
138: Therefore, Bob will find no interference pattern on his
139: screen in his single counts.
140:
141: On the other hand, suppose she positions it at the focus $k$ of the lens.
142: She will detect a click at point
143: $k$. Since she cannot distinguish whether her detection was generated by a
144: photon coming through the upper or lower path, by Feynman's dictum, both paths
145: interfere at $k$, and, because of path entanglement, correspondingly she is
146: expected to leave Bob's photon in the superposition state
147: \be
148: \label{M}
149: |M\rangle_B \equiv \frac{1}{\sqrt{2}}(\alpha |H\rangle_B + \beta |V\rangle_B)
150: \hspace{1.0cm}(|\alpha|^2 + |\beta|^2 = 1),
151: \ee
152: where $\alpha$ and $\beta$ are path dependent phase factors
153: (cf. Section \ref{degen}). Therefore, in this case, Bob
154: will find a Young's double slit pattern in the coincidence counts, produced by
155: the interfering $H$ and $V$ rays, rotated by the half-wave plate HWP2.
156: Further, because the coincidence counts involve only one possible outcome for
157: Alice, he in fact finds the interference pattern in his singles counts.
158:
159: What evidence do we have that Bob's photon is indeed
160: projected into the state Eq. (\ref{M}) by Alice's focal
161: plane measurement? One way to understand it is to view Alice's photon $A$ as a
162: handle on Bob's photon's path. Registering $A$ at $k$, irreversibly
163: destroys path information about Bob's photon, so the two $B$ paths are
164: free to interfere, but otherwise, not.
165: This complementaristic interpretation is well borne out by
166: implementations of the delayed choice experiment \cite{her95}. Furthermore,
167: coincidence
168: measurements on two-particle interference experiments \cite{int} clearly
169: show that the point where one of the entangled pair is registered
170: acts like a virtual source for the origin of the modes whose interference
171: determines the probability distribution for the localization of its
172: twin photon. % In fact, inserting appropriate phase factors $\alpha$ and
173: % $\beta$, we can obtain the correct coincidence count interference.
174: A further corroboration of this reasoning, closer to the
175: experiment at hand, comes
176: from the Innsbruck experiment \cite{zei00}, where it is found that
177: when the signal beam is focussed to a detector using a lens, an interference
178: pattern is detected in the idler coincidence counts, as discussed earlier.
179:
180: Finally, we want to
181: note that visualizing state $|M\rangle_B$ as the projection that results from
182: Alice's measurement is a sort of concession to our quantum mechanical
183: intuition, but is not necessary for the calculation of probabilities in
184: the quantum optical formalism, as shown in the following section.
185: Bob discerns whether Alice measured at the focal plane or not
186: depending on whether or not he finds the
187: tell-tale interference pattern. Clearly, this classical signal can be
188: transmitted arbitrarily fast by increasing $f$ and/or $d$ indefinitely.
189: This completes the basic idea of the proposed nonlocal classical signaling
190: test, which will be examined critically in the next two sections.
191:
192: \section{Derivation}\label{main}
193:
194: We now derive quantitatively the result given in the preceding section.
195: The four-mode state vector of the SPDC field incident on Alice's and Bob's
196: detectors is given by:
197: \be
198: \label{spdc}
199: |\Psi\rangle = |{\rm vac}\rangle + \epsilon(|hv\rangle - |vh\rangle)
200: \ee
201: where $|{\rm vac}\rangle$ is the vacuum state,
202: $|h\rangle$ and $|v\rangle$ are the Fock state modes propagating along the
203: $H$ and $V$ arms of the experiment and $\epsilon (\ll 1)$ depends on the
204: pump field strength and the crystal nonlinearity.
205: The positive frequency part of the electric field at an arbitrary point $z$
206: on Bob's screen is:
207: \be
208: \label{bobfjeld}
209: E_z^{(+)} = e^{ikr_D}\left(e^{ikr_1}\hat{h} + e^{ikr_2}\hat{v}\right),
210: \ee
211: where $\hat{h}$ and $\hat{v}$ are the annihilation operators for the
212: $h$ and $v$ modes, respectively. $r_D$ is the distance from the source
213: $o$ via PBS2 to the upper/lower slit on Bob's double slit diaphragm;
214: $r_1$ ($r_2$) is the distance from the upper (lower) slit to $z$ (Figure
215: \ref{yi}).
216:
217: If Alice positions her dectector at point $l$ or $m$ on the plane at
218: distance $f - g$ from the lens, the positive frequency part of the electric
219: field at point $l$ or $m$ is
220: \bex
221: \label{alicefjeldb}
222: E_l^{(+)} = e^{ikr_L}\hat{v}, \hspace{1.0cm}
223: E_m^{(+)} = e^{ikr_M}\hat{h},
224: \eex
225: where $r_L$ ($r_M$) is the distance from
226: $o$ via PBS1 along the upper (lower) path through the lens upto point $l$
227: ($m$). For simplicity, we set $r_L = r_M$. Now,
228: if Alice positions her dectector at the focal plane, the positive frequency
229: part of the electric field at point $k$ is
230: \be
231: \label{alicefjelda}
232: E_k^{(+)} = e^{ikr_K}\left(\hat{h} + \hat{v}\right),
233: \ee
234: where $r_K$ is the distance from
235: $o$ via PBS1 along the upper or lower path through the lens upto point $k$.
236: Again, for simplicity, the distances along the two paths have been taken to be
237: identical.
238:
239: The coincidence count rate $R$ for simultaneous measurements by Alice and
240: Bob is given by the absolute square of the second order correlation function
241: $\langle\Psi|E^{(+)}_yE^{(+)}_z|\Psi\rangle$ ($y = k, l, m$). This
242: is proportional to the probability for Alice's and Bob's correlated
243: measurements. If Alice positions her dectector at point $l$ or $m$ on the
244: plane at distance $f - g$ from the lens, the coincidence rate
245: for detections by Alice and by Bob at $z$ is
246: \be
247: \label{Rg}
248: R_g \propto |\langle\Psi| E_y^{(+)}E_z^{(+)}|\Psi\rangle|^2
249: \propto \epsilon^2 ,
250: \ee
251: where $(y = l, m)$ and we have used Eqs. (\ref{spdc}), (\ref{bobfjeld}) and
252: (\ref{alicefjelda}). As the coincidence rates for Alice's both detections
253: are uniform, Bob finds a uniform intensity pattern on his screen.
254: On the other hand, if she positions her detector at focus $k$,
255: the coincidence rate is given by
256: \be
257: \label{Rf}
258: R_f \propto |\langle\Psi| E_k^{(+)}E_z^{(+)}|\Psi\rangle|^2 =
259: \epsilon^2\{1 + \cos(k\cdot[r_1 - r_2])\},
260: \ee
261: which is equivalent to a conventional Young's double slit pattern. Because of
262: the focussing, no other coincidence terms are involved. So the interference
263: pattern Eq. (\ref{Rf}) is in fact seen in Bob's singles counts.
264: In an actual implementation, Eqs. (\ref{Rg}) and
265: (\ref{Rf}) must be further modified to take
266: into consideration the single slit diffraction pattern and the profile of the
267: down-converted laser beam.
268:
269: Usually, in biphoton interference \cite{int} experiments,
270: Alice's photon is not focussed to a single point, but allowed to
271: spread out according to the beam profile or through a single/double
272: slit system. Therefore
273: Bob would see an interference pattern of the type Eq. (\ref{Rf})
274: averaged over various Alice's detection positions, which
275: smears Bob's pattern to a uniform distribution that is indistinguishable
276: from that in Eq. (\ref{Rg}). In the above experiment,
277: by the focussing of Alice's beam, this smearing is-- crucially-- checked.
278: % whereby Bob's coincidence counts are rendered tantamount to single counts.
279:
280: Are there some other reasons that come into play
281: that somehow restore the distinguishability between the two paths in Alice's
282: focal plane measurement? Ultimately, only an experimental test can adjudicate.
283: Suppose Bob's beam diverges slowly, i.e,
284: $s\theta \ll \lambda$, where $s$, $\theta$
285: and $\lambda$ are slit-width, divergence angle and wavelength, respectively
286: An unentangled laser beam satisfying this condition will produce a Young's
287: double slit pattern. After Alice's focal plane measurement,
288: Bob's beam is indeed disentangled into such an unentangled laser beam.
289: Of course, the wavefront
290: of Bob's disentangled beam will be modified because only annular regions about
291: the lens's principal axis have the same phase (since they have the same
292: $r_K$). Nevertheless, provided the
293: double slit is positioned symmetrically about the lens's axis, and Bob's
294: downconverted (entangled) beam satisfies the usual interference criterion
295: $s\theta \ll \lambda$, then a possible interesting outcome for the above
296: experiment can be expected.
297:
298: We note that unlike the case with position-momentum entanglement, a spreading
299: of polarization-entangled light of laser is not required by the
300: Uncertainty principle. Therefore, in principle, polarization-entangled thin
301: pencils satisfying the above slow divergence condition are easier to prepare
302: than position-momentum entangled beams. This is
303: the reason why the present experiment is simpler to implement than
304: that presented in Ref. \cite{sri2223}, in which considerable optical
305: preprocessing is needed to bring about a position-momentum entangled equivalent
306: of the above experiment, and which, incidentally,
307: is closer to the Innsbruck experiment \cite{zei00}. In this regard, we note
308: that that no sudden spreading of
309: $B$ occurs on account of $A$'s localization even if the input is light
310: position-momentum entangled \cite{kim}.
311:
312: Two suggestions for the preparation of path-entangled pencils: (1) blocking
313: out photons $A$ and $B$ with a shield, except at two small opposing holes,
314: one on each shield, from which fiber optic cables of equal
315: length lead to the respective polarizing beam splitter; (2)
316: selective Bell state measurement using appropriate linear optics
317: on two seperable thin laser beams (in this connection, cf. Ref. \cite{lam01}).
318:
319: \section{Quantum mechanical picture}\label{degen}
320:
321: Although interference experiments rightly belong to the domain of quantum
322: optics (QO), many of them can usually be translated into quantum mechanical
323: language (for example, cf. Ref. \cite{lou} as regards the Mach-Zehnder
324: interferometer and Ref. \cite{sri154} as regards the delayed choice experiment
325: \cite{whe94,hom94}). Sometimes the
326: latter version can be easier to physically interpret.
327: Reverting back to the QM notation
328: of Section \ref{xper}, we find the reduced density matrix
329: for Bob's photon if Alice measures in the `focal plane'
330: basis, namely $(\rho_f)_B$, and that if she measures off the focal plane,
331: namely $(\rho_g)_B$, to be
332: \bex
333: (\rho_f)_B &=& |M\rangle_B\langle M|_B = \frac{1}{2}(\alpha|H\rangle_B +
334: \beta|V\rangle_B)(\alpha^*\langle H|_B + \beta^*\langle V|_B) \nonumber \\
335: (\rho_g)_B &=& \frac{1}{2}\left(|H\rangle_B\langle H|_B +
336: |V\rangle_B\langle V|_B\right).
337: \eex
338: The classicality of the signal is an expression of the fact that
339: $(\rho_f)_B \ne (\rho_g)_B$.
340:
341: In order to trace the origin of the classical signal, let us interpret the
342: results of the preceding section in the QM Schr\"odinger picture by inserting
343: the appropriate phase factors.
344: On each path,
345: \be
346: \label{transA}
347: |H\rangle_X \longrightarrow e^{ik\cdot x_{HX}}|H\rangle_X;
348: \hspace{1.0cm}
349: |V\rangle_X \longrightarrow e^{ik\cdot x_{VX}}|V\rangle_X,
350: \ee
351: where $x_{HX}$ ($x_{VX}$) is the distance along the $H$ ($V$) path on beam $X$
352: ($X = A, B$) from the source at $o$. Beyond the double slit, $B$ is
353: transformed into Bob's screen measurement basis according to:
354: \be
355: \label{transB}
356: |H\rangle_B \longrightarrow e^{ik\cdot r_1(z)}|z\rangle
357: \hspace{1.0cm}
358: |V\rangle_B \longrightarrow e^{ik\cdot r_2(z)}|z\rangle ,
359: \ee
360: where $|z\rangle$ is the eigenstate corresponding to $B$ being found at $z$,
361: an arbitrary detector element. Bob's measurement that localizes his particle
362: at $z$ is given by the usual von Neumann projector,
363: $\hat{P}_z \equiv |z\rangle\langle z|$.
364:
365: After inserting the spatial dependences Eqs. (\ref{transA}) and (\ref{transB})
366: into the biphoton state vector Eq. (\ref{output}), we have
367: \be
368: \label{output0}
369: |\psi\rangle_{AB} \longrightarrow \sum_z \left(|Hz\rangle_{AB}
370: e^{ik[r_A + r_D + r_2(z)]} - |Vz\rangle_{AB}e^{ik[r_A + r_D + r_1(z)]}\right),
371: \ee
372: with $r_A$ set to $r_L = r_M$, for off-focal-plane measurement, or to $r_K$,
373: for focal plane measurement.
374:
375: These two measurement planes are equivalent
376: to two different observables. The spectral decomposition for the observable
377: corresponding to measurement on the plane at distance $f - g$ from the lens
378: can be written as
379: \be
380: \label{Og}
381: \hat{O}_g = l|V\rangle_A\langle V|_A + m|H\rangle_A\langle H|_A.
382: \ee
383: According to the von Neumann projection postulate \cite{vN}, the probability to
384: find $m$ or $n$ is given by the expectation value of the corresponding projector
385: \be
386: \label{Ebc}
387: \hat{P}_l \equiv |V\rangle_A\langle V|_A, \hspace{1.0cm}
388: \hat{P}_m \equiv |H\rangle_A\langle H|_A
389: \ee
390: in the state $|\Psi\rangle_{AB}$ in Eq. (\ref{output0}) (setting $r_K$ to
391: $r_L = r_M$) \cite{vN}.
392:
393: The observable corresponding to measurement on the focal plane
394: is seen to be degenerate
395: in space, i.e., both eigenstates $|H\rangle$ and $|V\rangle$ have the
396: same position eigenvalue $k$. Hence, in analogy with Eq. (\ref{Og}), we write:
397: \be
398: \label{Of}
399: \hat{O}_f = k|V\rangle_A\langle V|_A + k|H\rangle_A\langle H|_A.
400: \ee
401: (An `energetic' analogy would be switching a magnetic field on or off to render
402: the spin eigenstates of an entangled electron non-degenerate or degenerate in
403: a local energy measurement.)
404: A measurement of position yields the value $k$, but what is the corresponding
405: projector? Clearly the von Neumann projection postulate cannot
406: handle this case and must somehow be extended. The problem of degenerate
407: measurement was first considered by L\"uders \cite{lud51} and has been the
408: subject of recent renewed interest \cite{bro01}.
409:
410: Two related problems here are that of: (a)
411: calculating the probability for obtaining the eigenvalue corresponding to the
412: degenerate subspace and, (b) determining the state in which the system is left
413: if the degenerate eigenvalue is found.
414: The latter problem for QO is necessarily
415: different from that in QM because no particle annihilation figures in QM,
416: unlike in QO. Hence, a QM interpretation is applicable
417: only with respect to the former problem. In L\"uders' \cite{lud51}
418: formalism for extending the projection postulate, the projector $\hat{P}_k$
419: for Alice's focal plane measurement is given by $\hat{P}_k = \hat{P}_l +
420: \hat{P}_m$. If valid, this extension
421: would indeed be sufficient to prohibit the classical signaling.
422: However, one can verify that $\langle (\hat{P}_l + \hat{P}_m) \otimes
423: \hat{P}_z\rangle$ does not reproduce the
424: interference Eq. (\ref{Rf}), where $\langle\cdots\rangle$ represents
425: expectation value with respect to the state Eq.
426: (\ref{output0}) setting $r_A = r_K$. The reason is that
427: it does not permit crosstalk between Bob's $H$ and $V$ modes, needed to
428: explain Bob's coincident interference pattern in Eq. (\ref{Rf}), and
429: more generally, the interference seen in coincidence with focal plane
430: detections of $A$ in the Innsbruck experiment \cite{zei00}.
431:
432: It turns out that the form of
433: the degenerate projector that agrees with Eq. (\ref{Rf}), and thus
434: rightly represents $A$'s electric field in Eq. (\ref{alicefjelda}), is
435: \be
436: \label{Ea}
437: \hat{P}_k \equiv \hat{P}_{l+m} =
438: (|H\rangle_A + |V\rangle_A)(\langle H|_A + \langle V|_A)
439: \ne \hat{P}_l + \hat{P}_m,
440: % + |H\rangle_A\langle V|_A + |V\rangle_A\langle H|_A,
441: \ee
442: in view of Eq. (\ref{Ebc}). By direct computation,
443: one can verify that $\langle \hat{P}_{l+m} \otimes \hat{P}_z\rangle$
444: indeed reproduces the interference Eq. (\ref{Rf}).
445: This implies that {\em in projecting a state vector to a state
446: corresponding to a degenerate eigenvalue, the amplitudes of the eigenstates
447: in the degenerate subspace superpose}. In other words, the probability
448: amplitude that a measurement finds a degenerate
449: eigenvalue is given by the sum of the degenerate amplitudes.
450: The consequences of this for state representation and state vector reduction,
451: both in seperable and entangled systems, and experimental tests of whether the
452: validity of such a `coherent projection' and of the associated `coherent
453: reduction' can be extended beyond QO to the case of energy degeneracy in QM
454: proper, are taken up in the future. Coherent reduction brings further richness
455: to the essential quantum phenomenon of superposition. What is encouraging is
456: that quantum optical tests for it are well feasible (in a related vein, cf.
457: Ref. \cite{ghi98}).
458:
459: According to the foregoing analysis, the classical nonlocal signal in the
460: proposed experiment owes its origin to the fact that Alice can choose to make
461: either a von Neumann measurement
462: in the non-degenerate off-focal-plane basis, or a degenerate measurement in
463: the focal plane basis, thereby disentangling Bob's photon by projecting it
464: completely or coherently. As a result, it requires only the EPR
465: channel and no additional classical communication channel.
466: This is not incompatible with the conclusion of
467: Refs. \cite{nosig}, where implicitly only complete von Neumann measurements in
468: a non-degenerate basis are considered.
469:
470: The new result derived here is in fact implicit in the
471: Innsbruck experiment \cite{zei00}, and we would not expect its possible
472: positive outcome if the Innsbruck experiment would not have found
473: interferences in Bob's coincidence counts for Alice's
474: focal plane measurement. And yet this latter counterfactual hypothesis would
475: not be possible without the abandonment of the usual quantum optical formalism
476: for calculating interferences at second order. Therefore, the classicality of
477: the signal in a sense lurks in the familiar double slit interference.
478:
479: \section{Conclusion}
480:
481: Feynman noted that the central mystery of QM--
482: namely, superposition-- is encapsulated by the double-slit interference
483: \cite{fey65}. Coherent reduction, as discussed above,
484: adds a further perspective to this `mystery'.
485: Technically speaking, an (improbable) positive outcome of the
486: experiment is not incompatible with QM itself, since
487: the features of QM that guarantee relativistic causality-- namely,
488: linearity \cite{gis90,rig95}, unitarity \cite{woo82}
489: and the tensor product character of the Hilbert space of composite quantum
490: systems \cite{pea00}-- are essentially non-relativistic. On the other hand, a
491: null result for the proposed experiment, although less obviously explained,
492: would be easier to accept.
493:
494: % {\small I thank Mr. B. S. Ramachandra, Ms. P. Biswas and Mr. B. Sahoo
495: % for discussions.}
496:
497: \begin{thebibliography}{}
498: \bibitem{epr} A. Einstein, B. Podolsky and N. Rosen, Phys. Rev. {\bf 47}, 777 (1935).
499: \bibitem{int} R. Ghosh, and L. Mandel, Phys. Rev. Lett. {\bf 59},
500: 1903 (1987); C. K. Hong, Z. Y. Ou and L. Mandel, Phys. Rev. Lett. {\bf 59},
501: 2044 (1987); Y. H. Shih and C. O. Alley, {\em ibid.} {\bf 61}, 50
502: (1988); J. G. Rarity and P. R. Tapster, {\em ibid.} {\bf 64}, 2495 (1990);
503: J. Brendel, E. Mohler and W. Martienssen, {\em ibid.} {\bf 66}, 1142
504: (1991); T. S. Larchuk, R. A. Campos J. G. Rarity, P. R. Tapster, J. G.
505: Jakeman, B. E. A. Saleh and M. C. Teich, {\em ibid.} {\bf 70} 1603 (1993);
506: A. M. Steinberg, P. G. Kwiat and R. Y. Chiao, {\em ibid.} {\bf 71} 708 (1993);
507: T. E. Keiss, Y. H. Shih, A. . Sergienko and C. O. Alley, {\em ibid.}
508: {\bf 71} 3893 (1993);
509: P. G. Kwiat, A. M. Steinberg, and R. Chiao, Phys. Rev.
510: A {\bf 47}, 2472 (1993); D. V. Strekalov, A. V. Sergienko, D. N. Klyshko, and
511: Y. H. Shih, Phys. Rev. Lett. {\bf 74}, 3600 (1995); T. B. Pittman, Y. H.
512: Shih, D. V. Strekalov and A. V. Sergienko, Phys. Rev. {\bf A52} R3429 (1995).
513: \bibitem{exp} A. Aspect, P. Grangier, and G. Roger, Phys. Rev. Lett.
514: {\bf 49}, 91 (1982); W. Tittel, J. Brendel, H. Zbinden, and N. Gisin,
515: Phys. Rev. Lett. {\bf 81}, 3563 (1998). G. Weihs, T. Jennewein,
516: C. Simon, H. Weinfurter, and A. Zeilinger, Phys. Rev. Lett. {\bf 81},
517: 5039 (1998).
518: \bibitem{sca00} V. Scarani, W. Tittel, H. Zbinden, N. Gisin,
519: Phys. Lett. A {\bf 276} 1 (2000); % also eprint quant-ph/0007008;
520: H. Zbinden, J. Brendel, N. Gisin and W. Tittel,
521: Phys. Rev. A {\bf 63}, 022111-1 (2001). % eprint quant-ph/0007009 (2000).
522: \bibitem{hen97} H. Stapp, Am. J. Phys. {\bf 65}, 300 (1997); H. Stapp,
523: eprint quant-ph/0010047.
524: \bibitem{ben93} C. H. Bennett, G. Brassard, C. Cr\'epeau, R. Josza, A. Peres
525: and W. K. Wootters, Phys. Rev. Lett. {\bf 70}, 1895 (1993).
526: \bibitem{ben00}
527: C. H. Bennett, D. P. DiVincenzo, P. W. Shor, J. A. Smolin,
528: B. M. Terhal, W. K. Wootters, Phys. Rev. Lett. {\bf 87} (2001) p.077902;
529: eprint quant-ph/0006044.
530: \bibitem{sri0104081} R. Srikanth, e-print/0104081.
531: \bibitem{bel64} J. S. Bell, Physics {\bf 1}, 195 (1964); F. Clauser,
532: M. A. Horne, A. Shimony, R. A. Holt, Phys. Rev. Lett. 23, 880 (1969).
533: % \bibitem{sta78} H. Stapp.
534: \bibitem{nosig} P. H. Eberhard, {\em Nuovo Cimento} {\bf 46B}, 392 (1978);
535: P. H. Eberhard and R. R. Ross, Found. Phys. Lett. 2 127 (1989);
536: D. Bohm and B. J. Hiley, {\it The Undivided Universe: An
537: ontological interpretation of Quantum Theory} (Routledge, London and
538: New York, 1993); P. J. Bussey, Phys. Lett. {\bf 90}A, 9 (1982);
539: T. F. Jordan, Phys. Lett. {\bf 94}A, 264 (1983);
540: A. J. M. Garrett, Found. Phys. {\bf 20}, No. 4, 381 (1990).
541: C. D. Cantrell, and M. O. Scully, Phys. Rep. {\bf 43}, 499 (1978).
542: \bibitem{zei00} A. Zeilinger, Rev. Mod. Phys. {\bf 71}, S288 (1999);
543: A. Dopfer, PhD thesis (Univ. of Innsbruck 1998).
544: \bibitem{kwi95} P. G. Kwiat, K. Mattle, H. Weinfurter, A. Zeilinger,
545: A. V. Sergienko and Y. Shih, Phys. Rev. Lett., 75, 4337 (1995).
546: \bibitem{bit01} G. Bitman, W. P. Grice, J. Moreau and L. Zhang, eprint
547: quant-ph/0106122.
548: \bibitem{her95} T. J. Herzog, P. G. Kwiat, H. Weinfurter and A. Zeilinger,
549: Phys. Rev. Lett. {\bf 75}, 3034 (1995); Y-H. Kim, R. Yu, S. P. Kulik,
550: Y. Shih and M. Scully, Phys. Rev. Lett. {\bf 84}, 1 (2000).
551: \bibitem{sri2223} R. Srikanth, e-print quant-ph/0101022; quant-ph/0101023.
552: \bibitem{kim} Y.- H. Kim \& Y. Shih, Found. Phys. {\bf 29}, 1849 (1999).
553: \bibitem{lam01} A. Lamas-Linares, C. Mikkelsen, J. C. Howell, D. Bouwmeester,
554: eprint quant-ph/0103056.
555: \bibitem{lou} R. Loudon, {\em Quantum Theory of Light}, (Oxford University
556: Press, 2000).
557: \bibitem{sri154} R. Srikanth, to appear as a Note in Current Science (2001);
558: eprint quant-ph/0106154.
559: \bibitem{whe94} J. A. Wheeler, {\em At Home in the Universe}, (AIP New York
560: 1994), p. 120,311.
561: \bibitem{hom94} D. Home, {\em Conceptual Foundations of Quantum Mechanics}
562: (Plenum, New York 1994), p. 274-275.
563: \bibitem{vN} J. v. Neumann, {\em Matematiske Grundlagen der Quantenmechanik}
564: (Springer, Berlin 1932); translation into English by R. T. Beyer, {\em
565: Mathematical Foundations of Quantum Mechanics} (Princeton Univ. Press 1971).
566: \bibitem{lud51} G. L\"uders, Annalen der Physik {\bf 8}, 322 (1951).
567: \bibitem{bro01}
568: S.L. Adler, D.C. Brody, T.A. Brun and L. P. Hughston, eprint quant-ph/0107153;
569: D. Brody, L. Hughston, eprint quant-ph/0104032.
570: \bibitem{ghi98} G. Ghirardi, eprint quant-ph/9810028.
571: \bibitem{fey65} R. P. Feynman, R. B. Leighton and M. Sands, {\em The Feynman
572: Lectures on Physics}, Vol. III (Addison Wesley 1965).
573: \bibitem{gis90} N. Gisin, Phys. Lett. A 143, 1 (1990).
574: \bibitem{rig95} M. Rigo and N. Gisin, Jl. Phys. A 28, 7375 (1995).
575: \bibitem{woo82} W. K. Wooters and W. H. Zurek, Nature {\bf 299}, 802 (1982).
576: \bibitem{pea00} K. A. Peacock and B. S. Hepburn, quant-ph/9906036.
577:
578: \end{thebibliography}
579:
580: \newpage
581:
582: %\vspace*{2.0cm}
583:
584: \begin{figure}
585: \centerline{\psfig{file=yi.eps,width=16.5cm}}
586: \vspace*{0.5cm}
587: \caption{Alice and Bob share a thin pencil of polarization-entangled biphotons
588: in pure state, from an EPR source at $o$. By means of a pair of polarizing
589: beam splitters (PBS1 and PBS2) and half-wave plates (HWP1 and HWP2),
590: polarization entanglement is converted to path-entanglement. Depending on
591: whether she observes her photon at the focus of her lens, or off the focal
592: plane, she cannot or can obtain path information for Bob's photon, thereby
593: permiting or prohibiting the latter's interference. The part of the experiment
594: enclosed in the dashed box prepares the biphoton in used to prepare the
595: biphoton in a path-entangled state.}
596: \label{yi}
597: \end{figure}
598: \end{document}
599:
600:
601: