Foundations of Physics Letters, Vol. 12, No.

2, 1999

WHY BOHM'S QUANTUM THEORY?

H. D. Zeh
Institut fUr Theoretische Physik, Universitiit Heidelberg
Philosophenweg 19
D-69120 Heidelberg, Germany
zeh@urz. uni-heidelberg. de

Received 22 February 1999

This is a brief reply to S. Goldstein's article "Quantum theory with-

out observers" in Physics Today. It is pointed out that Bohm's pilot
wave theory is successful only because it keeps Schrodinger's (exact)
wave mechanics unchanged, while the rest of it is observationally
meaningless and solely based on classical prejudice.

Key words: quantum theory, pilot wave theory, quantum trajecto-

ries, decoherence.

In his recent article on "Quantum theory without observers" [1],

Sheldon Goldstein raised a number of important questions about
quantum theory. Even though I cannot quite understand what a
universal physical theory without any concept of observers could
mean, I agree with many of his critical remarks. Bell, who once ob-
jected "against measurement," also pointed out that Bohm's quan-
tum theory depends on the assumption that only its hypothetical
classical variables can directly affect the consciousness of an observer
f2]. However, Goldstein's characterization of the consequences that
have to be drawn from his criticism appears one-sided, since he ne-
glects many essential aspects.
In particular, he does not mention at all that Bohm's classical
trajectories can neither be experimentally confirmed nor refuted if
the theory is exactly valid. It is always possible to postulate oth-
erwise unobservable (hence arbitrary) causes for stochastic events.
Bohm's presumed ensemble of classical configurations is merely dy-
namically consistent because of their presumed unobservable dynam-
197

0894-9875/9910400-0197$16.00/0 10 1999 Plenum Publishing Corporation

198 Zeh

ics. The resulting trajectories are thus entirely based on classical

prejudice. Goldstein quotes Bell that "it should be clear from the
smallness of the scintillation on the screen that we have to do with
a particle." Should it then not be as clear that the photon (or any
other boson) is a particle, too, rather than the consequence of a
field on three-dimensional space, as is assumed in Bohm's theory?
What is observed as a local object is in all cases the position of a
macroscopic "pointer" (such as a flash on the screen or a track in
the bubble chamber).
Appropriately localized wave packets representing pointer po-
sitions could be explained as emerging in a fundamental irreversible
"spontaneous localization" process by means of a modified Schrodin-
ger equation, such as it has been proposed by Pearle or Ghirardi, Ri-
mini, and Weber (GRW) [3) and also mentioned by Goldstein. Why
then re-introduce classical variables? There is no need (or any good
reason) even for a Heisenberg picture in terms of particles, nor for a
wave-particle "dualism". And why always represent these classical
variables by a (crucial) ensemble, while the wave function (the other
element of reality in Bohm's theory) is normally regarded as given,
but tacitly excluded from being directly experienced by an observer?
The nonlinear terms in spontaneous localization models are
chosen to become relevant in measurements or similar situations in
order to describe von Neumann's collapse of the wave function. How-
ever, equivalent effects in the density matrix of any relevant system
arise without any modification of the Schrodinger equation if only
the unavoidable environment of this system is realistically taken into
account by what Goldstein calls "Zurek's decoherence" [4,5). There-
fore, these two quite different dynamical mechanisms (the disappear-
ance and the dislocalization of phase relations) can both explain the
emergence of apparent "particles" and other quasi-classical (local)
properties (such as Bell's "small scintillations"). They cannot prac-
tically be distinguished from one another in these cases, provided
only the usual probability rules apply somewhere and for something
along the chain of interactions between pointer and observer( s) in or-
der to justify the concept of a density matrix. Goldstein's claim that
"the environment acts as an observer" is quite wrong: decoherence
is an objective physical process that is essential (and unavoidable)
also in Bohm's theory for correctly guiding the postulated classical
pointer positions or other macroscopic variables.
The application of a probabilistic collapse just where environ-
mental decoherence occurs anyway is thus as much a prejudice as the
existence of classical particles and fields. "Appearences are mislead-
ing." Here I fully agree with Goldstein (although appearance requires
an observer). In contrast to Bohm, Pearle and GRW proposed modi-
fications of quantum theory that can be experimentally refuted (and
very probably have been unless modified again [6)). While the search
Bohm's Quantum Theory 199

for deviations from the Schrodinger equation is a reasonable ende-

vour, one must warn young scientists against sp,ending their time
on calculating Bohm trajectories or "unraveling ' entangled (open)
systems - as suggestive as these pictures may appear to the tra-
ditionalistic mind. They would be investigating mere phantoms.
Such calculations would be meaningful if they led to new observable
consequences. However, precisely this is excluded by construction
of Bohm's theory, while a quantum system that is entangled with
its environment clearly does not possess any wave function by its
own. The idea of "quantum trajectories" for all systems of inter-
est, now quite popular in quantum optics, is (1) inconsistent (as it
would depend on the precise choice of the systems), (2) incompat-
ible with a Schrodinger equation (even if that were complemented
by a phenomonological collapse), and (3) in conflict with relevant
experiments.
Most experiments are local and not relevant for this purpose
(thus permitting the concept of a local density matrix, or an ap-
parent ensemble of wave functions representing it). However, all
basic quantum experiments performed during recent decades have
confirmed consequences of a nonlocal wave function( al), defined on
a high-dimensional space that we are used to interpret as a classi-
cal configuration 8pace. These consequences appear as "paradoxes"
when described in terms of particles or spatial waves (fields). The
Schrodinger equation, which unavoidably leads to drastic (eventu-
ally universal) entanglement between all macroscopic systems (and
thereby local(y to decoherence) [7], would not have to be modified
at all by means of the apparently required probalistic terms if one
dropped the further prejudice that there is only one state of each
observer in this "nonlocal reality". A universally exact Schrodinger
equation would require that there is an ever increasing number of dif-
ferent "versions" of the quasi-classical world with all its observers,
which are individually described by robust wave packets in appropri-
ate factor spaces of this high-dimensional space. According to wave
mechanics, observers can only exist in dynamically branching, there-
after essentially autonomous components of the global wave function.
These observer versions would practically lack any possibility of com-
municating with one another, and therefore need not be assumed to
disappear from reality with precisely one exception (as required by
the collapse). Because of their dynamical autonomy, these global
branch wave functions describe "consistent histories" to their ob-
servers. In order to be subjectively and individually experienced,
a component of the global wave function seems to have to possess
the product form tPobsWrest, with a sufficiently complex and robust
observer state tPobs, and regardless of the existence of any cla118ical
observer variables. The observer is the only "system" that has to be
conceptually defined in principle.
200 Zeh

All we have then still to assume in order to explain Born's

probabilities in the form of observable frequencies of measurement
results is that we, described by certain correlated 1/Jobs 's, happen
to live in a branch that does not possess a very small norm. This
robust component would represent an "apparent reality" to us (in the
operational sense), that also forms the basis of Zurek's "existential
interpretation" r8].
It is usually overlooked that Bohm's theory contains the same
"many worlds" of dynamically separate branches as the Everett in-
terpretation (now regarded as "empty" wave components), since it is
based on precisely the same ("absolutely real") global wave function
[2]. Its robust components branch by means of decoherence (rather
than combining by means of recoherence) because of a fundamental
initial condition to the global wave function that is also responsi-
ble for the Second Law [9) - not because of "increasing knowledge"
about a classical state (the reduction of Bohm's ensemble). Only the
"occupied" wave packet itself is thus meaningful, while the assumed
classical trajectory would merely point at it: "This is where we are
in the quantum world." However, this can be done without using
a trajectory. Any pointing finer than compatible with the width of
the robust wave packet is empirically unjustified. John Bell (who
rejected Everett's interpretation for being "extravagant") seems to
have realized this equivalence before he began to favor spontaneous
localization (such as GRW) over Bohm's theory [10]. This is another
(historical) fact simply neglected by Goldstein when he heavily relies
on Bell's words in his arguments.

REFERENCES
1. S. Goldstein, Phys. Today 51, (March) 42, (April) 38 (1998).
2. J .S. Bell, in ~uantum Gravity 2, C. Isham, R. Penrose, and D.
Sciama, eds. Clarendon, Oxford, 1981 ).
3. Ph. Pearle, P ys. Rev. D 13, 857 (1976). G.C. Ghirardi, A. Ri-
mini, and T. Weber, Phys. Rev. D 34, 470 (1986).
4. E. Joos, Phys. Rev. D 36, 3285 (1987). M. Tegmark, Found.
Phys. Lett. 6, 571 (1993).
5. D. Giulini, E. Joos, C. Kiefer, J. Kupsch, 1.-0. Stamatescu, and
H.D. Zeh, Decoherence and the Appearance of a Classical World
in Quantum Theory (Springer, New York, 1996).
6. Ph. Pearle and E. Squires, Phys. Rev. Lett. 73, 1 (1994).
7. H.D. Zeh, Found. Phys. 1, 69 (1970).
8. W.H. Zurek, Phil. Trans. R. Soc. Lond. A 356, 1793 (1998).
9. H.D. Zeh, The Physical Basis of the Direction of Time, 3rd edn.
(Springer, Berlin, 1999).
10. J.S. Bell, private correspondence, 1981.