Found Phys (2012) 42:15341555

DOI 10.1007/s10701-012-9681-5
Quantum Decoherence in a Pragmatist View: Dispelling
Feynmans Mystery
Richard Healey
Received: 3 April 2012 / Accepted: 6 September 2012 / Published online: 21 September 2012
Springer Science+Business Media, LLC 2012
Abstract The quantumtheory of decoherence plays an important role in a pragmatist
interpretation of quantum theory. It governs the descriptive content of claims about
values of physical magnitudes and offers advice on when to use quantum probabili-
ties as a guide to their truth. The content of a claim is to be understood in terms of
its role in inferences. This promises a better treatment of meaning than that offered
by Bohr. Quantum theory models physical systems with no mention of measurement:
it is decoherence, not measurement, that licenses application of Borns probability
rule. So quantum theory also offers advice on its own application. I show how this
works in a simple model of decoherence, and then in applications to both labora-
tory experiments and natural systems. Applications to quantum eld theory and the
measurement problem will be discussed elsewhere.
Keywords Decoherence Quantum interpretation Pragmatist
1 Introduction
This paper aims to show how a pragmatist account of the content and function of
statements in quantum theory is able to use decoherence to dispel the mystery Feyn-
man located at the heart of that theory. By decoherence, I mean the delocalization
of phase of a quantum system through interaction with its environment. Feynman [1,
Vol. III, Chap. 1] located the mystery already in a two-slit (or hole) interference
experiment with individual particleshe chose electrons. Focusing attention on the
proposition (A) that each electron either goes through hole 1 or it goes through hole
2 [and not both], he rehearsed a familiar argument with the (false) conclusion that
no interference fringes will appear on a detection screen placed behind the holes. As
R. Healey ()
Philosophy Department, University of Arizona, Tucson, AZ 85721-0027, USA
e-mail: [email protected]
Found Phys (2012) 42:15341555 1535
well as (A), the argument assumes that there are well-dened probabilities Pr(R&1),
Pr(R&2) for an electron to be detected in region R of the screen after passing through
hole 1, 2 respectively, whose values are independent of whether the other hole is open
or closed. Feynman [1, 19] concluded
if one has a piece of apparatus which is capable of determining whether the
electrons go through hole 1 or hole 2, then one can say it goes through either
hole 1 or hole 2. [otherwise] one may not say that an electron goes through
either hole 1 or hole 2. If one does say that, and starts to make any deduc-
tions from the statement, he will make errors in the analysis. This is the logical
tightrope on which we must walk if we wish to describe nature successfully.
But how could the absence of a piece of apparatus revoke ones right to free
speech? Presumably although one can assert (A) in any circumstances, Feynmans
advice was that one should do so only when the apparatus is present, because only
then is (A) meaningful and conducive to correct inferences. But the mystery remains:
How can the presence of a piece of apparatus render (A) both meaningful and correct,
and what exactly is meant by the presence of such a piece of apparatus? I will offer
answers to these questions as a way of dispelling the mystery.
An alternative is to locate the aw in the familiar argument in its assumptions
about probabilities. Grifths [2] consistent histories approach permits one to assume
(A) only in a framework corresponding to a history set {C

} satisfying a consis-
tency condition to ensure a well-dened probability for each history (specied by a
time-indexed sequence of projectors, each taken von Neumann-fashion to represent
a corresponding property of a system at that time). On this approach, Feynmans ar-
gument breaks down because there is no framework in which Pr(R&1), Pr(R&2) are
well-dened. But there is a framework that assigns probabilities Pr(S&1), Pr(S&2)
to histories of an electron passing through hole 1 (hole 2) and then being detected
somewhere (S). In such a framework, (A) may be assigned probability 1 and taken
as meaningful and true.
Now we have another mystery: (A) is meaningless in a framework including a
history corresponding to passage through hole 1(2) and being detected in region R,
but meaningful and true in another framework containing no such histories. The con-
sistent histories approach supplies formal framework-relativized rules governing the
making of statements ascribing properties to systems like electrons, but offers no ac-
count of the content of such statements and does not explain the point of making
them, in conformity to the rules.
While noting the inuence of Grifths prior work, Gell-Mann and Hartle [3, 4]
(see also Omns [5, 6]) stress the importance of environmentally-induced decoher-
ence in rendering a family of histories (at least approximately) consistent (or deco-
herent in their rival usage of that term). Gell-Mann and Hartle present their view as
a development of an Everettian formulation of quantum theory from which a more
standard formulation may be retrieved, given appropriate assumptions.
1
Their main
motivation for imposing decoherence/consistency conditions on histories is to ensure
1
But note that the characteristically democratic Everettian talk of multiple real branches no longer ap-
pears in the later [7].
1536 Found Phys (2012) 42:15341555
the application of probabilistic rules of quantum theorynot just to measurement
outcomes, but to properties of quantum systems whether or not these happen to be
measured. If successful, their approach would dispel part of Feynmans mystery by
replacing talk of the presence of a piece of apparatus by an account of how interac-
tions within a closed system can decohere the state of a subsystem through its interac-
tion with the rest of the system. But they say no more than Grifths about the content
of a statement that such a subsystem has a property corresponding to a projection,
or how to reconcile such a statement with another statement that it has a property
corresponding to a non-commuting projection from an alternative decoherent set of
histories.
Two things are needed to dispel Feynmans mystery:
(1) An account of how a statement ascribing a property to a quantum system acquires
any content at all, and what content it has in different circumstances.
(2) An account of the use of the probabilistic rules of quantum theory that shows
how and why their use must be restricted so as to ensure consistency.
I offer account (1) in Sect. 3.2, and (2) in Sect. 3.3 of this paper. The accounts are
connected by the natural requirement that the probabilistic rules of quantum theory
can be applied only to statements with well-dened content. These accounts depend
on what I call a pragmatist interpretation of quantum theory that I outlined in [8].
This governs the descriptive content of claims about values of physical magnitudes
and advises an agent on when to apply the Born Rule as a guide to their truth.
But, on this interpretation, it does so without representing the dynamic behavior
of physical systems: for while it is quantum states that are subject to environmental
decoherence, the quantum state does not serve to represent or describe physical sys-
tems. Because the familiar view that quantum models of environmental decoherence
offer representations of a physical process conicts with a non-representational view
of the quantum state, I explain in Sect. 2 how the quantum state functions, according
to this pragmatist interpretation.
Section 3 then shows in a simple model how decoherence governs the content of
descriptive claims about a qubit. The content of a claimis to be understood in terms of
what inferences an agent may draw from that claim and what would entitle an agent
to make it. This inferentialist pragmatism about content promises a better treatment
of meaning than that offered by Bohr and his followers.
2
Quantum theory assigns probabilities to claims about the values of magnitudes
through the Born Rule. But it is now well established that these magnitudes cannot
consistently all be taken simultaneously to possess precise values on each system,
distributed over a collection of similar systems in such a way that the fractions of
systems with particular values for each magnitude match the corresponding proba-
bilities owing from the Born Rule.
3
Consistency then restricts each application of
the Born Rule to a proper subset of all magnitudes. Conventionally, one species
2
Here I am indebted to the writings of Brandom [9, 10] and Price [11, 12]. Bohr expressed his views in a
number of essays collected in Bohr [1315].
3
See, for example, [1620].
Found Phys (2012) 42:15341555 1537
probabilities only for measurement outcomes, and postulates that only those magni-
tudes represented by pairwise commuting self-adjoint operators are simultaneously
measurable. But Bell [18, 21] raised powerful objections against incorporating mea-
surement into the basic principles of quantum theory, either as a primitive term or
when cashed out in equally unsatisfactory terms such as irreversible amplication,
classical system, or conscious observation. The pragmatist interpretation of [8]
relies on quantum models of environmental decoherence that involve no reference to
measurement to secure consistent application of the Born Rule. Section 3 shows how
this works.
The general account of Sect. 3 is illustrated in Sect. 4 by applying it to some more
realistic examples, both in a laboratory setting and in the universe at large. These
are intended to show how the details of environmental decoherence can affect the
signicance of descriptive claims licensed by a quantum state, and to exhibit both the
practical use of the Born Rule and its limitations.
In conclusion the paper summarizes the role of decoherence within the pragmatist
interpretation of quantum theory outlined in [8], contrasting this with the very differ-
ent role allotted to it in a recent pragmatist-inspired paper [22]; and it notes questions
that still need to be answered to understand just how decoherence can play this role.
2 The Function of the Quantum State
The delocalization of phase in a systems quantum state due to interaction with sys-
tems constituting its environment is generally regarded as a physical process. Quan-
tum models compare and contrast this process with other, more familiar, physical
processes such as dissipation due to energy loss into the environment. Quantum de-
coherence need not be accompanied by dissipation, though when it is, it is typically
a much faster process. Some say that quantum decoherence occurs as a result of a
system acting on its environment, whereas dissipation occurs because the system is
acted on by its environment. Such language in which discussions of environmental
decoherence are couched is thoroughly physical. Master equations and other mathe-
matical treatments of quantum decoherence are taken to represent how the physical
condition of a system changes in response to its interaction with its environment
when that interaction is represented by an interaction Hamiltonian. But the immedi-
ate content of such treatments concerns the evolution of a systems quantum state.
To read a representation of the evolution of a quantum state as a description of the
changing condition of the system to which it pertains is to adopt a particular interpre-
tative stance toward quantum states. It is to assume that a quantum state provides a
description of the physical condition of a system to which it is assigned.
But this is only one, disputed, view of the function of the quantum state. On the
present pragmatist understanding,
4
the function of the quantumstate is not to describe
4
Fuchs [23, 24] is a pragmatist who has proposed an alternative understanding of a quantum state as
merely a bearer of subjective probabilities representing an agents actual degrees of belief. He takes these
to be restricted only by minimal rationality requirements (unfortunately also called coherence!) necessary
and sufcient to warrant their mathematical representation as a probability measure. Fuchss subjectivist
quantum state represents an agents beliefs: it does not prescribe them. In the papers conclusion I comment
on his very different take on the role of quantum decoherence.
1538 Found Phys (2012) 42:15341555
but to prescribe: Aquantumstate does not provide even an incomplete description of a
physical system to which is assigned. Instead, by assigning a quantum state, an agent
using quantum theory takes the rst step in a procedure that licenses it
5
to express
claims about physical systems in descriptive language and then warrants it in partially
believing some such claims. The language in which these claims are expressed is not
the language of quantumstates or operators, and the claims are not about probabilities
or measurement results: they are about properties of physical systems. In this paper
I restrict attention to properties of a system corresponding to projections onto the
systems Hilbert space: so if M is a magnitude on a system, then the property M
locating its value in Borel set corresponds to the projection

P
M
() from

Ms
spectral measure.
An agent must be physically situatable to use quantum theory because a quantum
state and consequent Born probabilities can be assigned to a system only relative to
the physical situation of an (actual or hypothetical) agent for whom these assignments
would yield good advice. What one agent should believe may be quite different from
what another agent in a different physical, and therefore epistemic, situation should
nd credible. This relational character of quantum states and Born probabilities does
not make these subjective, and it may be neglected whenever users of quantum theory
nd themselves in relevantly similar physical situations.
6
Claims about properties
of a system are also objective, but, unlike claims pertaining to quantum states and
Born probabilities, they are not relational in this way: Whether they are true does not
depend on the physical situation of any actual or hypothetical agent.
A quantum state is objective because it provides authoritative guidance to an agent
on two important matters. It provides sound advice both on the content of statements
about properties of physical systems and on the credibility of some of these state-
ments. Environmental decoherence gures in both these roles of the quantum state.
Section 3.2 shows how decoherence enables the quantum state to play the rst ad-
visory role: Sect. 3.3 is concerned with its contribution to the second. Note that this
pragmatist interpretation does not deny that environmental decoherence involves a
physical process: but it does deny that the role of a systems quantum state is to de-
scribe or represent properties of systems involved in such a process.
An agent requires guidance in assessing the content of statements about systems
of interest. In the passage quoted in the introduction, Feynman begins by applying
a common view of meaning in quantum theory: that assignment of a value to an
observable on a system is meaningful only in the presence of some apparatus capable
of measuring the value of that observable. But some general account of meaning must
be offered in support of this claim, and the extreme operationist account that is most
naturally associated with it would be unacceptably vague even if it were otherwise
defensible.
5
I use the impersonal pronoun throughout to remind the reader that on this view an agent need be neither
human nor even conscious. In these respects my usage of agent is similar to Gell-Mann and Hartles
usage of their term IGUS (for Information Gathering and Utilizing System).
6
The relational character of quantum states has been emphasized by others, including [25], who showed
how much different state assignments can differ.
Found Phys (2012) 42:15341555 1539
Contemporary pragmatist accounts of meaning have the resources to provide a
better account of the meaning of a statement about a property of a system. It is strik-
ing that such an account is suggested by the way Feynman [1, 1.9] qualies his initial
ban on asserting proposition (A). A pragmatist like Brandom [9, 10] takes the content
of any claim to be articulated by the inferences in which it may gure as premise or
conclusion. These inferences may vary with the context in which a claim arises, so
the content of the claim depends on that context. The quantum state of a system mod-
ulates the content of claims about its properties by specifying the context in which
they arise. This depends on the nature and degree of environmental decoherence suf-
fered by this quantum state. A statement about a property M of a system whose
quantum state has extensively decohered in a basis of eigenstates of

M has a corre-
spondingly well-dened meaning: a rich content accrues to it via the large variety of
inferences that may legitimately be drawn to and from the statement in that context.
Only when a claim M about a system has a rich content because it is fully
integrated into a web of reliable inferences is it appropriate to apply the Born Rule to
assess the claims credibility. An agent may then apply the Born Rule to evaluate the
probability of the claim using the appropriate quantum state. If [

P
M
(),

P
N
( )] =
0, then the Born Rule will never be applicable to both claims M, N at once,
since there is no context in which they both have a sufciently well-dened content.
3 The Content and Credibility of Property Claims
3.1 A Simple Model of Decoherence
Consider a simple model of decoherence introduced by Zurek [26] and further dis-
cussed in Cucchetti, Paz and Zurek [27]. This features a single quantum system A
interacting with a second environment system E as in [27].
7
A is a single qubit,
and its environment E is modeled by a collection of N qubits. One can think of each
qubit as realized by a spin
1
2
system, so that | (|) represent z-spin up (down)
eigenstates of the Pauli spin operator
z
of A, while |
k
(|
k
) represent z-spin up
(down) eigenstates of
k
z
for the kth environment spin subsystem.
The individual Hamiltonians

H
A
,

H
E
of A and E are assumed to be zero, while
the interaction Hamiltonian

H
AE
has the form

H
AE
=
1
2

z

N

k=1
g
k

k
z
. (1)
If A, E are assumed to be initially assigned pure, uncorrelated states

A
=
_
a| +b|
_
, (2)

E
=
N

k=1
_

k
|
k
+
k
|
k
_
(3)
7
Zureks original model also included a third system S: his choice of notation then was intended to help
his reader bear in mind an application of the model to a system S interacting with a quantum apparatus A.
1540 Found Phys (2012) 42:15341555
then the initial state
(0) =
A

E
(4)
evolves according to the Schrdinger equation, becoming
(t ) =
_
a|

(t )
_
+b|

(t )
__
(5)
at time t where

(t )
_
=
N

k=1
_

k
e
ig
k
t
|
k
+
k
e
ig
k
t
|
k
_
=

(t )
_
. (6)
The state of A, calculated by tracing over the Hilbert space of E, is therefore

A
(t ) =|a|
2
|| +ab

r(t )|| +a

br

(t )|| +|b|
2
||. (7)
The coefcient r(t ) =E

(t )|E

(t ) appearing in the off-diagonal terms of

A
here
is
r(t ) =
N

k=1
_
cos 2g
k
t +i
_
|
k
|
2
|
k
|
2
_
sin2g
k
t
_
. (8)
Cucchetti, Paz and Zurek [27] show that |r(t )| tends to decrease rapidly with increas-
ing N and very quickly approaches zero with increasing t . More precisely, while
|r(t )|
2
uctuates, its average magnitude at any time is proportional to 2
N
, and, for
fairly generic values of the g
k
, it decreases with time according to the Gaussian rule
|r(t )|
2
e

2
t
2
, where depends on the distribution of the g
k
as well as the initial
state of E. This result is relatively insensitive to the initial state of E, which need not
be assumed to have the product form (3), though if the environment is initially in an
eigenstate of (1) |r(t )| = 1 so the state of A will suffer no decoherence. Since r(t )
is an almost periodic function of t for nite N, it will continue to return arbitrarily
closely to 1 at various times: but for N corresponding to a macroscopic environment
Zurek [26] estimated that the corresponding recurrence time exceeds the age of the
universe.
3.2 The Content of Property Claims in This Simple Model
Suppose an agent is considering what claims to entertain about the system A in this
simple model. The agent is not explicitly represented in the model itself. But since any
assignment of quantum states is from the perspective of some (actual or hypothetical)
physically situated agent, we must assume the agent has implicitly adopted such a
perspective by assigning the states that gure in the model. Magnitudes pertaining to
A correspond to self-adjoint operators on the Hilbert space H
A
. In this simple model,
any such operator

Q may be expressed as a real linear sum of Pauli spin operators
and the identity operator on H
A
as follows

Q=x
x
+y
y
+z
z
+c

I. (9)
Found Phys (2012) 42:15341555 1541
So we are considering claims about A of the form K : Q , where is a Borel
set of real numbers, and Q corresponds uniquely to the operator

Q. After setting
S
i
(/2)
i
(i =x, y, z) these include
I =1, (A)
S
x
{+/2, /2}, (B)
S
x
=+/2, (C)
S
x
=/2, (C

)
S
z
{+/2, /2}, (D)
S
z
=+/2, (E)
S
z
=/2. (E

)
The agents primary interest is in its entitlement to believe a claim K about A, in-
cluding each of the claims (A)(E

). But its rst concern is what content is expressed

by such a claim, and so it should consult the quantum state of A.
Claim (A) is vacuous. It never warrants further claims about A. What is the content
of each of (B)(E

) given only the initial state (2) the agent assigns to A? That depends
on the inferential role of each claim. The agent may be tempted to infer claim (D)
about A from the fact that (2) expresses this state as a superposition of eigenstates
of
z
. But the initial state of A may be expressed equally well as a superposition of
eigenstates |, | of
x

A
=c| +d|, where c =
1

2
(a +b), d =
1

2
(a b) (10)
(or indeed of any operator

Q). So if the content of a claim of the form K then de-
pended only on the state (2) then our agent should be equally tempted to make claim
(B) (as well as every other similar claim assigning some eigenvalue of

Q to every
magnitude Q in that state).
The simple model represents no piece of apparatus capable of determining whether
(C) or (C

) is true, so Feynman would warn against saying (B) unless our agent de-
clines to make any inferences from (B). (B) is an exclusive disjunction, and the prob-
lematic inferences to be barred would proceed by deriving a (false) conclusion from
each disjunct separately and hence drawing that conclusion on the basis of the dis-
junction alone. In the two-slit experiment, the assumption that each electron goes
through one slit or the other leads to the false conclusion that the interference pattern
on the screen is the sum of a pattern formed by electrons going through slit 1 and a
pattern formed by electrons going through slit 2. But to derive that conclusion, one
needs further to assume that the behavior of an electron going through slit 1(2) is the
same whether or not slit 2(1) is openan assumption rejected by Bohmians, among
others.
This illustrates an important point. The inferences that contribute to the content of
a property claim like (B) are not restricted to mathematically and logically valid in-
ferences, but include what Sellars [28] called material inferences. Indeed, according
1542 Found Phys (2012) 42:15341555
to an inferentialist account of content like Brandoms [9, 10] it is precisely such ma-
terial inferences that contribute essentially to empirical content. But, as in this case,
what material inferences a claim licenses will depend on what other assumptions are
made.
The content of (C) and (C

) must be restricted so as to exclude their use even in

hypothetical material inferences when A is in state (2). Without such a restriction,
our agent could infer that S
x
,S
z
(and indeed all other spin components) have precise
real values together in state (2). While this is not a contradiction, it does conict with
generally accepted background assumptions.
8
State (2) is an eigenstate of the operator

(sin cos )
x
+(sin sin)
y
+cos
z
, (11)
where a = cos /2exp(i/2), b = sin/2exp(+i/2). This state may be repre-
sented on the Bloch sphere by a unit vector n with angular coordinates (, ). The
operator

S
n
= (/2)

corresponds to a component S
n
of angular momentum in
a spatial direction n with spherical coordinates (, ) dened with respect to the
(x, y, z) Cartesian coordinate system. Consider the claim
S
n
=+/2 (F)
Since (2) is an eigenstate of

S
n
with eigenvalue +/2, our agent may be tempted to
make claim (F) solely on the basis of that initial state assignment to A. But before
doing so, it should assess (F)s content.
Since the content of (F) is a function of its inferential role, the agent must con-
sider what could entitle it to infer (F) and what it could infer from (F). The agent
could immediately infer (F) from (2) in accordance with the interpretative principle
EigenState to EigenValue:
(SV) If a systems quantum state satises

P
i
= , where

P
i
projects onto
the eigenspace with eigenvalue q
i
of an operator

Q corresponding to magnitude
Q, then Q has value q
i
.
But the principle (SV) should be rejected as incompatible with the pragmatist de-
nial that a systems quantumstate provides any kind of description of it. Alternatively,
our agent might think to infer (F) using the EPR [29, p. 777] sufcient condition of
reality.
If, without in any way disturbing a system, we can predict with certainty (i.e.
with probability equal to unity) the value of a physical quantity, then there
exists an element of reality corresponding to this physical quantity.
The thought might be that application of the Born Rule to (2) would assign proba-
bility unity to the claim S
n
=+/2, and then (F) follows from (EPR)s reality condi-
tion. But this thought is mistaken whatever the status of that criterion. One is entitled
to apply the Born Rule to state (2) and assign a probability (unity) to (F) only if (F)
8
A similar conclusion in the 3-dimensional Hilbert space of a spin 1 system would be inconsistent with
Gleasons theorem [16].
Found Phys (2012) 42:15341555 1543
has sufcient content to permit that application. But the empirical content of (F) is
exactly what is in question. Clearly, it would be circular to assume that (F) has suf-
cient content to entitle one to apply the Born Rule to (F) in state (2) in order to argue
that (F) has any signicant empirical content! In fact the Born Rule is not applicable
to this claim in the simple model, in which the only interaction to which A is subject
is modeled by (1): to be entitled to apply the Born Rule to assign a probability to
(F) at t = 0, our agent would have to realize that A is subject to an interaction that
decohered eigenstates of

S
n
at t =0.
In state (2) one may infer from (F) to any claims validly deducible by logic and
mathematics alone, such as these:
S
n
{+/2, /2},
(S
n
=+/2) or (S
n
=+),
S
2
n
=
2
/4.
So it is not strictly correct to say that (F) is vacuous in this case. But in order to have
any physical content, (F) would have to permit material inferences that are neither
logically nor mathematically valid.
In classical physics, property claims typically permit material inferences of two
kinds: dynamic inferences and measurement inferences. An assumption of continuity
guarantees that ascription to a magnitude of a value in set at time t licenses a mate-
rial inference to its value in set

at t +, where

is close to for sufciently

small : and ascription to a magnitude of a value in set at time t licenses a material
inference that the result of a sufciently carefully conducted measurement at t +
would nd a value in

. Since our agent can make neither kind of material inference

from (F), (F) lacks physical content hereit is empirically vacuous, as are (C) and
(C

). The initial state (2) licenses an agent to make no physically signicant claims
about A.
But the agents resources are not conned to the assignment of an initial state to
A. Using this simple model of decoherence, our agent also species how the initial
quantum state of A evolves under the inuence of interaction with its environment.
It is the role of decoherence here that endows certain claims about A with empirical
signicance, according to the interpretation outlined in [8].
The quantum state initially assigned to A evolves, so that after a remarkably short
time T it will come to approach the diagonal form

A
=|a|
2
|| +|b|
2
||. (12)
If =0 or (i.e. |a|
2
or |b|
2
=1) in state (2), then claim (F) reduces respectively to
(E) or to (E

). Each of these is now physically signicantnot because of the initial

state (2), but as a result of how the simple model treats As interaction with E. It
is only because of this environmental decoherence that our agent can entertain any
physically signicant claims about As properties: these include (D), (E) and (E

), but
not (A)(C

).
1544 Found Phys (2012) 42:15341555
While (A)(C

) remain empirically vacuous in state (12), (D), (E) and (E

) have
empirical content because of the material inferences each supports. Our agent may
use (D) in inferences that assume that just one disjunct is true in state (12), even
with no empirical basis for claiming (E) as against (E

) (or vice versa). The material

inference to (D) at time T is not justied by the state of A alone, but because the
state of A continues to remain very close to (12) for an extended interval including
T , and is in that sense stable against the environmental interaction our agent models
by (1). Although principle (SV) is false, a pragmatist interpretation of the quantum
state does endorse this particular consequence of (SV) because at T each of (E), (E

)
permits the dynamic and measurement inferences discussed three paragraphs back.
This illustrates the importance of environmental decoherence in endowing a prop-
erty claim with empirical content. Deployment of the simple model of decoherence
cannot by itself justify an agent in making a material inference either to (E) or to (E

).
Rather, as Sect. 3.3 explains, assignment of state (12) to A in the context of this model
justies the practical inference involved in adopting degree of belief (credence) |a|
2
in (E) and credence |b|
2
in (E

). On this pragmatist interpretation it is a basic assump-

tion of any application of a quantum model that an agent can subsequently come to be
warranted in believing a claim like (E) as against (E

) (or vice versa) by experience.

It follows that the process of observation or experimentation which would give rise
to an agents experience after applying a quantum model can nowhere be represented
within the model that agent is applying. It does not follow that this process cannot be
modeled by another agent, although this second model could not then be extended to
encompass processes that included that agents experiences.
Environmental decoherence is not perfect, even in this simple model. As it evolves,
the state of A arrived at by tracing over E in state (5) will be exactly diagonal in some
orthogonal basis of eigenstates of

for almost always extremely close, but not

equal, to zero. (Here varies over the angle of inclination to the z axis, and over
the azimuthal angle from the x axis). Consider a claim about A of the form L for
some pair (, ):
S

{+/2, /2} L
If is close enough to zero, then (1) will very rapidly, and quite stably, bring the state
of A almost as close to diagonal in a basis of eigenstates of

S

as of

S
z
. A material
inference to a claim of the form L in the context of the model will be almost as good
as the inference to (D), and the inferential power of a claim of the form L will be
almost as great as that of (D). More generally, the empirical content of a claim of the
form L here is a function of , varying continuously from its maximum value for
=0, to zero for =/2. It corresponds to the reliability of the inference from
the claim that S

has one of its eigenvalues at time t to the conclusion that S

has
that same eigenvalue at t + and that this would also be the result of a well-conducted
measurement of S

at time t +.
What is the relation between the interaction modeled by (1), the possibility of
measuring the value of a magnitude Q in the simple model, and what it takes to have
a piece of apparatus which is capable of determining the value of a magnitude Q? The
model itself makes no mention of any apparatus or measurement. But in applying the
model, an agent is effectively committed to counting the interaction modeled by (1) as
Found Phys (2012) 42:15341555 1545
itself a potential measurement of S
z
that excludes measurements of other magnitudes
S

with far from 0, while simultaneously serving as a somewhat less reliable

measurement of magnitudes S

with very close to 0, . An agent makes this

commitment by taking it for granted that the process being modeled had or will have
a determinate outcome it could come to recognize as indicating the value of S
z
by
examining either A or some part of E. The agent is thereby committed to regarding
the whole system being modeled as effectively including an apparatus capable of
determining the value of S
z
, but excluding any apparatus capable of determining the
value of any magnitude S

with far from 0, .

3.3 The Born Rule in the Simple Model
The references of footnote 3 show there is no consistent simultaneous assignment of
a Born probability to every statement ascribing a precise value to a magnitude on a
system. According to [8], the function of the Born Rule is to advise an agent on how
strongly it should believe certain claims. Since an agent can attach some degree of
belief only to an empirically signicant claim, this immediately restricts applications
of the Born Rule to empirically signicant property claims.
As Sect. 3.2 made clear, not every property claim concerning a system is equally
empirically signicant even when that system is subject to environmental decoher-
ence: Empirical signicance comes in degrees here. As varies, the empirical sig-
nicance of a claim of the form L varies accordingly, as does that of any claim about
A of the form M:
S

. M
Empirical signicance has no natural cut-off here, or in any application of quantum
theory. So this restriction on application of the Born Rule does not yield a precise
selection criterion. This is a classic case of vagueness. The Born Rule is clearly ap-
plicable to claims (D), (E), (E) in the simple model, and clearly inapplicable to claims
(A), (B), (C), (C

). But the limits of applicability of the Born Rule to a claim of the

form L or M may be set anywhere within a wide (but equally indeterminate) range
of values of in the neighborhoods of 0, . Does this vagueness matter?
Some take quantum theory to be fundamental because it provides our most ac-
curate descriptions of naturecall this fundamental
a
. But Bell [18, pp. 125126]
criticized contemporary formulations of quantum theory on the grounds that these
are fundamentally approximate and intrinsically inexact. Surely, he asked in [21],
after 62 years we should have an exact formulation of some serious part of quantum
mechanics? Bell denied that quantum theory is fundamental
a
, because contempo-
rary formulations are in terms of observables rather than what he called beables:
It is not easy to identify precisely which physical processes are to be given the
status of observations and which are to be relegated to the limbo between one
observation and another. So it could be hoped that some increase in precision
might be possible by concentration on the beables, which can be described in
classical terms, because they are there. [18, p. 52].
Bell would surely have rejected the pragmatist interpretation of [8]. He would have
taken the vagueness inherent in the conditions of applicability of the Born Rule to
1546 Found Phys (2012) 42:15341555
introduce an unacceptable imprecision into the theory, and regarded quantum theory
under this interpretation as not a serious theorya serious candidate for the job of
truly describing nature. But, in this pragmatist view, quantum theory achieves its
unprecedented success without describing nature, either vaguely or precisely.
Interpreted along the lines of [8], an agent does not use a quantum-theoretic model
to represent physical systems: quantum theory is not itself in the business of describ-
ing physical reality. Since quantum theory does not yield descriptions of nature, it
is clearly not a fundamental
a
theory. But it is fundamental in another sense: it gives
us our best and only way of predicting and explaining a host of otherwise puzzling
phenomena. We do this using quantum modelsnot to describe reality but to advise
us on what to believe about it. Any such use depends on application of the Born Rule.
So the predictive and explanatory successes we achieve using quantum theory depend
on judicious application of that rule.
Now one can see why any vagueness associated with application of the Born Rule
does not matter. Application of a theory or rule always requires judgment, and this is
no exception. In applying any physical theory one must rst decide how to model the
part or aspect of the physical world on which the application is targeted. The model of
Sect. 3.1 was called simple because it has few if any real world targetsa wise agent
would rarely if ever decide to apply it. When applying a quantum-theoretic model, an
agent must make a further decision about which claims are apt for application of the
Born Rule. Here, too, good judgment is called for.
Models of quantum theory are not inherently imprecise. Their specication need
contain none of Bells [18, p. 215] proscribed words measurement, appara-
tus, environment, microscopic, macroscopic, reversible, irreversible, ob-
servable, information or measurement, though one may use any of these words
harmlessly in commenting on the model with a view to its intended applications, as
several of these words were used in Sect. 3.1.
9
Any element of imprecision or in-
exactness can enter only when a quantum model is applied to a specic physical
situation.
The Born Rule itself is in no way imprecise or inexact: specically, a statement of
the Rule should not contain measurement or any other similarly problematic terms
The Born Rule simply assigns a mathematical probability measure to all claims about
A of the form K for every single magnitude Q in the simple model. In a model with a
higher-dimensional Hilbert space for A, the Born Rule also assigns a joint probability
measure to sets of claims {K
i
}, where the corresponding

Q
i
pairwise commute. In
applying the model, an agent needs to judge which of these mathematical measures
should be taken to govern credence and which lack cognitive signicance in this
application. Previous experience, as ltered through vague categories such as those
criticized by Bell may improve this judgment. But, just as in classical physics, quan-
tum theory can help structure the agents deliberation by making available enlarged
models that take account of the interaction of the target system with its environment,
whether this is thought of as an experimental arrangement or just the natural physical
9
It does seem necessary to use a word like system to say what is ascribed a quantum state in a quantum
model. But with no mention of any apparatus, the model cannot enshrine the shifty split between system
and apparatus of which Bell complained.
Found Phys (2012) 42:15341555 1547
situation in which the target system nds itself. In application, quantum theory is no
more inexact than classical physics.
4 Examples of Property Claims and Use of the Born Rule
In this section, I extend the discussion to more realistic examples. Laboratory exper-
iments in a controlled environment provide the clearest examples. But the universe
at large supplies an environment for natural processes that furnish some of the most
interesting applications of quantum theory.
4.1 Molecular Interference Lithography of C
60
Juffman et al. [30] prepared a beam of C
60
molecules with well-dened velocity v,
passed them through two gratings of a Talbot-Lau interferometer in a high vacuum,
and collected them on a carefully prepared silicon surface placed at the Talbot dis-
tance. They then moved the silicon about a meter into a second high vacuum cham-
ber and scanned the surface with a scanning tunneling electron microscope (STM)
capable of imaging individual atoms on the surface of the silicon. After running the
microscope over a square area of approximately 2 m
2
they were able to produce an
image of some one to two thousand C
60
molecules forming an interference pattern.
They reported that the surface binding of the fullerenes was so strong that they could
not observe any clustering, even over two weeks. Clearly they felt no compunction in
attributing very well dened, stable, positions to the molecules on the silicon surface,
and even recommended developing this experiment into a technique for controlled
deposition for nano-technological applications.
Assuming the process is stationary, one can assign fullerenes in the beam a quan-
tum state at each position z along the interferometer axis by replacing t in the
Schrdinger equation by z/v. To account for the interference pattern in this experi-
ment, one can then use the Schrdinger equation to calculate the quantum state at the
silicon surface and apply the Born Rule to calculate a fullerene probability density at a
particular location at distance x from the interferometer axis in a direction perpendic-
ular to the slit orientation. Application of the Born Rule would lead one condently
to expect formation of the observed interference pattern for a large enough number
of fullerenes, while acknowledging that this condence falls short of certainty.
There was no mention of decoherence in this account, which simply assumed one
can apply the Born Rule to claims of the form x for fullerenes at the silicon
surface. But the account involved no such application to fullerenes passing through
the interferometer from the oven to the surface. The title of Juffman et al. [30] points
to a common way of talking about single-particle interference experiments like this.
One says that a C
60
molecule acts like a particle at the surface (so it is meaningful to
ascribe it a determinate position there) but a wave in the interferometer (so it is mean-
ingless to say that it passed through just one slit of the central grating). The pragmatist
interpretation outlined in [8] endorses this way of talking, but only as a gloss on the
more nuanced account made possible by the application of an inferentialist view of
empirical content within a quantum model of decoherence.
1548 Found Phys (2012) 42:15341555
It is impracticable to formulate and solve the Schrdinger equation for the entire
many-body quantum interaction that begins with the binding of a C
60
molecule to the
silicon surface. It is clear that this will rapidly and strongly couple the C
60
molecule to
an environment of an exponentially increasing number of degrees of freedom, even-
tually involving the entire laboratory and beyond. But it is not unreasonable to apply a
canonical model of quantum Brownian motion in which the center of mass x-position
of a C
60
molecule is linearly coupled to a bath of harmonic oscillators corresponding
to the modes of the entire silicon crystal to which it is bound by an assumed simple
harmonic potential. In this model, the relevant degree of freedom of the molecule
picks out a system A interacting with an environment system E modeling the silicon
surface. This model has been studied by Paz, Habib and Zurek [31] and others.
10
An
agent may appeal to this model in assessing the empirical signicance of property
claims concerning a C
60
molecule at the silicon surface, and to justify application of
the Born Rule to certain of these, even while acknowledging that a more complex
model might offer wiser counsel.
In this case, the model of quantum Brownian motion shows that after a remark-
ably short time the quantum state of A will be expressible as a mixture of narrow
Gaussians, each approximating a point (x, p
x
) in classical phase space. Moreover,
the weights of these states will be, and will for a long time remain, equal to the cor-
responding phase space probability densities as calculated from the Wigner functions
of such Gaussians. An agent may therefore associate a high degree of empirical con-
tent with a claim locating a C
60
molecule at a particular place on the silicon surface,
and is therefore entitled to apply the Born Rule to claims of the form x , p
x

using the quantum state deduced by applying the Schrdinger equation to calculate
how the initial quantum state of a C
60
molecule evolves before it reaches the silicon
surface. The high empirical content of such claims follows from the justiability of
material inferences to claims about values of (x, p
x
) at different times, as attestable
by repeated measurements by the STM.
In the experiment of Juffman et al. [30], the center of mass fullerene wave-function
suffered negligible decoherence in the interferometer. So here an agent should assign
claims of the form x about C
60
molecules minuscule empirical content before
they reach the silicon surface and so decline to apply the Born Rule to them. The
next experiment is interesting precisely because it incorporates just such decoherence
within the interferometer.
4.2 Inuence of Molecular Temperature on C
70
Coherence
Hackermller et al. [33] investigated the effects of increased temperature in matter
wave interferometer experiments in which fullerenes lose their quantum behavior
by thermal emission of radiation. They prepared a beam of C
70
molecules of well-
dened velocity, passed them through two gratings of a Talbot-Lau interferometer in
a high vacuum, and detected those that passed through a third movable grating set at
the appropriate Talbot distance and used as a scanning mask, by ionizing them and
collecting the ions at a detector. Each molecule is sufciently large and complex to
10
Schlosshauer [32] gives a more recent review.
Found Phys (2012) 42:15341555 1549
be assigned a temperature as it stores a considerable amount of energy in its internal
degrees of freedom.
11
Interaction with the electromagnetic vacuum may result in
emission of photons with an intensity and frequency that increases as the internal
temperature is raised. Entanglement between such photon states and the state of the
emitting molecule tends to induce environmental decoherence.
Hackermller et al. [33] present a theoretical model of this decoherence that
ts their observations quite well, as the observed interference dies away when the
molecules temperature is raised from 1000 K to 3000 K. Hornberger, Sipe and Arndt
[34] give a more detailed exposition. An agent can use this model to assess the em-
pirical signicance of property claims concerning a C
70
molecule as it traverses the
interferometer. The model treats photon emission by a fullerene as a sequence of in-
dependent, separate events, analogous to collisions with gas particles (which occur
relatively rarely, given the extremely low pressure inside the interferometer). Since
the process is stationary, the reduced quantum state for the C
70
molecules center of
mass degree of freedom may be expressed as a function of position R in the interfer-
ometer as

(R
1
, R
2
) =(R
1
, R
2
)(R
1
R
2
) (13)
where the decoherence function (R
1
R
2
) represents the effect of decoherence on
the off-diagonal elements of the un-decohered reduced state (R
1
, R
2
). (It has no
effect on the diagonal elements, since lim
R
1
R
2
(R
1
R
2
) =1.)
Assuming photon emission is both isotropic and independent of R, is a function
only of the spectral photon emission rate. This is not the same as for a macroscopic
black body since the emitting particle is small and not in thermal equilibrium with a
heat bath, and the emission is assumed to take place at a xed internal energy rather
than temperature. Given the assumption that the emitting molecule has a denite
energy E, it can be associated with a microcanonical temperature T

given in terms
of the entropy S(E) by
T

(E) =
_
S(E)
E
_
1
. (14)
So to assign an internal temperature to a beam of C
70
molecules at each point in
the interferometer one must assume that each is always in a state of denite energy
except during photon emission, which is associated with a transition between energy
states. But what justies this assumption?
While it is decoherence of their center of mass quantumstate that is the focus of the
model, one must also consider decoherence of the internal state of the C
70
molecules.
Here a different model of decoherence is more appropriate. While the fullerenes in-
teract strongly with the electromagnetic eld of the laser beams that heat them before
entering the interferometer, their electromagnetic interactions inside it are very weak
at room temperature. Paz and Zurek [35] showed that in this quantum limit, the
reduced internal quantum state of a system rapidly becomes approximately diago-
nal in a basis of energy eigenstates as result of its interaction with the environment.
This is what justies an agent in claiming that each fullerene has a denite internal
11
In fact, assignment of a temperature to each molecule is a step that requires justication, as we shall see.
1550 Found Phys (2012) 42:15341555
energy within the interferometer, which may change if it emits a photon. So decoher-
ence plays a double role here. The model of (center of mass) decoherence relevant to
assigning empirical content to claims concerning the position of a fullerene in the in-
terferometer shows howthat decoherence depends on the fullerenes temperature. But
the justication for assigning empirical content to a claim asserting such dependence
rests on an independent model of the fullerenes (internal energy) decoherence.
12
Hackermller et al. [33] detected their fullerenes by ionizing them after the scan-
ning mask and measuring the intensity of detected ions. The detection process in-
volves focusing any ions produced on a conversion electrode, then detecting the emit-
ted electrons. How far does this indirect method of observing the interference pattern
affect the application of the Born Rule and the assignment of content to claims of
the form x about fullerenes at the scanning mask? As [34] shows, the geometry
of the interferometer in the experiment of Hackermller et al. [33] is such that the
probability density for fullerene x-position at the scanning mask predicted by unre-
ective application of the Born Rule to the un-decohered reduced state (R
1
, R
2
)
corresponds to a smoothed image of the rst gratinga pattern with the same period
d as the rst grating, with maximum intensity at the center of the image of a slit win-
dow in the rst grating and minimum intensity at the center of the image of a wall in
the rst grating.
The detector system employed by Hackermller et al. [33] operates by measuring
the ionization intensity for all C
70
molecules passing through the scanning mask at a
particular x-setting. So it is insensitive to through which slit in that third grating any
particular fullerene may (or may not) have passed. Since no apparatus is capable of
detecting through which slit of the second or third grating a fullerene passes, Feyn-
man would forbid one to say the fullerene passed through slit 1 or . . . or slit i or . . .
or slit N
j
of either the 2nd or 3rd (scanning mask) grating (j =2, 3).
On the present pragmatist interpretation, because no signicant decoherence oc-
curred at either grating, a claim of the form x
i
about a fullerene at the second or
third grating has little or no empirical content, where
i
species the opening interval
of the ith slit of either grating (i =1, 2, . . . , N
j
). The exclusive disjunction
(x
1
) . . . (x
i
) . . . (x
N
3
) (15)
regarding the position of a fullerene at the scanning mask therefore also lacks the em-
pirical content required to license its use as a premise in any inference. ((15) asserts
that exactly one (x
i
) (i =1, . . . , N
3
) holds.)
Since the object of the experiment of Hackermller et al. [33] is to investigate the
effect of thermal decoherence on the fringes produced by quantum interference of
single fullerenes, it is obviously important to be able to justify application of the Born
Rule to both decohered and un-decohered reduced states of fullerenes to compare
12
There is a further subtlety here, since no claimabout the entropy S or the microcanonical temperature T

dened in terms of it is a property claim (of the form M ). The entropy is a function of the quantum
state given by the von Neumann expression S = Tr( log), and so the interpretation of [8] denies
that a claim assigning a value either to S or to T

is a property claim. But both S and T

are still just as

objective as the quantum state assignment , and a property claim ascribing a value to E has objective and
well-dened content.
Found Phys (2012) 42:15341555 1551
their predicted detection intensities. However, neither a claim of the form x
i
nor
an exclusive disjunction (15) of such claims has enough empirical content to license
application of the Born Rule to such a claim about a fullerene at the third grating.
Yet there is a related statement to which the Born Rule may perhaps be justiably
applied, namely the inclusive disjunction
(x
1
) . . . (x
i
) . . . (x
N
3
). (16)
One could use a claim of this form to say that a fullerene passed through the third
grating, without thereby implying that it passed through some particular slit in that
grating. Such a claim has some inferential power in this situation. It supports the ma-
terial inference to the true conclusion that fullerenes will be detected by the ionization
detector if, but only if, the slit windows in the third grating are not blocked: One is
entitled to make the claim if, but only if, a beam of fullerenes emerges from the oven
(as conrmed by ionization detectors placed in front of the rst grating to measure
the initial beam temperature). But one may question whether these inferences alone
endow (16) with enough empirical content to permit one to apply the Born Rule here,
since interaction with the third grating produces little decoherence in the fullerenes
quantum state.
The blue laser beam used to ionize fullerenes after the third grating does interact
strongly with them and substantially decoheres the beam, effectively localizing ion-
ized fullerenes. At this point, a claim of the form x
i
about an ionized fullerene
has acquired a high degree of empirical content, and so one is clearly entitled to apply
the Born Rule to assign a probability to it, as well as to (15) and (16). But the rule
is here applied to the quantum state after the third gratinga state that played no
part in the analysis of the experiment! Ionization-induced decoherence is irrelevant
to an application of the Born Rule to the quantum state of the fullerenes at the third
grating.
If one remains dissatised by the justication for applying the Born Rule offered
two paragraphs back, another approach is available. This is to focus instead on appli-
cation of the Born Rule to a claim of the form x
TOT
, where
TOT
is an interval
covering the whole range of x-positions where the blue laser is capable of ioniz-
ing fullerenes so that the resulting ion can impact the detector electrode and elicit a
detection signal. Localization of ionization events gives a high degree of empirical
signicance to such a claim about a fullerene in an ionization event. The experiment
can be taken to show directly how the frequency of these events varies with the x-
setting of the scanning mask when the fullerenes are heated to a specic temperature
before entering the interferometer. One can redescribe such a frequency as a measure
of the probability of detecting a fullerene at an x-position in a slit window rather
than a wall in the scanning mask while denying the signicance of the claim that
the fullerene had any x-position as it encountered the scanning mask. This guarded
way of speaking is common when quantum physicists discuss applications of the
Born Rule. It was subjected to withering criticism by Bell [18]. But Bells objec-
tions do not apply when such talk is cashed out in terms of empirically signicant
claims to which the Born Rule may be justiably applied, such as x
TOT
in this
example.
1552 Found Phys (2012) 42:15341555
4.3 Quantum Theory in the Universe
Quantum theory has been applied successfully to a wide range of terrestrial phe-
nomena outside the laboratory. Such applications have already provided the basis for
a thriving technology. We have overwhelming reasons to believe that some natural
phenomena to which quantum theory has been successfully applied occurred long
before there were laboratories, physicists or any other agents capable of observing or
applying quantum theory to them: we use the quantum theory of radioactive decay to
date them!
Quantum theory has been successfully applied to yield an understanding of such
extraterrestrial phenomena as the structure of the sun and many other kinds of stars in
our galaxy and far beyond, as well as their modes of energy production, nucleosynthe-
sis, birth and death. Quantum theories of the Standard Model have been successfully
applied to give us a detailed, quantitative understanding of the evolution of matter
in the early universe. Quantum theories have been applied (albeit as yet more specu-
latively) to predict the existence and nature of radiation from black holes and to the
development of large-scale structure in the extremely early universe. How is quantum
decoherence relevant to such applications of quantum theory in the wild?
Much of what we know about the solar system, and almost everything we know
about what lies outside it, is based on evidence provided by analyzing electromag-
netic radiation, especially that emitted or absorbed by excited atoms and molecules.
Since quantum theory provides us with our best understanding of the processes in-
volved in the emission and absorption of radiation by atoms and molecules, it is
only by applying quantum theory to phenomena that occurred far away (and in many
cases long ago) that we can justify knowledge claims based on this evidence. No sin-
gle, simple model of decoherence can be expected to encompass all such phenomena.
But in many cases the atoms and molecules involved will be in an environment that
decoheres their internal states in an energy basis (cf. the discussion of emission by
the C
70
molecules in Sect. 4.2). It is such decoherence that justies one in assuming
that emission or absorption occurs between states of well-dened internal energy, and
so applying the Born Rule to calculate absorption or emission probabilities.
Environmental decoherence in an energy basis provides a similar justication for
application of the Born Rule to calculate rates of nuclear reaction and decay in stellar
nucleosynthesis. Here it is the nuclear energy levels that decohere in consequence of
environmental interactions.
Applications of the Standard Model to primordial nucleosynthesis in the rst three
minutes of the Big Bang have successfully accounted for observed cosmic abundance
of helium, deuterium and lithium-7. These applications depend on calculations of
rates for weak interaction processes including conversion of a proton and electron
into a neutron and neutrino. Such a calculation proceeds by applying the Born Rule
in the context of perturbation theory via Fermis golden rule (or some elaboration to
take account of higher order terms in a perturbation expansion). This application may
once again be justied by the decoherence due to interaction with the early universe
environment provided by the highly excited state of the quantized electromagnetic
eld. A companion paper will discuss quantum eld-theoretic models of decoher-
ence.
Found Phys (2012) 42:15341555 1553
5 Summary and Outlook
If a quantum state does not describe or represent physical properties of a system to
which it is assigned, it must have some other function. In the pragmatist interpretation
outlined in [8] it has one role within a model of quantum theory and a second role
in guiding the application of a model. Within a model, the quantum state functions
as input to the Born Rule for calculating probabilities of claims of the form M .
There is no mention of measurement, either in these claims or in the Born Rule itself.
Within a model, Bells [18, 21] requirement is metthat the theory should be fully
formulated in mathematical terms, with nothing left to the discretion of the theoretical
physicist. However, in order to apply a quantum model it is necessary to assess the
signicance of claims concerning the system(s) which are the intended target of the
application.
Physicists have acquired practical expertise in this task, guided by informal max-
ims like Wheelers No phenomenon is a real phenomenon until it is an observed
phenomenon and by Bohrs view that the entire experimental setup provides us with
the dening conditions for the application of classical concepts in the domain of quan-
tum physics. But while the application of models in physics always requires skill and
judgment, Bell [18, 21] was surely right to complain about reliance here on such
vague and anthropocentric terms as measurement, observation and experimental
setup.
There is nothing anthropocentric about environmental delocalization of coherence.
In the pragmatist interpretation outlined in [8] quantum models of decoherence gov-
ern the signicance of property claims and their suitability for application of the Born
Rule. The extension of a quantum model of a target system to include its environment
yields a principled and non-anthropocentric way of using quantum theory to guide its
own application.
This does not eliminate the need for judgment and discretion. Existing quantum
models of environmental decoherence incorporate many idealizations and/or approx-
imations known to hold only in highly controlled laboratory experiments. Choosing
one model rather than another as a basis for assessing the signicance of property
claims and their suitability for application of the Born Rule requires skill and judg-
ment, especially when the intended target system is in an uncontrolled and epistemi-
cally inaccessible environment. By using an inappropriate model of decoherence, an
agent might come to apply the Born Rule to an unsuitable property claim. But any
such mistake would be subject to correction by standard scientic methods.
By contrast, there is no foundational place for decoherence in the Quantum
Bayesian program of [22] (see also [23, 24]) which regards decoherence not as an ex-
ternal physical process but as an agents internal application of a rationality constraint
(the Reection Principle) on what he believes about how his beliefs will change on
acquiring data through acting on the world. The Quantum Bayesian agrees with the
pragmatist of [8] that the evolving quantum state of a system interacting with its en-
vironment does not represent changing properties of that system. But he disagrees by
taking it to represent an individual agents decisions to act and consequential experi-
ences.
1554 Found Phys (2012) 42:15341555
In the pragmatist view of [8], on the other hand, the function of the evolving
quantum state is to advise a situated agent on the content and credibility of prop-
erty claimstypically those the agent is not currently in a position to check for itself.
The quantum theory of decoherence plays the important role of modeling the evolu-
tion of the quantum state so an agent can elicit this advice. Neither property claims
nor quantum states represent the agents actions or experiences, and there is no talk
of measurement, either in a model of decoherence or in the Born Rule.
The Quantum Bayesian offers no account of what gives a claim about a systems
property its content, and how that content depends on the situation of the system. By
taking the subject matter of quantum theory to be restricted to an individual agents
decisions and experiences, the Quantum Bayesian is unable to account for the the-
orys application to situations such as those mentioned in Sect. 4.3 where no decision
is in question and nothing will be experienced. Quantum Bayesianism would deprive
us of much of the descriptive and explanatory power quantum theory gives us. In the
pragmatist view of [8], decoherence is not merely important in understanding quan-
tum measurement: It plays a foundational role in guiding our beliefs about systems
we have not measured or cannot measure because of our physical and spatiotemporal
limitations.
The foundational signicance of environmental delocalization of coherence in the
pragmatist interpretation outlined in [8] makes it important to improve our under-
standing by constructing, analyzing and (if possible) testing new and more realistic
models. To my knowledge, there has not been much interest in modeling the envi-
ronmental decoherence suffered by astrophysical or cosmological systems including
those mentioned in Sect. 4.3. This attitude may seem justiable if quantum theory
is simply a tool for calculating probabilities for outcomes of laboratory experiments.
But it is indefensible from the pragmatist view of quantum theory outlined in [8].
Acknowledgements This publication was made possible through the support of a grant from the John
Templeton Foundation. The opinions expressed in this publication are those of the author and do not
necessarily reect the views of the John Templeton Foundation.
References
1. Feynman, R.P.: The Feynman Lectures on Physics. Addison-Wesley, Reading (1963)
2. Grifths, R.B.: Consistent Quantum Theory. Cambridge University Press, Cambridge (2002)
3. Gell-Mann, M., Hartle, J.B.: Quantum mechanics in the light of quantum cosmology. In: Ezawa, H.,
Murayama, Y., Nomura, S. (eds.) Proceedings of the 3rd International Symposium on the Foundations
of Quantum Mechanics in the Light of New Technology. Physical Society of Japan, Tokyo (1990)
4. Gell-Mann, M., Hartle, J.B.: Quantum mechanics in the light of quantum cosmology. In: Zurek, W.H.
(ed.) Complexity, Entropy and the Physics of information. Addison-Wesley, Redwood City (1990)
5. Omns, R.: The Interpretation of Quantum Mechanics. Princeton University Press, Princeton (1994)
6. Omns, R.: Understanding Quantum Mechanics. Princeton University Press, Princeton (1999)
7. Gell-Mann, M., Hartle, J.B.: Decoherent histories quantum mechanics with one real ne-grained
history. arXiv:1106.0767v3 [quant-ph] 28 Nov. (2011)
8. Healey, R.: Quantum theory: a pragmatist view. Br. J. Philos. Sci. (2012). doi:10.1093/bjps/axr054
9. Brandom, R.: Making it Explicit. Harvard University Press, Cambridge (1994)
10. Brandom, R.: Articulating Reasons. Harvard University Press, Cambridge (2000)
11. Price, H.: Naturalism Without Mirrors. Oxford University Press, Oxford (2011)
12. Price, H.: Facts and the Function of Truth. Blackwell, Oxford (1988)
Found Phys (2012) 42:15341555 1555
13. Bohr, N.: Atomic Theory and the Description of Nature. Cambridge University Press, Cambridge
(1934)
14. Bohr, N.: Atomic Physics and Human Knowledge. Wiley, New York (1958)
15. Bohr, N.: Essays 19581962 on Atomic Physics and Human Knowledge. Wiley, New York (1963)
16. Gleason, A.M.: Measures on the closed subspaces of a Hilbert space. J. Math. Mech. 6, 885893
(1957)
17. Bell, J.S.: On the Einstein-Podolsky-Rosen paradox. Physics 1 (1964)
18. Bell, J.S.: Speakable and Unspeakable in Quantum Mechanics. Revised edn. Cambridge University
Press, Cambridge (2004)
19. Kochen, S., Specker, E.: The problem of hidden variables in quantum mechanics. J. Math. Mech. 17,
5981 (1967)
20. Mermin, D.: Hidden Variables and the Two Theorems of John Bell. Rev. Mod. Phys. 65, 803815
(1993)
21. Bell, J.S.: Against measurement. Phys. World 8, 3340 (1990)
22. Fuchs, C.A., Schack, R.: Bayesian conditioning, the reection principle, and quantum decoherence.
arXiv:1103.5950v1 [quant-ph] 30 Mar. (2011)
23. Fuchs, C.A.: Quantum mechanics as quantum information, mostly. J. Mod. Opt. 50(6/7), 9871023
(2003)
24. Fuchs, C.A.: QBism, the perimeter of quantum Bayesianism. arXiv:1003.5209v1 (2010)
25. Brun, T., Finkelstein, J., Mermin, N.D.: How much state assignments can differ. Phys. Rev. A 65,
032315 (2002)
26. Zurek, W.H.: Environment-induced superselection rules. Phys. Rev. D 26, 18621880 (1982)
27. Cucchetti, F.M., Paz, J.P., Zurek, W.H.: Decoherence from spin environments. Phys. Rev. A 72,
052113 (2005)
28. Sellars, W.: Inference and meaning. Mind 62, 313338 (1953)
29. Einstein, A., Podolsky, B., Rosen, N.: Can quantum-mechanical description of physical reality be
considered complete? Phys. Rev. 47, 777780 (1935)
30. Juffman, T. et al.: Wave and particle in molecular interference lithography. Phys. Rev. Lett. 103,
263601 (2009)
31. Paz, J.P., Habib, S., Zurek, W.H.: Reduction of the wave packet: preferred observable and decoherence
time scale. Phys. Rev. D 47, 488501 (1993)
32. Schlosshauer, M.: Decoherence and the Quantum-to-Classical Transition. Springer, Berlin (2007)
33. Hackermller, L., Hornberger, K., Brezger, B., Zeilinger, A., Arndt, M.: Decoherence of matter waves
by thermal emission of radiation. Nature 427, 711714 (2004)
34. Hornberger, K., Sipe, J.E., Arndt, M.: Theory of decoherence in a matter wave Talbot-Lau interfer-
ometer. Phys. Rev. A 70, 053618 (2004)
35. Paz, J.P., Zurek, W.H.: Quantum limit of decoherence: Environment induced superselection of energy
eigenstates. Phys. Rev. Lett. 82, 51815185 (1999)