<<

. 21
( 62 .)



>>

idea that an individual thing™s intrinsic properties determine its place in
the causal nexus. The relational properties of that individual thing are
determined not simply by its intrinsic properties, but by those together
Context-Sensitive Realizations 123

with the intrinsic properties of the object(s) to which it is related. For
the particularist, it is still the intrinsic properties of individual things that
realize all the properties that there are. There are several problems that
particularism faces.1
First, it suggests a general strategy for understanding relational prop-
erties that presupposes that they can always be viewed in terms of rela-
tions to particular objects. Certainly, many relational properties in sci-
ence can be so understood. But many of the examples I provided in
the last chapter invoke relational properties that lack this feature, and
this points to the dif¬culty of assimilating all relational properties and
kinds to particularism. For example, thinking about water, having a spe-
ci¬c level of evolutionary ¬tness, and being a homeowner are relational
properties, but not ones that simply involve relations between the bearer
of the property and one or more particular objects. Rather, they are
relational in presupposing the location of the bearer of the property
in some larger system or environment: in a certain linguistic commu-
nity, in an (usually, the actual) environment, in a society with certain
kinds of institutions and practices. An individual™s intrinsic properties
do not determine whether it has any of these relational properties. Sim-
ply adding to these the intrinsic properties of other individuals (and
which individuals?) does not lead to a realization base for these relational
properties.
Second, since particularism is a form of microphysical determinism,
then not any old intrinsic property serves as the ultimate realizer for
all the properties that there are, but just those intrinsic properties that
very small things have. Thus, as well as having to show how all relational
properties are determined as are quantitative comparative properties,
particularism must also show how all of the intrinsic properties of the not
so small are realized by just the intrinsic properties of the smaller things
that constitute them. But this seems false even at the chemical level,
let alone at “higher” levels. For example, salt is physically composed of
sodium and chlorine. Yet its having the chemical composition that it does,
and the intrinsic properties it has, is not realized simply by the intrinsic
properties of sodium and the intrinsic properties of chlorine, but those
together with the realizing sodium and chlorine standing in a certain
relation to one another (in particular, having a certain spatial relation to
one another). Likewise, organism-level properties that we might think of
as intrinsic, such as running speed or bodily musculature, are not simply
realized by the intrinsic properties of organism parts, but by those plus
how those parts are put together.
Individualism and Externalism in Cognitive Sciences
124

This second problem suggests a natural revision to particularism.
Rather than try, as does particularism, to maintain that there is a sense
in which the relational properties of individual things can be reduced
to intrinsic properties, instead consider relations themselves as ontolog-
ically basic. Call such a view relational microphysicalism. This form of mi-
crophysicalism plays up the fact that when we turn to physics itself we
¬nd a range of relations: from states of entanglement in quantum me-
chanics, to space-time separation in relativity theory, to velocity in good
old Newtonian mechanics. Since such relations occur within physics “
indeed, within microphysics “ the idea is to take them, together with the
intrinsic properties of the physically very small, as what realizes all other
properties and relations. Relational microphysicalism thus abandons the
putative asymmetry between intrinsic and relational properties, but as a
form of smallism maintains the asymmetry between the small and the not
so small.
David Lewis™s endorsement of the thesis of Humean supervenience
expresses a form of relational microphysicalism. In the preface to the
second volume of his collected papers, he expresses this doctrine as
holding that
all there is to the world is a vast mosaic of local matters of particular fact, just one
little thing and then another . . . we have local qualities: perfectly natural intrinsic
properties which need nothing bigger than a point at which to be instantiated.
For short: we have an arrangement of qualities. And that is all.2

Lewis™s references to mosaics and arrangements suggest relational micro-
physicalism rather than particularism.
Relational microphysicalism avoids the second problem facing partic-
ularism, but still faces the ¬rst problem, that of accounting for relational
properties that do not involve relations between objectlike relata. In ad-
dition, while it countenances relations, it is only those relations that, in
the ¬rst instance, hold between the smallest objects there are. Relational
microphysicalism is aggregative about what there is in the world in that
any composite entity, C, and its properties, are realized by its physical con-
stituents, A and B, their properties, together with the relations between
A and B. Yet C™s relational properties will not be realized by A, B, and the
relations between them. So quite apart from relational properties that do
not involve objectlike relata, and even granting that relational microphys-
icalism is true of an object™s intrinsic properties, provided that aggregated
individuals have some relational properties, relational microphysicalism
remains unable to account for all there is.
Context-Sensitive Realizations 125

3 dispositions and science
Many properties in the physical sciences are dispositional: They are ten-
dencies that the objects that have them manifest in certain circumstances
or under certain conditions. While the manifestation of dispositional
properties may require those circumstances or conditions to obtain, what
is often called the categorical base of the disposition surely does not: The
base is intrinsic to the bearer of the disposition. Given this popular view
of dispositions, one might develop several lines of argument that indicate
that there is something mistaken about the context-sensitive view of re-
alization. For example: Because dispositions are intrinsic and important
in the physical sciences, in traf¬cking in unanalyzed relational proper-
ties the fragile sciences lose their mooring from physicalism. Or: Because
we can understand the role of “context” in making sense of dispositions
without abandoning the idea that intrinsic properties are ontologically
basic in some sense, we can preserve the asymmetry between intrinsic
and relational properties, or between the small and the not so small,
that I have used the context-sensitive view of realization to challenge. Or:
Even if there is a sense in which some properties have a wide realiza-
tion (manifestation), there is a deeper sense in which they must have an
entity-bounded realization (base).3
These lines of argument could, no doubt, be ¬lled in. But my aim
here, in part, is to suggest that to do so will be in vain, for the view of
dispositions on which they rest is problematic for much the reason that
the standard view of realization is problematic. Many of the dispositional
properties in the physical sciences are themselves relational.
Consider dispositional properties in chemistry, such as acidity and mis-
cibility. Whether a given liquid is acidic or miscible, whether it has those
properties, is metaphysically determined by more than facts about its con-
stitution. This is because these dispositional properties are dispositions
to have certain effects on other chemical substances and kinds, and so
whether they have those dispositions “ not just whether they manifest
these dispositions “ is determined by facts about those other substances.
The very presence of the disposition, not just its manifestation, involves
the physical con¬guration of the world beyond the bearer of the disposi-
tion. These dispositional properties have a wide physical realization.
This point is re¬‚ected in standard de¬nitions of an acid “ for exam-
ple, as a proton donor, or as an electron-pair acceptor. For whether a
substance with a given physical structure has the disposition to donate
protons or to accept electron pairs depends on facts about the broader
Individualism and Externalism in Cognitive Sciences
126

chemical system in which that substance exists. If just these facts were
different, a liquid that is actually an acid could lose this disposition, and
could do so even were its chemical composition to remain unchanged.
While successive concepts of an acid “ from that of Arrhenius, to those of
Bronsted and Lowry and Lewis “ characterize an acid in what we might
think of as increasingly “purely” dispositional terms, that is, in ways that
increasingly abstract away from the environment, all of these concepts
make either explicit or implicit reference to the character of solvents
and bases which lie beyond the boundary of the thing that is an acid.
Whether, say, hydrogen chloride has the disposition to donate a proton
when placed in water depends not only on the intrinsic chemical char-
acter of HCl, but also on that of H2 O and the relative strengths of the
forces governing their interaction. A physical duplicate of a substance
that is acidic in the actual world may not be acidic in a world where the
facts about solvents and chemical forces are different, even once we hold
¬xed the facts about other physical forces as “background conditions.”
In this respect, acidity is like ¬tness or any of the other examples I have
cited from the fragile sciences. The salience of the fact that the core re-
alization of acidity is individualistic should not obscure our view of what
metaphysically determines the presence of the disposition itself.4
This requires endorsing that at least some dispositions, including some
dispositions in the physical sciences, are extrinsic or wide: Whether some-
thing has the disposition is not determined by that thing™s physical con-
stitution. Jennifer McKitrick has recently provided a range of common-
sense examples of extrinsic dispositions “ the power to open a door,
weight, the disposition to dissolve the contents of my pocket, vulnera-
bility, visibility, and recognizability “ and defended this idea against the
predominant view of dispositions as essentially intrinsic. Some of the
examples that I provided in Chapter 5, such as ¬tness and folk psycho-
logical states, are often understood as dispositional properties within the
fragile sciences, and if they are relational, as I have argued, then they
must be wide dispositional properties. In fact, such wide dispositions are
widespread across the sciences, including acidity, miscibility, and solu-
bility in the chemical sciences; conductivity, heat sensitivity, and rigidity
in the electronic and engineering sciences; and trustworthiness, fertility,
and stability in the cognitive, biological, and social sciences.5
Part of my point here is to preempt a series of connected ideas con-
cerning dispositional properties and the nature of science: that an ap-
peal to dispositional properties provides refuge for the standard view
of realization and the asymmetry that it posits between intrinsic and
Context-Sensitive Realizations 127

table 6.1 Realization in Psychology, Biology, and Chemistry

Psychological Biological Chemical
being in pain being a predator being acidic
P
nociceptive system predator-prey system system of acids and
S
bases
person animal liquid solution
IB
certain level of having sharp claws having an empty
core R
activity in C-¬bers electron shell
CR plus states of rest CR plus properties of CR plus properties
total R
of nociceptive rest of IB and prey of rest of IB plus
system (thalamus, that, together with those of solvents
somatosensory IB, constitutes the and bases that,
cortex, etc.) predator-prey together with IB,
system (e.g., speed constitute the
of IB, body of prey) acid-base system
conditions of rest of conditions of the conditions of the
BCs
(holding the nervous ecological physical world
of world system of IB and environment more generally,
beyond S) IB™s other bodily occupied by S, e.g., e.g., particular
systems terrain, seasonal value of gravity
conditions on Earth, laws of
nature



relational properties; that “real dispositions” in the fundamental sciences
are intrinsic; that the ubiquity of relational properties in the fragile sci-
ences casts them under some sort of metaphysical shadow dissipated by
the light of the physical sciences. But part of my point is to extend the
reach of the context-sensitive view of realization and the framework that
I introduced in articulating it. To this end, Table 6.1 shows the parallels
between psychological, biological, and chemical examples that I have dis-
cussed within this framework. As this table shows, both entity-bounded
and wide realizations can be readily represented within this framework.
The symmetry between the two, I want to suggest, implies the irrelevance
of the boundary of the individual as a constraint on what counts as a
realization of a given property.
Relational properties permeate the fragile sciences, and we should
thus expect wide realization and the strategy of integrative synthesis to
be the norm across those sciences. While entity-bounded realization and
constitutive decomposition are certainly predominant in the physical
sciences, there is also a range of properties in these sciences that are,
like those in the fragile sciences, implicitly relational, including many
Individualism and Externalism in Cognitive Sciences
128

dispositional properties. Such properties and kinds have wide realiza-
tions. This is one more reason to be skeptical of the idea that there is any-
thing metaphysically distinctive or suspicious about the fragile sciences,
or about the nature of relational properties and kinds.
Having examined aspects of physicalism in general in light of the
context-sensitive view of realization, let us return to focus on physical-
ism about the mind in particular. Rejecting the constitutivity thesis in
the standard view most obviously requires giving up reductive physicalist
views of the mind. But this also has perhaps less obvious implications for
nonreductive materialism.


4 nonreductive materialism
The con¬‚ict between the suf¬ciency and constitutivity theses provides a
novel way of expressing a long-acknowledged tension between external-
ism and reductive forms of physicalism in the philosophy of mind. Yet it
also points to largely unrecognized inadequacies in a number of ways of
expressing nonreductive materialism. Those expressions have also relied
on the standard view of realization.
Nonreductive materialism has sometimes been formulated in terms of
the acceptance of a “token-token” identity thesis or via a compositional
view of realization. Consider each view in turn and what our discussion
thus far implies about it.6
The token identity theory claims that tokens of mental and physical
states may be identical even if types of mental and physical states are
not identical, where the relevant physical states are intrinsic states of the
brain. Since such states are intrinsic, physical states of the individual,
the token identity theory, at least as usually articulated, presupposes the
constitutivity thesis. But our exploration of the varieties of realization
suggests that this thesis and the token identity theory are false for at least
many mental states. This is because the total realizations of a range of
mental states are wide and thus are not intrinsic states of the brain. At
most, it is the core realizations of mental and physical states that are
identical, but this does not help us identify mental and physical states.
Compositional views of realization, and thus physicalism, likewise take
the relevant composed entity to be the individual or her central nervous
system, and in so doing, rely on the constitutivity thesis. They thus face
the same problem that the token identity theory faces. In short, both of
these common expressions of nonreductive materialism have relied on
the standard view of realization. If either view allows the relevant tokens
Context-Sensitive Realizations 129

or composed entity to be larger than the individual who instantiates the

<<

. 21
( 62 .)



>>