<<

. 20
( 62 .)



>>

different environments. That is, the numbers that represent viability and
fertility may vary solely because of an organism™s environmental location.
This relational aspect to the property of ¬tness is often masked by the
fact that an organism™s environment usually plays (in effect) the role of a
constant in many of the contexts in which the concept of ¬tness is put to
work. Yet the properties “has a probability of surviving of 0.7” and “has
an expected number of offspring of 2.2” are incompletely speci¬ed in
a way that makes them meaningless without an implicit reference to an
environment. What metaphysically suf¬ces for a given organism to have
a speci¬c level of ¬tness is not instantiated entirely in that organism: The
total realization of ¬tness (and its determinate forms) is wide, not entity
bounded. Here the relevant wide system is the organism plus something
like its niche, broadly construed to include its location within a particular
population structure and other relational aspects of its existence.
The same general point is true of the total realizations of properties of
phenotypic traits or behaviors. What makes a trait or behavior an adap-
tation, for example, is something about its evolutionary history, not just
something about the individual who has the trait or exhibits the behavior.
Whatever we want to say about their core realizations, their total realiza-
tions are not individualistic. There are ontological and not just epistemic
Individualism and Externalism in Cognitive Sciences
116

reasons why you can™t tell whether, say, the presence of wings in a given
species is an adaptation or a spandrel simply by inspecting existing mem-
bers of that species. These examples show that total realizations do not
always satisfy the constitutivity thesis; they also exemplify the pattern of
integrative synthesis depicted in Figure 5.1b.
To this point I have been discussing wide realizations whose core part
is located within an individual. One might well wonder whether a re-
alization could be wide in that not only does its noncore part extend
beyond the individual but so too does its core part. I shall call this type of
wide realization a radically wide realization:
(e) radically wide realization: a wide realization whose core part is not
located entirely within IB, the individual who has P
The clearest examples of radically wide realizations are those of social
actions that involve engaging with the world and have further social and
institutional background conditions. And so we move from the biological
to the social sciences in thinking about the full range of the applicability
of the notion of wide realization in the fragile sciences.
Consider actions such as making a withdrawal from a bank, committing
a felony, or voting, each of which we might do by signing a piece of paper
in certain circumstances. Here, not only the noncore part of the total
realization extends beyond the individual agent, but so too does the most
natural candidate for the core realization of these actions, signing a piece
of paper. This is an action of an individual, as are the actions that it in
turn realizes, even though the action extends beyond the boundary of
that individual. The relevant system explored via integrative synthesis “
whether it be the banking system, criminal justice system, or electoral
system “ likewise extends beyond the boundary of the individual agent
and has its own background conditions.
The same is true of many of our ways of classifying agents in the so-
cial sciences. In economics, for example, individuals are homeowners,
consumers, and wage earners. Someone falls into one of these categories
by virtue of the relationship between what she does and has done, and
the legal, social, and economic institutions and practices to which she is
subject. An agent enters into verbal and written agreements, hands over
cash, or performs certain tasks. Since these actions or behaviors them-
selves, however, literally extend into the world beyond the individual,
their realizations do not stop at the skin. Behavior that stops at the skin “
the bodily movement that an agent makes “ is not even a core realization
of the corresponding action.
Metaphysics, Mind, and Science: Two Views 117

10 two views reconsidered
This chapter has been more analytical in its style than those that have pre-
ceded it, and it may pay to step back from the thick of it all to see where
we are. In sections 7“9, I have proposed four ways in which physical re-
alizations are context sensitive, the ¬rst and second of which challenge
the suf¬ciency thesis (section 7), and the third and fourth of which chal-
lenge the constitutivity thesis (sections 8 and 9). In section 8, I introduced
(ordinary) wide realizations, exempli¬ed by intentional mental proper-
ties, whose noncore part extends beyond the boundary of the individual
who has those properties. What I called radically wide realizations in sec-
tion 9 are wide realizations whose core part also extends beyond that
boundary, and I suggested that they were exempli¬ed by social actions.
My more general point in these sections was that there are a range of
properties in the cognitive, biological, and social sciences that have wide
rather than entity-bounded realizations.
Those who would like to salvage the standard view of realization can
shuf¬‚e where they locate the particular examples I have introduced in this
four-fold schema. But since the suf¬ciency and constitutivity theses are
jointly satis¬ed in none of the four forms of context-sensitive realization,
there will remain a problem for the standard view, no matter how much
shuf¬‚ing is done.
On the standard view, realizations are metaphysically determinative of
the properties they realize and physically constitutive of the individuals
who possess those properties. On the alternative, context-sensitive view,
realizations are located within systems, and those systems in turn are lo-
cated within broader environments. The context-sensitive view allows that
the standard view is correct about some properties and their realizations “
what I called entity-bounded realizations in section 8 “ but insists that
there is a wide variety of properties that involve another form of context
sensitivity, that of wide realization. In effect, the context-sensitive view
implies that there is an individualistic bias within the standard view, one
that not only ignores the perhaps innocuous context sensitivity of both
core and total realizations, but that overlooks or denies the possibility of
wide realizations altogether.
I have been arguing that there is a sort of tension within the standard
view. For what metaphysically determines the presence of a property of-
ten extends beyond the boundary of the individual who has that prop-
erty. And, conversely, the physical constituents of the individual bearing
a given property are often contributory or partial realizers rather than
Individualism and Externalism in Cognitive Sciences
118

metaphysically determinative realizers of that property. In advocating the
endorsement of wide realizations, I have, in effect, embraced the suf¬-
ciency thesis over the constitutivity thesis.
With this in mind, let us return to reconsider the relationship between
realization and appeals to neural correlates, mechanisms, and localiza-
tions within the cognitive sciences. Just as realizations are assumed to be
entity bounded on the standard view, psychologists and cognitive neuro-
scientists typically think of neural mechanisms as entity bounded, and so
have an individualistic view of these. Given the tension within the stan-
dard view of realization, it seems clear that cognitive scientists themselves
would likely opt for the constitutivity over the suf¬ciency thesis. If that is
so, however, then this represents a signi¬cant break from the physicalist
orthodoxy that “the physical” determines “the mental.”
Whatever psychologists and cognitive neuroscientists think about re-
jecting the metaphysical suf¬ciency thesis, it is a view deeply embedded
within the philosophy of mind. First, the view of realization as a determi-
native relation is one that reductionists such as Jaegwon Kim have readily
drawn on in arguing that nonreductionists occupy an unstable position;
this view is not likely to be seen by such reductionists as merely optional.
Second, while this view of realization can be seen as just one of a cluster of
intuitions, it is central to understanding the cluster as a whole. For exam-
ple, it connects the notion of realization with that of supervenience in an
intuitive way by treating them as correlative notions; it explains the sense
in which realizers provide a metaphysical basis for what they realize; and
it allows us to make sense of the idea that mental properties are “nothing
over and above” their realizations. This should make us wary of simply
dropping the suf¬ciency thesis. To settle, in effect, for partial realizations,
such as core realizations, as the metaphysical substrate to the mind that
best corresponds to what cognitive scientists mean in talking of the neural
correlates, mechanisms, and localizations for cognitive capacities would
require a softening in how physicalism is to be understood.
The context-sensitive view suggests another option, one that both pre-
serves this determination thesis in some cases and brackets consideration
of it in others. While some cognitive capacities and the mechanisms that
realize them have an entity-bounded realization, others merely have neu-
ral correlates, or are partially realized in speci¬c regions of the brain. The
context-sensitive view of realization goes hand in hand with a context-
sensitive view of mechanisms. Neural mechanisms, like neural realiza-
tions, are always located within cognitive systems, and I have argued
that cognitive systems sometimes extend beyond the boundary of the
Metaphysics, Mind, and Science: Two Views 119

individual property bearer. When they do, the mechanisms are wide in
that they are individuated by reference to this system.
By identifying a range of properties across the cognitive, biological,
and social sciences to which the context-sensitive view seems to naturally
apply, I have made a start on articulating that view and perhaps shown its
promise. I have thus far, however, done little by way of really defending
the context-sensitive view of realization. That is the task of Chapter 6.
6

Context-Sensitive Realizations




1 adjusting one™s metaphysics
The argument of Chapter 5 provides a prima facie motivation for further
exploring the context-sensitive view of realization. In this chapter, these
explorations are of three kinds. Together they constitute a philosophical
defense of the context-sensitive view.
The ¬rst concerns context-sensitive realizations and physicalism. Phys-
icalism has been formulated both as a view of the mind-body relation
and as a more encompassing metaphysical position, and one of the aims
of the next three sections is to show what the context-sensitive view of
realization implies about both of these forms of physicalism. Although
there are forms of physicalism in the philosophy of mind that allow for
the possibility that minds have nonphysical realizations, it remains true
that the expression “physical realization” would be regarded by many as
containing a redundancy. There are, so far as I know, no developed pro-
posals for just how the mind would be “realized in” nonphysical stuff.
Given this, it is important to show that the context-sensitive view of real-
ization preserves robust forms of physicalism, even if, as I shall argue, it
requires either giving up or revising several strands to physicalist thought.
In this regard, in sections 2“4, I shall discuss, respectively, the thesis of
microphysical determinism, the nature of dispositions, and nonreductive
materialism.
The second kind of exploration considers modi¬cations to the stan-
dard view of realization that address its putative shortcomings as a general
view of realization. One general suspicion often directed at so-called novel
views is that their putative insights can be captured within the framework

120
Context-Sensitive Realizations 121

of existing views. Being clear about the overall metaphysical package that
the context-sensitive view brings with it is one way to dissipate such suspi-
cions, or at least to shift the burden of proof to those who harbor them.
In section 5, I shall examine a version of the standard view that ascribes a
key role to background conditions without buying into the full range of
context-sensitive realizations advocated within the context-sensitive view.
Here it will be useful to introduce an analogy to the place of background
conditions in discussions of causation, both in articulating the modi¬ed
standard view and in revealing its shortcomings relative to the context-
sensitive view. In section 6, I examine several other attempts to encom-
pass some form of context sensitivity within the framework provided by
the standard view. If the argument of these sections is correct, then the
context-sensitive view offers a richer metaphysical package than do these
variations on the standard view.
Finally, I shall conclude this exploration of the context-sensitive view
by tacking to the other side, responding to claims that taking the context-
sensitive view seriously requires adjustments in our conception of minds,
individuals, and the relationship between them that are considerably
more radical than those I have entertained thus far. Those who both
are sympathetic to the context-sensitive view and think that it marks an
important departure from business as usual in a materialist metaphysics
of mind might well think that my own remarks in this chapter and the
last are unduly conservative or restrictive. If mental states have wide re-
alizations, for example, why think that we can (or would want to) hold
onto realism about the mind, or restrict ourselves to viewing subjectivity
as attaching to individuals at all? I shall take up these sorts of question in
sections 7“9.

2 microphysical determinism, relations, and smallism
In the previous chapter, I said that the fragile sciences often traf¬c in
properties that have a wide realization, and that when they do the cor-
responding methodology of integrative synthesis is appropriate. These
sciences often investigate the relational properties of individuals, and
what (totally) realizes such properties is not contained within the bound-
ary of those individuals. Thus, an adequate account of the metaphysics of
realization must move beyond the standard view with its exclusive focus
on entity-bounded realizations and the constitutivity thesis.
We could express this criticism of the standard view by saying that
it manifests a form of smallism, discrimination in favor of the small, in
Individualism and Externalism in Cognitive Sciences
122

directing us always to what constitutes individuals, rather than what they in
turn constitute. Indeed, the problem that relational properties and kinds
pose for the standard view of realization is a version of what in Chapter 1
I identi¬ed as a general problem for smallist views in metaphysics. Since
the standard view has been developed as part of a physicalist view of the
mind, and the problem that I claim exists is quite general, we should
consider the resources that physicalists have to address it.
A common expression of physicalism says that all facts, properties,
events, processes, states, and entities supervene on the totality of physical
facts, properties, and so on. This is to say that there is some or other
physical base set of facts, properties, and so on for any fact, property,
and so on. It is a short step from here to gloss physicalism as the view that
there are physical realizations for all of the facts, properties, and so on that
there are. Given that both supervenience and realization are relations of
metaphysical determination that, so to speak, run in opposite directions “
from realizations and to supervening properties “ we might be tempted
to think of these two metaphysical notions as getting at much the same
idea. If that is so, then our short step is very short indeed, perhaps just
that of logical closure.
Physicalism, thus, is a kind of thesis of metaphysical determination:
The physical determines the rest. Microphysical determinism is a more
speci¬c form of physical determinism, one more directly relevant to as-
sessing smallist metaphysical views. It is the general view that the proper-
ties of microphysical entities metaphysically determine all the properties
there are. Consider two forms of microphysical determinism that, in dif-
ferent ways, attempt to accommodate relational properties.
The ¬rst of these Paul Teller has called local physicalism and particular-
ism. It holds that the microphysical properties of any given entity, plus
the microphysical properties of all those entities to which it is related,
determine all of the properties it has, including its relational properties.
Suppose that Tom has a mass of 100 kg and Susan has a mass of 80 kg,
and so Tom has the relational property of being heavier than Susan,
and stands in the relation “is heavier than” to Susan. On the particu-
larist version of microphysical determinism, this relational property of
Tom™s, and the corresponding relation between Tom and Susan is de-
termined by Tom™s microphysical properties and Susan™s microphysical
properties. What particularism preserves from corpuscularianism is the

<<

. 20
( 62 .)



>>