<<

. 12
( 62 .)



>>



those that develop in the normal course of an organism™s life. In what
follows, I shall take for granted the criticisms that Khalidi and Samuels
make of such views.18
Khalidi has defended the adequacy of a triggering view of innateness
over what he calls the disease model and plasticity accounts of nativist
views and argued that domain-speci¬city stands in an evidential rather
than a constitutive relationship to nativism. In essence, Khalidi™s account
makes do with just one of the two dimensions that I have posited to
the debate over nativism, that encapsulated by the external minimalism
thesis. Khalidi begins by registering the variety of cognitive scientists “
including Hirschfeld, Gelman, Landau, Cosmides, Tooby, Keil, Gopnik,
and Meltzoff “ whom he considers as viewing innateness and domain-
speci¬city as constitutively linked. He then states his view that an “innate
mental state or cognitive capacity is one that would be triggered by the en-
vironment,” a view articulated earlier by the philosopher Stephen Stich.
The remainder of Khalidi™s paper criticizes views that attempt to show
Disciplining the Individual and the Mind
62

that there is some constitutive link between this notion of innateness and
domain-speci¬city.19
There are two methodological problems with this view and how it is
defended. The ¬rst is that it is rather philosophically high-handed in
implying that many of those at the center of contemporary debates over
nativism are conceptually confused about what nativism is. No doubt,
there is some confusion. But defending a view of nativism that implies
that a wide variety of researchers working on nativism are infected with
this particular confusion is like insisting that lawyers are conceptually
confused when they treat spoken agreements as contracts on the grounds
that one has analyzed the concept of a contract in terms of signed, written
agreements. The implication itself should be grounds for rethinking the
grounds of the analysis.
The second problem is that the arguments that Khalidi provides can at
best show that internal richness is neither identical to, nor reducible to,
nor entailed by something like the external minimalism thesis. This point
should be granted by a proponent of the two-dimensional view. Indeed,
one might take this, in conjunction with how researchers themselves talk
about nativism, as a reason for insisting that we need at least two inde-
pendent dimensions to understand the debate over nativism. Moreover,
since the arguments against incorporating something like the internal
richness thesis into the account of nativism are all arguments that pre-
suppose the priority of a one-dimensional triggering criterion, they beg
the question against a proponent of the two-dimensional view.
Richard Samuels provides another one-dimensional analysis of innate
cognitive structures, the primitivist account, that proceeds in two steps. Ac-
cording to Samuels, “a psychological structure is innate just in case it is a
psychological primitive,” and (step two) psychological primitives are struc-
tures posited by some correct scienti¬c psychological theory, structures
whose acquisition is not in turn explained by any such theory. (Putative
counterexamples to the suf¬ciency of the account lead Samuels to mod-
ify it by making an additional appeal to the acquisition of the structure
“in the normal course of events”; I ignore this here since nothing I say
turns on this modi¬cation.)20
There are three related problems with this view: it (a) places a bur-
den on the notion of a complete psychological theory heavier than that
notion can bear; (b) presupposes a tidy alignment between psychologi-
cal structures and the mechanisms that mediate their acquisition, and
(c) entails that psychologically complex structures cannot be innate.
What these problems imply is that the primitivist account is uninformative
Nativism on My Mind 63

[(a)] or mistaken [(b) and (c)] about what psychological structures are
innate. I spell out each problem in turn.
Consider (a). Whether a structure is innate, on Samuels™s of¬cial view,
turns on there not being any correct psychological theory that explains
its acquisition. Given that even by the most optimistic of lights there
remain signi¬cant and fundamental disagreements about the acquisition
of almost every postulated cognitive structure, one might wonder just
what structures a correct psychological theory will postulate as acquired.
Even the current best candidates for innate psychological structures, such
as those governing intuitive mechanics and intuitive mathematics, have
previously been viewed within developmental psychology as having been
acquired. Without a more complete psychological theory it is dif¬cult to
see what their status will be when psychologists call it a day.21
Suppose that the sorts of inferences that we can make from the current
state of psychological theorizing to a complete and correct psychologi-
cal theory are limited, as I am implying. Then the primitivist account is
about as useful as an analysis that claims that innate cognitive structures
are those that God planted in our minds. Either (even both) of these
analyses might be true, but it seems desirable for a correct analysis to
be more informative than is either. To draw a contrast, it may be that
the fundamental physical entities in the universe are those posited in a
correct physical theory. But this constitutes a part of an insightful char-
acterization of (say) physicalism only given that current physical theory
provides us with some solid epistemic guidance as to what a correct physi-
cal theory looks like, and what it postulates as fundamental. I suggest that
this is precisely what we lack at the moment in the cognitive sciences.
Let us put the of¬cial talk of “correct psychological theories” to one
side and turn to the heart of the primitivist view, which holds that “innate
cognitive structures are ones that are not acquired by any psychological
process or mechanism.”22 Since there is an obvious sense in which all
of an organism™s structures are acquired “ all organisms develop (say,
from a fertilized egg) from not having to having them “ this account
implies that innate structures are those acquired by nonpsychological
mechanisms. This brings me to (b): that the account presupposes a
neat alignment between the nativist status of psychological structures
and the kinds of mechanisms that mediate their acquisition. Those ac-
quired by nonpsychological mechanisms are innate, while those acquired
by psychological mechanisms are not innate. This appears to make good
sense of cases where the mechanisms responsible for some psychological
structure are paradigmatically psychological (for example, learning) or
Disciplining the Individual and the Mind
64

paradigmatically nonpsychological (for example, genetic). But I think the
appearances are misleading, and very few, if any, structures are acquired
exclusively either by psychological or by nonpsychological mechanisms.
Consider ¬rst the sorts of case where the primitivist account might
seem adequate. For example, the psychological structures needed to read
or to consciously solve algebra problems are not innate because they
are acquired via the psychological mechanisms or processes of teaching,
learning, study, and instruction. And the psychological structures needed
for identifying something as a physical object are innate because they are
acquired simply by “triggering,” that is, by brute-causal exposure. Re¬‚ec-
tion on both of these cases, however, brings out why (b) is a problem.
Reading and consciously solving algebra problems are abilities ac-
quired through psychological processes, no doubt. But their acquisition
usually also involves mechanisms that are “social,” such as instruction,
and others that are “biological,” such as those for neural storage. In-
deed, any psychological ability that is acquired once a child already has
acquired a lot of psychological structures and mechanisms, however they
are acquired, is likely itself to be acquired through the agency of both
psychological and nonpsychological mechanisms. One reason for the
importance of psychological mechanisms is the way in which they are in-
tegrated with both “higher” and “lower” level mechanisms, and thus the
correspondence proposed by Samuels between “noninnate” and “psycho-
logical” acquisition processes is problematic.
There is much the same problem for the proposed link between “in-
nate” and “nonpsychological” mechanisms of acquisition. Brute-causal
exposure to physical objects and how they behave mechanically may
be suf¬cient to activate the intuitive physics of normal newborn or very
young infants. But there are surely psychological mechanisms that such
infants are equipped with that play a role in how their intuitive physics op-
erates. For example, they have mechanisms that govern how they behave,
such as how long they look at a stimulus, or other mechanisms that govern
preferential looking or other forms of early behavior. That such behav-
ior is subject to psychological generalizations, such as “Find unexpected
stimuli interesting” and “Look longer at objects that are interesting,” is
presupposed in the very experimental paradigms for the investigation of
the proposed innate structures.
The basic problem that the entwinement of psychological and nonpsy-
chological mechanisms poses for the primitivist account is as follows.
If innate psychological structures are those acquired without any psy-
chological mechanisms or only by nonpsychological mechanisms, then
Nativism on My Mind 65

no or almost no psychological structures are innate. Importantly, the
psychological structures that are our current best bet for being innate
turn out not to be innate, on such a view. And that suggests that the
primitivist account is mistaken.
Finally, consider (c). By a “psychologically complex” structure I mean
a structure that has parts whose coordination and integration physically
constitute that structure, and which themselves are psychological. For
such a structure, there may be a correct psychological theory for how
that structure is acquired that is given in terms of how those parts develop
and are integrated. But for all that, the structure itself might be innate.
Indeed, those who view such structures as innate often take the prime
psychological task to be to functionally decompose the structure and un-
derstand the developmental trajectory of those parts. For example, Leslie
and Baron-Cohen posit an innate theory of mind module (ToMM), and
propose a particular compositional structure that is psychological and
whose development is explained in psychological and neuropsycholog-
ical terms. This is true more generally of domain-speci¬c theories (for
example, of physical objects, of biology), which are complex in something
like the sense above but whose innateness is not thereby ruled out. One
could insist that the psychologists who talk in this way are mistaken. But
such insistence would seem high-handed in much the way that I claimed
was Khalidi™s account.


6 satisfying some desiderata
Samuels also posits ¬ve constraints on the adequacy of any account of
innateness. These can be summarized group-wise as follows:
(i) X is innate ’ (X is not learned) and ’ (the environment plays
/
no role in the acquisition of X)
(ii) the account must make sense of the chief arguments for/against
nativism, of why nativism matters to cognitive science, and of the
logical geography to the debate.
Samuels calls (i) “conceptual constraints” and views them as more foun-
dational than (ii). All ¬ve desiderata that Samuels offers are plausible
constraints, and despite its other problems, the primitivist account prima
facie does a reasonable job of satisfying all of them.
But the two-dimensional view does a better job here. Since the external
minimalism thesis is one dimension in the view, that view readily meets
both of Samuels™ “conceptual constraints.” I have already directed much
Disciplining the Individual and the Mind
66

of my discussion at showing how the two-dimensional view fares regarding
the “logical geography” of the debate over nativism. The only point I shall
underscore here is that, in contrast to both one-dimensional views we have
considered in the previous section (as well as others we have not), the
two-dimensional view makes it natural to see why “being innate” is not
an all-or-nothing matter, and how we might distinguish between stronger
and more moderate positions in the debate over nativism.
The two remaining constraints that Samuels posits are that the ac-
count make sense of the arguments for nativism and that it show why the
debate over nativism is of signi¬cance within cognitive science. Samuels
himself considers two chief arguments for the innateness of cognitive
structures “ those which appeal to the poverty of the stimulus and to
early development “ and to those we might add arguments that cite the
universal manifestation of a psychological structure, or that invoke neu-
rally speci¬c de¬cits. Endorsement of the external minimalism thesis
itself would make each of these arguments plausible ones to offer, but it
is worth noting that the internal richness thesis also provides grounding
for at least these arguments. For if that thesis is true and cognition is
governed by internally rich units, then we might expect normal environ-
ments to be suf¬cient for the abilities that those units imbue an organism
with, and so develop early and universally, and for those abilities to be
subject to impairment through speci¬c neural damage. As importantly,
the two-dimensional view also helps to explain why rejecting one of these
arguments “ say, the poverty of the stimulus argument “ or viewing it as
having limited signi¬cance, need not involve a wholesale abandonment
of nativism or the arguments for it.
Finally on the desiderata front, the two-dimensional view suggests that
part of the signi¬cance of the debate over nativism lies in arguments over
two distinct issues: the causal role of the environment in the acquisition of
cognitive structures and abilities and the nature of the internal resources
that individuals antecedently bring to bear on cognitive tasks. Positions
on each of these issues carry with them methodological implications for
how to study cognition and particular cognitive abilities. One™s view of
the external minimalism thesis affects (or re¬‚ects) whether one thinks
that cognitive scientists should invest their time in studying the structure
of an organism™s environment. And one™s view of the internal richness
thesis does the same with respect to the sort of internal structures posited
or sought inside the cognizer. The two-dimensional view also makes it
clear that a position on one of these issues leaves room for disagreement
about the other.
Nativism on My Mind 67

7 but could two dimensions be enough?
Suppose that we need two dimensions to understand the debate over na-
tivism and that these are the two that I have suggested. Might we need
more? For example, one might think, particularly if one views evolution-
ary psychology as a strong nativist position, that one should also view the
claim that rich mental structures are evolved or encoded in the genome
as part of the commitment of a strong nativist view. The developmental
psychologist Frank Keil has suggested that it is an important part of na-
tivist views of the mind that what is “built in” to an organism is not simply
perceptual abilities but something intuitively further “upstream” from
perception, something “more central,” something genuinely cognitive.
The philosopher Fiona Cowie has also identi¬ed ¬ve theses as constitut-
ing the “Chomskyan nativist” view of language acquisition. On Cowie™s
view, a ¬rm commitment to one of these, the thesis that the constraints
and principles that de¬ne the domain of language and its subdomains be
identi¬ed with universal grammar is what distinguishes the Chomskyan
view from all nonnativist alternatives. Perhaps we should view this di-
mension as essential to characterizing nativism more generally. Since we
double the grain of our analysis for each additional dimension we add,
why not enrich our view of the nativism debate further? Why think two is
enough, especially if one leads to impoverishment? But I think we should
be cautious here. More is not always better. Consider brie¬‚y the proposals
of Keil and Cowie in turn.23
Keil™s suggestion is motivated by the recognition that all parties to the
debate over nativism acknowledge that some psychological structures are
innate and that what distinguishes nativists from empiricists, apart from
their views of what I am calling internal richness, is how far upstream this
structure is to be found. I think that there is some truth in this and that
this dimension of “centrality” is important in understanding why nativist
views come in degrees. In some speci¬c debates over nativism, such as that
over concepts, this dimension of centrality may be signi¬cant enough to
be added as a distinct dimension. So it may be that it is useful to add this
as a third dimension to the account I have proposed, at least for some
purposes.
Yet this dimension doesn™t help much in understanding other partic-
ular nativist debates, such as that over the theory of mind, which occur,
so to speak, at one place in the stream, and whose alternatives to strong

<<

. 12
( 62 .)



>>