<< . .

( 29)

. . >>


The sketch of the structure of substance concepts presented so far has
been argued for almost entrely a priori. I have attempted a task analysis
for substance concepts and tried to show what follows if they are to
perform these tasks. Earlier I compared this project to Marr™s first level
of analysis in his theory of vision, where he gives a task analysis for vi-
sual perception. To fill in higher levels of analysis, explaining exactly
how these abilities are implemented, how the various kinds of sub-
stances are reidentified across encounters, how skills in reidentification
are acquired, how substance templates are acquired and how they oper-
ate, is a job, as I understand it, primarily for experimental psychology
and for research in child development. But I can try to help make the
questions clearer and offer some tentative suggestions about where one
might look for answers.
According to various estimates, children acquire from five to nine
words daily between the ages of two and six (Byrnes & Gelman 1991;
Clark 1991; Waxman 1991) “ Chomsky says, “about a word an hour
from ages two to eight with lexical items typically acquired on a single
exposure . . . ” (Chomsky 1995, p. 15). How is this possible? One obvi-
ous hypothesis here is that many concepts are developed prior to lan-
guage, and indeed, at least some must be, for the infant recognizes her
mother and the dog recognizes its master. Each has the capacity to re-
identify the relevant individual under diverse conditions, thus making it
possible to learn how to behave appropriately in their presence.
Some of the skills that are surely essential to reidentifying ordinary
substances have traditionally been classified as “motor” and “perceptual”
rather than “cognitive.” Perhaps the most basic of these is the ability to
track objects with the eyes, head, feet, hands, ears, and nose, and so forth.
Objects tracked in this way are not merely conceived to be the same but
are perceived as the same under certain conditions, the perception of
sameness bridging, for example, over motions of perceived and perceiver,
over changes in properties of the object, and over temporary disappear-
ances of the object behind other objects. The mechanisms responsible
for the ability to track and for perceptual “identity-” or “existence-con-
stancy” may well be largely endogenous (Dodwell, Humphrey, & Muir
1987; Nelson & Horowitz 1987; Spelke 1993) and also “cognitively im-
penetrable” (Shepard 1976, 1983).That is, no matter what you know re-

ally happened, under appropriate sensory stimulation, certain illusions of
constant identity persist. Even if the perceived object apparently flies
right through a brick wall, you still can™t help perceiving it as the same
object going in one side and coming out of the other. These basic abil-
ities seem to be the bottom layer upon which conceptions of substances
are built.
The mechanisms by which infants reidentify individuals perceptually
apparently do not rely upon properties of the tracked object remaining
the same but upon movement, spatial location, and trajectory (Gopnik
& Meltzoff 1996). Xu and Carey (1996) have produced experimental
evidence that ten-month-old infants, unlike twelve-month-old infants,
are not surprised if an object of one kind apparently turns into an ob-
ject of another kind, say, a yellow rubber duck into a white Styrofoam
ball, though they are surprised if an object they are tracking apparently
turns into two objects.Tracking in this property-blind way would make
it possible to observe, for various broad kinds of objects, what sorts of
things tend to remain the same and what sorts may change within a
short period, yielding clues for keeping conceptual track of substances.
While perceptually tracking a substance you can learn how it looks,
how it sounds, how it feels, smells, tastes, the manner in which it moves
and changes, and so forth.
Perceptual tracking allows the accumulation of information about a
substance over a period of time, information perceived as about the
same substance. Nor is it only individual objects that are tracked in this
way. If I am tracking Fido, I am also tracking the species dog, and also
fur and bone. Which of these I am tracking with my mind depends
upon which I am learning about or registering information about as I
go. And that is determined by which of these substances I identify on
other occasions as the one this learning concerns, that is, as being the
same substance again. As I dissect my specimen frog in the zoology lab-
oratory, whether I am conceptually tracking just the individual Kermit, or
tracking frogs, depends on whether I attempt to apply what I have
learned from my experience only to later meetings with Kermit or
whether to frogs in general.


For the usefulness of your knowledge of a substance to last, however,
you must also know how to reidentify the substance after a break, even

a lengthy break, in perceptual tracking. And unless the substance is an
individual space-time worm, you must be able to reidentify it also over
its objective discontinuities in space and time. The substance dog is not
space-time continuous, nor is wood. This kind of keeping track of a
substance I will call “conceptual tracking.”To track a substance concep-
tually is to understand rather than directly to perceive its being the same
one when you encounter it again. Perceptual tracking would seem to be
the beginning of conceptual tracking, but conceptual tracking or keep-
ing track must continue over long and wide interruptions in perceptual
tracking. Out of what materials are our abilities to track substances con-
ceptually built?
The mechanisms of perceptual constancy for properties are probably
the most important. These mechanisms may be fashioned in part
through experience and certainly they are tuned through experience,
but much of their basic structure may be endogenous (Dodwell et al.
1987; compare Gallistel, Brown, Carey, Gelman, & Keil 1993; Marler
1993). They cause distal qualities to appear as the same through wide
variation in proximal manifestations. For example, they allow the same
shape and size to be registered as the same despite alterations in angle
of observation and distance, colors to appear as the same under widely
varying lighting conditions, and voices to sound as the same voice
through distortions and extraneous noise. These mechanisms allow one
to be sensitive to the objective variances versus invariances characteriz-
ing a perceptually tracked object through changes in conditions of ob-
servation and in its changing relations to the tracker. And they allow
substances to be reidentified via their stable properties under very di-
verse conditions of perception.
Because the mechanisms of perceptual constancy are involved, how-
ever, it should not be thought that concepts of properties are always in-
volved in conceptual tracking of substances. Having concepts of prop-
erties, I am assuming, would be to represent properties, as such, in
thought. The thought of a property is not just a reaction caused by a
property; it must play an appropriate conceptual role (Section 7.4). Cer-
tainly a mere response to a presented property, such as a discriminating
reflex response, requires no concepts. The moth turns toward light, but
has no concept of light. Similarly, responding, say, to a certain configu-
ration of shape, color, texture, and motion with the thought squirrel
again is not, merely as such, to have thoughts of these shapes, colors, tex-
tures, and motions themselves. Indeed, adults don™t seem to have con-
cepts of the particular shapes and motions that are squirrel shapes and

squirrel motions except the analytical concepts squirrel-shaped and moves-
like-a-squirrel, these concepts presupposing rather than underlying the
concept squirrel.
This accords, of course, with the finding that children appreciate
holistic similarities before appreciating separate property dimensions
such as color and shape, suggesting that concepts of properties and other
abstract objects may not be required to have substance concepts. Appar-
ently it also accords with findings in neuroscience:
. . . more detailed investigation reveals that most sensory neurons respond to
complex combinations of stimulus features. For example, visual cells that re-
spond to oriented edges may also respond to color, motion and color disparity
(Pribram 1991, pp. 79“81). Moreover, it is not uncommon to find neurons in
visual cortex that are attuned to acoustic frequencies (Pribram 1991, p. 81, cit-
ing Bridgeman 1982; Pribram, Speielli, & Kamback 1967). Conversely, it has
been reported recently (Calvert, Bullmore, Brammer, Campbell, Williams,
McGuire,Woodruff, Iverson, & Davis 1997) that our understanding of face-to-
face communication is aided by the response of auditory neurons to visual stim-
uli. Finally . . . top-down signals in sensory systems can alter the receptive
fields of sensory neurons, that is, their response is context-sensitive (Pribram
1991 pp. 257“8).
Much of the persistence of talk about feature detectors in neuroscience can
be attributed to the same descriptivist assumptions that pervade philosophy and
cognitive science. If . . . that is what we . . . look for in the brain . . . to a large
extent that is what we will find. (MacLennan 1998, p. 78)

MacLennan goes on to claim that cases in which “a stimulus is pro-
jected into a very low-dimensional space,” are “comparatively rare and
secondary to the processing of concrete micro correlations, upon which
reidentification rests.” Apparently, holistic neural representations are
prior to representations of single properties.
Besides perceptual tracking abilities and other perceptual constancies
that may be largely built in, there is evidence that infants may have built
into them systems designed, specifically, to recognize human faces. It is
well known that they have a strong disposition from the earliest days to
track and study human faces (e.g., Johnson, Dziuawiec, Ellis, & Morton
1991). Also, many species that recognize individual conspecifics instinc-
tively use smell for this purpose, and in the early months human infants
also know Mama by smell (MacFarlane 1977). It appears that the infant
may know innately at least two good ways conceptually to track indi-
vidual conspecifics. Faces and personal odors are indicative of individual
identity; clothes, postures, and so forth, are not.

An extension of perceptual tracking through space is a kind of con-
ceptual tracking through space. Even some quite lowly species are
equipped with the capacity to keep track of their positions as they move
about within their immediate spatial locales.This is the same, of course,
as keeping track of where other things are in relation to them.3 Where
something was when you encountered it last is often a clue to where it
would or might be when you encounter it again, thus serving as an aid
to identifying it.4 Things that don™t move at all can easily be kept track
of this way, and things that move slowly or only intermittently can be
kept track of this way over short interruptions in perceptual tracking.
This extends the period over which other identifying properties can be
observed or committed to memory. For example, at the beginning of
term I often have concepts of various students that I am not yet able to
recognize anywhere outside my classroom. The look of a new face or
new kind of animal may take a while to sink in, perceptual tracking and
conceptual tracking through space filling in temporarily.
Sometimes, on the other hand, we may find a use for merely tem-
porary concepts rooted in this sort of tracking. Consider the concept
you have of your glass at a cocktail party.You keep track of it by keep-
ing it in your hand, or by setting it down somewhere that you remem-
ber. But if you turn your back and someone straightens up a bit, that
may be the end of the tracking trail for that glass. When the party is
over, you lose track anyway, but it doesn™t matter. Similarly, concepts of
individual dishes in a matching set in one™s own cupboard are likely to
be only temporary. If these dishes have no individual salient distinguish-
ing marks, and you have no cause to remember special happenings con-
cerning any of them, every time the dishes are done and put away
again, all your individual concepts of cups, glasses, and plates disappear,
and new concepts of the same old individual dishes must be born again
next meal. As an experiment, try to think, right now, serially, of each in-
dividual fork in your silver drawer.


When perceptual tracking and conceptual tracking in space are coupled
with exploratory manipulation, probing, and testing, this may reveal
properties and dispositions that prove to be better tracers of a substance
3 Whitehead is supposed to have claimed that he always knew where he was, but that some-
times he didn™t know where the other things were.
4 Compare Gareth Evans (1982, Section 8.3).

than more easily observed properties. An easy example is the tool bag of
tests and routines that chemists use in order to identify chemical stuffs.
Tests of this latter sort are typically employed with an explicit under-
standing of the properties one is looking for. One has a disposition, for
example, to make an explicit inference from “the stuff has gone green”
to “there™s copper in it” (Quine 1960). Identifications of this sort gen-
erally do presuppose the application of prior property concepts. Fur-
ther, any explicit knowledge that you have of the properties of a sub-
stance can help you to identify it, even if these properties are not
unique to it. No, we think, that can™t be Sally after all because Sally
doesn™t know French, or that can™t be real gold in the window because
real gold would cost more than that.
It is because knowledge of the properties of substances are so often
used in the process of identifying them that it is easy to assimilate hav-
ing a concept of a substance to having knowledge of properties that
would identify it, and to assimilate identifying to classifying, to applying
a description. But consider: Recognizing Mama by smell certainly is not
classifying her nor is it conceiving of her as whatever bears that smell. It
is more accurate to imagine it as a tokening of the mental term “Mama”
in response to a smell. The thought is of Mama, not of smells, but it
arises in response to a smell. Similarly, recognizing copper by the fact
that the stuff has gone green is not conceiving of it as being, just, a
green-turning stuff. Rather, one tokens a mental term for copper in re-
sponse to the knowledge it has gone green.What makes it a mental term
for copper is not that it occurs in response to knowledge of these or
those properties, but the fact that it serves as a repository for incoming
information5 about copper and its tokenings are controlled by previous
manifestations of copper in one™s experience. These include, of course
manifestations by which explicit knowledge about copper has previously
been gained (for example, through language “ see Chapter 6).
Accurate understanding of the ontological principle or principles
that ground a substance can certainly help in tracking it in difficult
cases. The psychologists Medin, Gelman, Keil, and Gopnik and Melt-
zoff, especially, have been interested in tracing the origin and develop-
ment of children™s understanding of these principles, and they have ob-
served that both children and adults appreciate that there must be some
such principles underlying their substance concepts. But they have not

5 That is, information derived from natural informationC encountered by the senses. On in-
formationC, see Appendix B.

been clear that understanding of this sort is not necessary to having a
concept of a substance, and that having or lacking such understanding
need make no difference to the extensions of one™s substance concepts.


So much for learning to track substances. But how does the child know
what questions she should expect to be answerable about each sub-
stance? This requires at least a rough grasp of relevant substance tem-
plates. As mentioned in Chapter 2, there is evidence that some prelim-
inary grasp of some substance templates, such as physical kinds, animal
kinds, plant kinds, artifact kinds, social kinds and so forth, may be en-
dogenous (Atran 1989; Boyer 1998; Carey 1985; Gallistel et al. 1993;
Gelman and Coley 1991; Keil 1979, 1989; Markman 1989; Marler
1993; Spelke 1989, 1993). Mandler (1997, 1998) claims that “the earli-
est conceptual distinctions infants make is at the level of animal and ve-
hicle, not at the level of dog and cat” (1998, p. 79). If true, this is an in-
teresting contrast to the order in which they acquire the words “dog”
and “cat” versus the words “animal” and “vehicle.” This makes sense,
however, when we consider how few things there are to be learned
about either animal or vehicle (as such) on the one hand and how im-
portant these are as substance templates on the other. What is most in-
teresting about animals, for example, is that they divide into species, and
that roughly the same sorts of questions can be asked about each of
these species, and answered once and for all after one or a few observa-
tions. Since animal is not something there is much to find out about,
there also is not much to say about it. So it is not surprising that the
word “animal” enters the child™s vocabulary rather late. But, since rec-
ognizing the substance template animal is crucial to learning about the
various species of animals, it is equally unsurprising that animals might
be recognized as such very early. Indeed, as the various psychologists
mentioned above suggest, grasp of such substance templates may have a
strong boost from endogenous factors.
However acquired, an adult possesses innumerable substance tem-
plates of more and less generality. The ability to recognize substance in-
stances falling under these templates immediately supplies not only an-
swerable questions to ask of these substances but the ability to learn
how to track each new instance encountered very efficiently. Things
that are likely to vary in posture but not size or color can be tracked us-
ing size and color but not posture; things that can be more than one

place at once (kinds and stuffs) are not tracked by place, or rather, their
tracking can carry over from one place to another, and so forth. Huge
numbers of substances are not merely substances, but bring with them
templates for more concrete substances falling under them. For ex-
ample, the ability to identify cats is easily applied to discovering what
sorts of questions can be asked about individual cats. What color is this
cat (it won™t change as with chameleons), is it tame or untamed (not ap-
plicable to flies), and does it have feline leukemia (not applicable to
dogs) or a loud purr?
Whether we have built in templates and ways of conceptually track-
ing stuffs or real kinds of any particular sort is clearly a matter for em-
pirical research “ research of the sort that the psychologists mentioned
above, among others, have been doing, though I am suggesting a some-
what different framework for interpretation of experimental results.
Without doubt, the results of more traditional studies of concept for-
mation also cast light on how conceptual tracking develops. Although
tradition has pretty single-mindedly taken substance concepts to be
classifiers, much experimental work is easily reinterpreted as implicitly
addressed to the question how we track substances and how we learn to
track them. Examining “the function” from “learning instances plus the
target items to categorize” to “the set of possible category judgments,”
as Billman put it (1992), should help us to discern what kinds of traces
are followed as people attempt conceptual tracking, at various ages, and
for different domains of real kinds. But I believe that experiments need
to be designed and interpreted with it in mind that the cognitive sys-
tems are designed by evolution and tuned by experience to find real-
world substances, not random logically possible ones. Close attention
needs to be paid to the details of real world ontology, to the principles
that hold real substances together, and the relevance of experiments us-
ing artificial objects and kinds should be carefully justified.
One more fundamental medium through which conceptual tracking
is achieved is language. That is what Chapter 6 is about.

Substance Concepts Through
Knowing the Meanings of Words


As I have described substance concepts, having these need not depend
on knowing words. Preverbal humans, indeed, any animal that collects
practical knowledge over time of how to relate to specific substances
needs to have concepts of these. On the other hand, it is clear that lan-
guage interacts with substance concepts in vigorous ways, completely
transforming the conceptual repertoire. Putnam (1975) argued for what
he called “the division of linguistic labor,” according to which laymen
can borrow on the concepts of experts. Though offering an entirely
different analysis, I will conclude similarly, that the public language plays
a crucial role both in the acquisition of substance concepts and also in
their completed structure.
The story so far about substance concepts seems to collide with the
obvious fact that many of these concepts, both for children and adults,
have been acquired without encountering the substances “themselves”
but only by hearing about them. With regard to these very same sub-
stances, moreover, we are often in the position that Kripke (1972) and
Putnam (1975) observed, having no unique descriptions of them in
mind either, so that descriptionist theories of how extensions are deter-
mined also do not fit these cases. I will argue that this entire problem
falls away if we view speech as a direct medium for the perception of
It is traditional to assume that gathering information by being told
things is a radically different sort of process from gathering information
directly through perception. But there is reason to think that this dif-

ference has been greatly exaggerated.1 In fact, uncritically believing
what you hear said may be surprisingly like uncritically believing what
you see. For example, there is experimental evidence that what one is
told goes directly into belief unless cognitive work is done to prevent
this, just as what one perceives through other media does. Loading the
cognitive systems with other tasks, such as having simultaneously to
count backwards by threes, has the effect of facilitating belief fixation
regarding whatever one hears or reads (Gilbert 1993).
There are two things that distinguish direct perception quite sharply
from the acquisition of information through language, but neither im-
plies a difference in immediacy. In direct perception, the spatial and
temporal relation of the perceiver to the object perceived is given,
whereas it is not normally given through language. If you see the cat,
you normally see also its spatial relation to you, and whatever you per-

<< . .

( 29)

. . >>

Copyright Design by: Sunlight webdesign