Content uploaded by Stephen Stich
Author content
All content in this area was uploaded by Stephen Stich on Dec 03, 2020
Content may be subject to copyright.
University of Arkansas Press
Dennett on Intentional Systems
Author(s): STEPHEN P. STICH
Source:
Philosophical Topics,
Vol. 12, No. 1, Functionalism and the Philosophy of Mind
(SPRING 1981), pp. 39-62
Published by: University of Arkansas Press
Stable URL: https://www.jstor.org/stable/43153844
Accessed: 04-04-2020 01:50 UTC
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/43153844?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
University of Arkansas Press
is collaborating with JSTOR to digitize, preserve and extend
access to
Philosophical Topics
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
Dennett on Intentional Systems
STEPHEN P. STICH
University of Maryland
During the last dozen years, Daniel Dennett has been elaborating an
interconnected - and increasingly influential - set of views in the
philosophy of mind, the philosophy of psychology, and those parts
of moral philosophy that deal with the notions of freedom,
responsibility and personhood. The central unifying theme running
through Dennett's writings on each of these topics is his concept of
an intentional system. He invokes the concept to "legitimize"
mentalistic predicates ( Brainstorms , p. xvii),1 to explain the
theoretical strategy of cognitive psychology and artificial
intelligence, and, ultimately, to attempt a reconciliation between
"our vision of ourselves as responsible, free, rational agents, and
our vision of ourselves as complex parts of the physical world of
science" (BS, p. x). My goal in this paper is to raise some doubts
about the "intentional coin" (BS, p. xviii) with which Dennett
proposes to purchase his moral and "mental treasures." Since I aim
to offer a critique of Dennett's views, it is inevitable that much of
what I say will be negative in tone. But this tone should not be
misconstrued. It is my view that Dennett's theories are of great
importance and will shape discussion in the philosophy of mind for
decades to come. Moreover, I think that much of what Dennett says
is close to being true. If we reconstruct his notion of an intentional
system to eliminate its instrumentalism and its unfortunate
infatuation with idealized rationality, we can use the result to give a
better account of common-sense mentalistic notions, and also to
give a clearer and more tenable account of the strategy of cognitive
science. Toward the end of this paper I will sketch the outlines of
such a "de-rationalized" cousin to Dennett's idea of an intentional
system.
39
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
I
In explaining the idea of an intentional system, Dennett's recurrent
illustration is the chess-playing computer. There are, he urges,
three quite different stances we might "adopt in trying to predict
and explain its behavior" (BS p. 237).
First there is the design stance. If one knows exactly how the
computer's program has been designed . . . one can predict the
computer's designed response to any move one makes. One's
prediction will come true provided only that the computer
performs as designed, that is, without breakdown. . . . The
essential feature of the design stance is that we make predictions
solely from knowledge or assumptions about the system's
design, often without making any examination of the innards of
the particular object.
Second, there is what we may call the physical stance. From this
stance our predictions are based on the actual state of the
particular system, and are worked out by applying whatever
knowledge we have of the laws of nature. . . . One seldom adopts
the physical stance in dealing with a computer just because the
number of critical variables in the physical constitution of a
computer would overwhelm the most prodigious human
calculator. . . . Attempting to give a physical account or prediction
of the chess playing computer would be a pointless and herculean
labor, but it would work in principle. One could predict the
response it would make in a chess game by tracing out the effects
of the input energies all the way through the computer until once
more type was pressed against paper and a response was printed.
There is a third stance one can adopt toward a system, and that
is the intentional stance. This tends to be the most appropriate
when the system one is dealing with is too complex to be dealt
with effectively from the other stances. In the case of a chess
playing computer one adopts this stance when one tries to predict
its response to one's move by figuring out what a good or
reasonable response would be, given the information the
computer has about the situation. Here one assumes not just the
absence of malfunction but the rationality of the design or
programming as well.
Whenever one can successfully adopt the intentional stance
toward an object, I call that object an intentional system. The
success of the stance is of course a matter settled pragmatically,
without reference to whether the object really has beliefs,
40
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
intentions, and so forth; so whether or not any computer can be
conscious, or have thoughts or desires, some computers
undeniably are intentional systems, for they are systems whose
behavior can be predicted, and most efficiently predicted, by
adopting the intentional stance towards them. [BS, pp. 237-38;
for a largely identical passage, cf. BS, pp. 4-7]
So any object will count as an intentional system if we can usefully
predict its behavior by assuming that it will behave rationally. And
what is it to behave rationally? Here, Dennett suggests, the full
answer must ultimately be provided by a new sort of theory,
intentional-system theory , which will provide us with a normative
account of rationality. This new theory "is envisaged as a close kin
of - and overlapping with - such already existing disciplines as
epistemic logic, decision theory and game theory, which are all
similarly abstract, normative and couched in intentional language"
(Three Kinds , p. 19). Of course, we already have some "rough and
ready principles" of rationality which we can and do press into
service pending a more detailed normative theory:
(1) A system's beliefs are those it ought to have, given its
perceptual capacities, its epistemic needs, and its biography.
Thus in general, its beliefs are both true and relevant to its life. . . .
(2) A system's desires are those it ought to have, given its
biological needs and the most practicable means of satisfying
them. Thus [naturally evolved] intentional systems desire
survival and procreation, and hence desire food, security, health,
sex, wealth, power, influence, and so forth, and also whatever
local arrangements tend (in their eyes - given their beliefs) to
further these ends in appropriate measure. . . .
(3) A system's behavior will consist of those acts that it would be
rational for an agent with those beliefs and desires to perform.
[TK, pp. 8-9]
Obviously these three principles are very rough and ready
indeed. However, we also have a wealth of more detailed
common-sense principles that anchor our intuitive notion of
rationality. Some of these, in turn, are systematized and improved
upon by existing theories in logic, evolutionary biology and decision
theory. But though the intentional-system theorist can count on
some help from these more developed disciplines, he still has a
great deal of work to do. Neither singly nor severally do these
disciplines tell us what beliefs a given organism or system ought to
41
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
have, what desires it ought to have, or how it should act, given the
beliefs and desires it has. Dennett has no illusions on the point. He
portrays intentional-system theory - the general normative theory
of rationality - as a discipline in its infancy. When the course of our
argument requires some substantive premises about what it would
be rational for a system to believe or do, we can follow Dennett's
lead and let our common-sense intuitions be our guide.
I have been stressing the role of a normative theory of rationality
in Dennett's account of the intentional stance. But there is a second,
equally important, component in his view. According to Dennett,
when we describe an organism or an artifact as an intentional
system, we are making no commitments about the internal physical
workings of the system. Nor are we saying anything about the design or
program of the system . Just as a single program or design description is
compatible with indefinitely many physical realizations, so too a
single intentional description is compatible with indefinitely many
different programs or design descriptions. To view an object as an
intentional system, we must attribute to it a substantial range of
beliefs and desires - the beliefs and desires it would be rational for
such an object to have, given its nature and history. However, we
need not assume that the beliefs and desires attributed correspond
in any systematic way to internal states characterized either
physically or functionally. Dennett makes the point vividly with the
example of two robots each designed to be identical to a given
person, Mary, when viewed from the intentional stance. The first
robot, Ruth, "has internal processes which 'model' Mary's as
closely as you like" (BS, p. 105). It is functionally identical to Mary,
though the two may be quite different physically. Since Mary and
Ruth share a common design or program, they will behave
identically. Thus any beliefs and desires we attribute to Mary we
may attribute also to Ruth, and the attributions will be equally
useful in predicting their behavior. The second robot, Sally, has a
program which is input-output equivalent to Ruth's, though it uses
a quite different computational strategy. "Sally may not be a very
good psychological model of Mary," since "Sally's response delays,
errors and the like may not match Mary's." But at the level of
common-sense descriptions of actions, all three will behave alike.
"... the ascription of all Mary's beliefs and desires (etc.) to Sally will
be just as predictive as their ascription to Ruth so far as prediction of
action goes" (BS, p. 105). So when we adopt the intentional stance,
Mary, Ruth and Sally are indistinguishable.
42
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
Dennett, then, is a self-professed instrumentalist about the
beliefs and desires we ascribe to an object when we adopt the
intentional stance toward it. "... the beliefs and other intentions of
an intentional system need [not] be represented 'within' the system in
any way for us to get a purchase on predicting its behavior by
ascribing such intentions to it" (BS, p. 277). Rather, these "putative
. . . states" can be relegated "to the role of idealized fictions in an
action-predicting, action-explaining calculus" (BS, p. 30). For
Dennett, the belief and desire states of an intentional system are not
what Reichenbach calls "illata - posited theoretical entities." Rather
they are "abstracta - calculation bound entities or logical
constructs" (TK, p. 13). Their status is analogous to the lines in a
parallelogram of forces (TK, p. 20). Of course, it is conceivable that
some objects which are usefully treated as intentional systems really
do have internal states that correspond to the beliefs and desires
ascribed to them in an intentional characterization. As some writers
have suggested, there might be functionally distinct neural belief
and desire stores where each belief and desire is inscribed in an
appropriate neural code. Dennett, however, thinks this is not likely
to be true for people, animals and other familiar intentional
systems.2 Be this as it may, the important point in the present
context is that when we describe an object in intentional-system
terms, we are quite explicitly not making any commitment about its
workings, beyond the minimal claim that whatever the mechanism
causally responsible for the behavior may be, it must be the sort of
mechanism which will produce behavior generally predictable by
assuming the intentional stance.
This completes my sketch of Dennett's notion of intentional
systems. Let us now consider what Dennett wants to do with the
notion. The principal project Dennett has in mind for intentional
systems is "legitimizing" (BS, p. xvii), or providing a sort of
"conceptual reduction" (TK, p. 30) of various notions in
common-sense or folk psychology. The sort of legitimizing Dennett
has in mind is explained by analogy with Church's Thesis. Church
proposed that the informal, intuitive mathematical concept of an
"effective" procedure be identified with the formal notion of a
recursive (or Turing-machine computable) function. The proposal
"is not provable, since it hinges on the intuitive and unformalizable
notion of an effective procedure, but it is generally accepted, and it
provides a very useful reduction of a fuzzy-but-useful
mathematical notion to a crisply defined notion of apparently equal
scope and greater power" (BS, p. xviii; cf. also TK, p. 30). It is
43
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
Dennett's hope to provide the same sort of legitimization of the
notions of folk psychology by showing how these notions can be
characterized in terms of the notions of intentional-system theory.
"... the claim that every mental phenomenon alluded to in folk
psychology is intentional-system-characterizable would, if true,
provide a reduction of the mental as ordinarily understood - a
domain whose boundaries are at best fixed by mutual
acknowledgement and shared intuition - to a clearly defined
domain of entities, whose principles of organization are familiar,
relatively formal and systematic, and entirely general" ( TK , pp.
30-31).
All this sounds reasonable enough - an exciting project, if
Dennett can pull it off. The effort looks even more intriguing when
we note how broadly Dennett intends to cast his net. It is his aim to
show not only that such "program receptive" (BS, p. 29) features
of mentality as belief and desire are intentional-
system-characterizable, but also that "program resistant features of
mentality" like pain, dreams, mental images, and even free will are
"captured in the net of intentional systems" (BS, p. xviii). But a dark
cloud looms on the horizon, one that will continue to plague us. In
much of his work Dennett exhibits an exasperating tendency to
make bold, flamboyant, fascinating claims in one breath, only to
take them back, or seem to, in the next. Thus, scarcely a page after
proclaiming his intention to show that a broad range of
common-sense mental phenomena are intentional-
system-characterizable and thus legitimized, Dennett proclaims
himself to be an eliminative materialist concerning these very same
phenomena. Beliefs, desires, pains, mental images,
experiences - as these are ordinarily understood "are not good
theoretical entities, however well entrenched" (BS, p. xx) the terms
'belief', 'pain', etc. may be in the habits of thought of our society. So
"we legislate the putative items right out of existence" (BS, p. xx).
How are we to make sense of this apparent contradiction?
There is, I think, a plausible - and uncontradictory -
interpretation of what Dennett is up to. The problem he is
grappling with is that the fit between our intuitive folk-
psychological notions and the intentional-system characterizations
he provides for them is just not as comfortable as the fit between the
intuitive notion of effective mathematical procedure and the formal
notion of Turing computability. Our folk-psychological concepts,
"like folk productions generally," are complex, messy, variegated
and in danger of incoherence (TK, p. 16). By contrast, notions
44
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
characterized in terms of intentional-system theory are - it is to be
hoped - coherent, sharply drawn and constructed with a
self-conscious eye for their subsequent incorporation into science
(TK, p. 6). The intentional-system analysans are intended to be
improvements on their analysanda. What they give us is not an
"anthropological" (TK, p. 6) portrait of our folk notions (warts and
all), but rather an improved version of "the parts of folk psychology
worth caring about" (TK, p. 30). So Dennett is an elimina tive
materialist about mental phenomena alluded to in warts-and-all
folk psychology; what are intentional-system-characterizable are
not the notions of folk psychology, but rather related successor
concepts which capture all that's worth caring about.
But now what are we to make of the claim that the intentional
system Ersätze capture all that's worth caring about in folk
psychology? What is worth caring about? Dennett concedes that an
"anthropological" study of unreconstructed folk notions which
includes "whatever folk actually include in their theory, however
misguided, incoherent, gratuitous some of it may be," (TK, p. 6)
would be a perfectly legitimate endeavor. Folk theory may be myth,
"but it is a myth we live in, so it is an 'important' phenomenon in
nature" (TK, p. 6). 3 However, Dennett does not share the
anthropologist's (or the cognitive simulator's) interest in the
idiosyncracies and contradictions embedded in our folk notions.
What is of interest to him, he strongly suggests, is "the
proto-scientific quest": "an attempt to prepare folk theory for
subsequent incorporation into or reduction to the rest of science,"
eliminating "all that is false or ill-founded" (TK, p. 6). If matters
stopped there, we could parse Dennett's "all that's worth caring
about" as "all that's worth caring about for the purposes of science."
But matters do not stop there. To see why, we will have to take a
detour to survey another central theme in Dennett's thinking.
As we have noted, a basic goal of Dennett's theory is to reconcile
"our vision of ourselves as responsible, free, rational agents, and
our vision of ourselves as complex parts of the physical world of
science" (BS, p. x). The conflict that threatens between these two
visions is a perennial philosophical preoccupation. "... the validity
of our conceptual scheme of moral agents having dignity, freedom
and responsibility stands or falls on the question: can men ever be
truly said to have beliefs, desires, intentions? If they can, there is at
least some hope of retaining a notion of the dignity of man; if they
cannot, if men never can be said truly to want or believe, then surely
they never can be said truly to act responsibly, or to have a
45
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
conception of justice, or to know the difference between right and
wrong" (BS, pp. 63-64). Yet many psychologists, most notoriously
Skinner, have denied that people have beliefs, desires and other
mental states.4 This threat to our view of ourselves as moral agents
does not arise only from rabid behaviorism. Dennett sees it lurking
also in certain recently fashionable philosophical theories about the
nature of mental states. Consider, for example, the type-type
identity theory which holds that every mental-state type is to be
identified with a physical-state type - a brain state characterized in
physico-chemical terms. What if it should turn out that there simply
is no physical-state type that is shared by all beings to whom we
commonly attribute the belief that snow is white? If we hang on to
the type-type identity theory, then this very plausible empirical
finding would seem to entail that there is no such mental state as
believing that snow is white. Much the same result threatens from
those versions of functionalism which hold that "each mental type
is identifiable as a functional type in the language of Turing machine
description" (BS, p. xvi). For "there is really no more reason to
believe you and I 'have the same program' in any relaxed and
abstract sense, considering the differences in our nature and
nurture, than that our brains have identical physico-chemical
descriptions" (BS, p. xvi). So if we adhere to functionalism, a
plausible result in cognitive psychology - the discovery that people
do not have the same programs - threatens to establish that people
do not have beliefs at all.5
We can now see one of the principal virtues of Dennett's
instrumentalism about intentional systems. Since describing an
object as an intentional system entails nothing whatever about
either the physico-chemical nature or the functional design of the
mechanism that causes the object's behavior, neither
neurophysiology nor "sub-personal cognitive psychology" (which
studies the functional organization or program of the organism)
could possibly show that the object was not an intentional system .
Thus if beliefs and desires (or some respectable Ersätze) can be
characterized in terms of intentional-system theory, we need have
no fear that advances in psychology or brain science might establish
that people do not really have beliefs and desires. So the viability of
our "conceptual scheme of moral agents" is sustained, in this
quarter at least.6
Now, finally, it is clear how Dennett's preoccupation with moral
themes bears on his elimina tive materialism. Recall that Dennett
proposes to trade our ungainly folk-psychological notions for
46
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
concepts characterized in terms of intentional systems. The claim is
not that the new concepts are identical with the old, but that they are
better. They are clearer, more systematic, free from the incoherence
lurking in folk notions, and they capture everything in folk psychology
that is worth caring about. One of the things worth caring about, for
Dennett, is the suitability of the clarified notions for incorporation
into science. However, if he is to succeed in insulating our moral
world-view from the threat posed by scientific psychology, then
there is obviously something else Dennett must count as worth
caring about. The new concepts built from intentional-system
notions must be as serviceable as the older folk notions in sustaining
our vision of ourselves as persons.
II
In this section I want to examine just how well Dennett's intentional
system Ersätze mirror the notions of folk psychology. My focus will
be on the "program receptive" notions of belief and desire, concepts
which should be easiest to purchase "with intentional coin," and
my claim will be that the fit between our common-sense notions
and Dennett's proffered replacements is a very poor one.7 Of
course, Dennett does not maintain that the fit is perfect, only that
intentional-system theory preserves "the parts of folk psychology
worth caring about" ( TK , p. 30). This is the doctrine I am concerned
to challenge. On my view, the move to an intentional-
system-characterized notion of belief would leave us unable to say a
great deal that we wish to say about ourselves and our fellows.
Moreover, the losses will be important ones. If we accept Dennett's
trade, we will have no coherent way to describe our cognitive
shortcomings nor the process by which we may learn to overcome
them. Equally unwelcome, the thriving scientific study of the
strengths and weaknesses of human reasoning would wither and
die, its hypotheses ruled literally incoherent. What is more, the
instrumentalism of Dennett's intentional-system notions seems to
fly in the face of some deeply rooted intuitions about responsibility
and moral agency. Throughout most of what follows, I will cleave to
the fiction that we already have a tolerably well worked out
normative theory of rationality, or could readily build one, though
in the closing pages I will offer some skeptical thoughts about how
likely this fiction is.
I begin with the problems posed by irrationality. An intentional
system, recall, is an ideally rational system; it believes, wants and
does just what it ought to, as stipulated by a normative theory of
47
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
rationality. People, by contrast, are not ideally rational, and therein
lies a devastating problem for Dennett. If we were to adopt his
suggestion and trade up to the intentional-system notions of belief
and desire (hereafter IS belief & IS desire), then we simply would
not be able to say all those things we need to say about ourselves and
our fellows when we deal with each other's idiosyncrasies,
shortcomings, and cognitive growth.
Consider belief. Presumably no system ought to hold
contradictory beliefs, and all systems ought to believe all the logical
truths, along with all the logical consequences of what they believe
(cf. BS, pp. 11, 20, 44; TK, p. 11). But people depart from this ideal in
a variety of ways. We generally fail to believe all logical
consequences of our beliefs - sometimes because the reasoning
required would be difficult, and sometimes because we simply fail
to take account of one or more of our beliefs. Suppose, for example,
that an astronaut set the controls incorrectly and has sent his craft
into a perilous spin. One possible explanation of his mistake would
be that the on-board computer was down, and he had to
hand-calculate the setting for the controls. He made a mistake in the
calculation, and thus came to have a mistaken belief about what the
setting should be. Another possibility is that, although he knew the
craft was in the gravitational field of a nearby asteroid - indeed he
could see it through the window - he simply forgot to take this into
account in figuring out where the control should be set. There is
nothing in the least paradoxical about these explanations. We offer
similar explanations all the time in explaining our own actions and
those of other people. Indeed, since these explanations are so
intimately bound up with our notions of excuse and blame,
quick-wittedness, absent-mindedness and a host of others, it
boggles the mind to try imagining how we would get on with each
other if we resolved to renounce them. But if, following Dennett, we
agree to swap the folk notion of belief for the intentional-system
notion, then renounce them we must. It simply makes no sense to
attribute inferential failings or inconsistent beliefs to an ideally
rational system.
Our intuitive grasp on the notion of rational desire is rather more
tenuous than our grasp on the analogous notion for belief. Still,
there seem to be many cases in which we want to ascribe desires to
people which are not rational on any plausible reading of that term.
Jones is a successful writer, in good health, with many friends and
admirers. But he says he wants to die, and ultimately takes his own
life. Smith has a dreadful allergy to chocolate, and he knows it. One
48
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
taste and he is condemned to a week of painful, debilitating hives.
But he really wants that chocolate bar at the checkout counter. After
staring at it for a minute, he buys it and gobbles it down. Brown
collects spiders. They are of no economic value, and he doesn't even
think they are very pretty. But it is his hobby. He wants to add to his
collection a specimen of a species found only in the desert. So,
despite his dislike of hot weather, he arranges to spend his vacation
spider hunting in Nevada. By my lights, both Jones' desire and
Smith's are simply irrational. As for Brown, "irrational" seems
much too strong. Yet it is certainly implausible to say that he ought to
want that spider. So, on Dennett's account, it is not a rational desire.
But idealized intentional systems have all and only the desires they
ought to have. Thus if we trade the common-sense notion of want
for Dennett's IS want, we simply will not be able to say that Brown
wants the spider or that Jones wants to die.
The existence of examples like the ones I have been sketching is
not news to Dennett. From his earliest paper on intentional systems
to his most recent, he has struggled with analogous cases.
Unfortunately, however, he is far from clear on what he proposes to
do about them. As I read him, there are two quite dif fernet lines that
he proposes; I will call them the hard line and the soft line. Neither is
carefully spelled out in Dennett's writings, and he often seems to
endorse both within a single paper. Once they have been sharply
stated, I think it will be clear that neither line is tenable.
The Hard Line
The hard line sticks firmly with the idealized notion of an intentional
system and tries to minimize the importance of the gap between IS
beliefs and IS desires and their folk-psychological namesakes. The
basic ploy here is to suggest that when folk psychology ascribes
contradictory beliefs to people or when it insists that a person does
not believe some of the consequences of his beliefs, folk psychology
undermines its own usefulness and threatens to lapse into
incoherence. When this happens, we are forced back to the design
stance or the physical stance:
The presumption of rationality is so strongly entrenched in our
inference habits that when our predictions [based on the
assumption] prove false, we at first cast about for adjustments in
the information possession conditions (he must not have heard,
he must not know English, he must not have seen x, . . . ) or goal
weightings, before questioning the rationality of the system as a
49
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
whole. In extreme cases personalities may prove to be so
unpredictable from the intentional stance that we abandon it, and
if we have accumulated a lot of evidence in the meanwhile about
the nature of response patterns in the individual, we may find
that a species of design stance can be effectively adopted. This is
the fundamentally different attitude we occasionally adopt
toward the insane. [BS, pp. 9-10]
Here, surely, Dennett is just wrong about what we do when
predictions based on idealized rationality prove false. When a
neighborhood boy gives me the wrong change from my purchase at
his lemonade stand, I do not assume that he believes quarters are
only worth 23tf , nor that he wants to cheat me out of the I am due.
My first assumption is that he is not yet very good at doing sums in
his head. Similarly, when a subject working on one of Wason &
Johnson-Laird's deceptively difficult reasoning tasks gets the
wrong answer, we are not likely to assume that he didn't
understand the instructions, nor that he didn't want to get the right
answer. Out first assumption is that he blew it; he made a mistake in
reasoning.8 What misleads Dennett here is that he is focusing on
cases of counter-intuitive or unfamiliar cognitive failings. When
someone seems to have made a mistake we can't readily imagine
ourselves ever making, we do indeed begin to wonder whether he
might perhaps have some unanticipated beliefs and desires. Or if a
person seems to be making enormous numbers of mistakes and
ending up with a substantial hoard of bizarre beliefs, we grow
increasingly reluctant to ascribe beliefs and desires to him at all.
Perhaps we count him among the insane. These facts will assume
some importance later on. But they are of little use to the hard-line
defense of intentional systems. For it is in the diverse domain of
more or less familiar inferential shortcomings that common sense
most readily and usefully portrays people as departing from an
idealized standard of rationality.
Dennett frequently suggests that we cannot coherently describe a
person whose beliefs depart from the idealized standard:
Conflict arises . . . when a person falls short of perfect rationality,
and avows beliefs that either are strongly disconfirmed by the
available empirical evidence or are self-contradictory or
contradict other avowals he has made. If we lean on the myth that
a man is perfectly rational, we must find his avowals less than
authoritative: "You can't mean - understand - what you're
saying!;" if we lean on his right as a speaking intentional system
50
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
to have his word accepted, we grant him an irrational set of
beliefs. Neither position provides a stable resting place; for, as we
saw earlier, intentional explanation and prediction cannot be
accommodated either to breakdown or to less than optimal design , so there
is no coherent intentional description of such an impasse. [BS, 20, last
emphasis added]9
In the paper from which the quote is taken, Dennett uses
'intentional description', 'intentional explanation' and the like for
both common-sense belief-desire accounts and idealized
intentional system accounts. The ambiguity this engenders is
crucial in evaluating his claim. On the idealized intentional systems
reading it is a tautology that "there is no coherent intentional
description of such an impasse." But on the common-sense reading
it is simply false. There is nothing at all incoherent about a
(common-sense) intentional description of a man who has
miscalculated the balance in his checking account!
The fact that folk psychology often comfortably and
unproblematically views people as departing from the standard of
full rationality often looms large in cases where questions of
morality and responsibility are salient. Consider the case of Oscar,
the engineer. It is his job to review planned operations at the factory
and halt those that might lead to explosion. But one day there is an
explosion and, bureaucracy being what it is, three years later Oscar
is called before a board of inquiry. Why didn't he halt the hazardous
operation? It looks bad for Oscar, since an independent expert has
testified that the data Oscar had logically entail a certain equation,
and it is a commonplace amongst competent safety engineers that
the equation is a sure sign of trouble. But Oscar has an impressive
defense. Granted the data he had entails the equation, and granted
any competent engineer would know that the equation is a sign of
trouble. But at the time of the accident neither Oscar nor anyone else
knew that the data logically entailed the equation. It was only six
months after the accident that Professor Brain at Cambridge proved
a fundamental theorem needed to show that the data entail the
equation. Without knowledge of the theorem, neither Oscar nor
anyone else could be expected to believe that the data entail the
equation.
At several places Dennett cites Quine as a fellow defender of the
view that the ascription of inconsistent beliefs is problematic.
To echo a theme I have long cherished in Quine' s work, all the
evidence - behavioral and internal - we acquire for the correctness
51
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
of one of these ascriptions is not only evidence against the other,
but the best sort of evidence. [Reply, p. 74]
However, Dennett misconstrues Quine's point. What Quine urges
is not that any inconsistency is evidence of bad translation (or bad
belief ascription), but rather that obvious inconsistency is a sign that
something has gone wrong. For Quine, unlike Dennett, sees
translation (and belief ascription) as a matter of putting ourselves in
our subject's shoes. And the self we put in those shoes, we are too
well aware, departs in many ways from the standard of optimal
rationality. The point can be made vividly by contrasting Oscar, our
safety engineer, with Otto, a lesser functionary. Otto is charged
with the responsibility of memorizing a list of contingency plans: If
the red light flashes, order the building evacuated; if the warning
light goes on, turn the big blue valve; if the buzzer sounds, alert the
manager. Now suppose that while he is on duty the red light flashes
but Otto fails to order an evacuation. There is a strong prima facie
case that Otto is to be held responsible for the consequences. Either
he failed to see the light (he was asleep or not paying due attention),
or he did not memorize the contingency plans as he was obligated
to, or he has some sinister motive. But, and this is the crucial point,
it will be no excuse for Otto to claim that he had memorized the plan,
saw the light, and was paying attention, but it just never occurred to
him to order the evacuation. It is in these cases of apparently blatant
or "incomprehensible" irrationality that we hunt first for hidden
motives or beliefs. For, absent these, the subject must be judged
irrational in a way we cannot imagine ourselves being irrational; and
it is this sort of irrationality that threatens the application of our
common-sense notions of belief and desire.
In Dennett's writings there are frequent hints of a second strategy
for defending the hard line, a strategy which relies on an
evolutionary argument. He cheerfully concedes that he has "left
[his] claim about the relation between rationality and evolutionary
considerations so open-ended that it is hard to argue against
efficiently" (R, p. 73). Still, I think it is important to try ringing some
arguments out of Dennett's vague meditations on this topic. As I
read him, Dennett is exploring a pair of ideas for showing that the
gap between IS notions and folk notions is much smaller than some
have feared. If he can show this, the hard line will have been
vindicated.
The first idea is suggested by a passage (BS, pp. 8-9) in which
Dennett asks whether we could adopt the intentional stance toward
exotic creatures encountered on an alien planet. His answer is that
52
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
we could, provided "we have reason to suppose that a process of
natural selection has been in effect ..." (BS, p. 8). The argument
seems to be that natural selection favors true beliefs, and thus will
favor cognitive processes which generally yield true beliefs in the
organism's natural environment. So if an organism is the product of
natural selection, we can safely assume that most of its beliefs will be
true, and most of its belief-forming strategies will be rational.
Departures from the normative standard required by the intentional
stance will be few and far between.
For two quite different reasons, this argument is untenable. First,
it is simply not the case that natural selection favors true beliefs over
false ones. What natural selection does favor, is beliefs which yield
selective advantage. And there are many environmental
circumstances in which false beliefs will be more useful than true
ones. In these circumstances, natural selection ought to favor
cognitive processes which yield suitable false beliefs and disfavor
processes which yield true beliefs. Moreover, even when having
true beliefs is optimal, natural selection may often favor a process
that yields false beliefs most of the time, but which has a high
probability of yielding true beliefs when it counts. Thus, for
example, in an environment with a wide variety of suitable foods,
an organism may do very well if it radically overgeneralizes about
what is inedible. If eating a certain food caused illness on a single
occasion, the organism would immediately come to believe (falsely,
let us assume) that all passingly similar foods are poisonous as well.
When it comes to food poisoning, better safe than sorry is a policy that
recommends itself to natural selection.10
The second fault in the argument I am attributing to Dennett is a
subtle but enormously important one. As stated, the argument slips
almost unnoticeably from the claim that natural selection favors
cognitive processes which yield true beliefs in the natural
environment to the claim that natural selection favors rational
belief-forming strategies. But, even if the first claim were true, the
second would not follow. There are many circumstances in which
inferential strategies which from a normative standpoint are
patently invalid will nonetheless generally yield the right answer.
The social-psychology literature is rich with illustrations of
inferential strategies which stand subjects in good stead ordinarily,
but which subjects readily overextend, with unhappy results.11
So long as we recognize a distinction between a normative theory
of inference or decision making and a set of inferential practices
which (in the right environment) generally get the right (or
selectively useful) answer, it will be clear that the two need not, and
53
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
generally do not, coincide. However, in a number of places Dennett
seems to be suggesting that there really is no distinction here, that
by "normative theory of inference and decision" he simply means
"practices favored by natural selection." This move is at the core of
the second idea I see in Dennett for using evolutionary notions to
buttress the hard line. (Cf. R, pp. 73-74.) And buttress it would! For
it would then become tautologous that naturally evolved creatures
are intentional systems, believing, wanting and doing what they
ought, save when they are malfunctioning. Yet Dennett will have to
pay a heavy price for turning the hard line into a tautology. For if this
is what he means by "normative theory of belief and decision," then
such established theories as deductive and inductive logic, decision
theory and game theory are of no help in assessing what an
organism "ought to believe." Natural selection, as we have already
noted, sometimes smiles upon cognitive processes that depart
substantially from the canons of logic and decision theory. So these
established theories and our guesses about how to extend them will
be of no help in assessing what an intentional system should
believe, desire or do. Instead, to predict from the intentional stance
we should need a detailed study of the organism's physiology, its
ecological environment and its history. But predicting from the
intentional stance, characterized in this way, is surely not to be
recommended when we "doubt the practicality of prediction from
the design or physical stance" (BS, p. 8). Nor, obviously, does this
intentional stance promise to yield belief and desire attributions that
are all but co-extensive with those made in common sense.
This is all I shall have to say by way of meeting the hard line head
on. I think it is fair to conclude that the hard line simply cannot be
maintained. The differences separating the IS notions of belief and
desire from their common-sense counterparts is anything but
insubstantial. Before turning to Dennett's soft line, we should note a
further unwelcome consequence of rejecting folk psychology in
favor of intentional-system theory. During the last decade,
cognitive psychologists have become increasingly interested in
studying the strengths and foibles of human reasoning. There is a
substantial and growing literature aimed at uncovering predictable
departures from normative standards of reasoning and decision
making, almost all of it implicitly or explicitly cast in the idiom of folk
psychology.12 Were we to replace folk notions with their
intentional-system analogues, we should have to conclude that all
of this work limning the boundaries of human rationality is simply
incoherent. For, as Dennett notes, "the presuppositions of
54
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
intentional explanation . . . put prediction of lapses in principle
beyond its scope. . ."13 (BS, p. 246).
The Soft Line
In contrast with the hard line, which tries to minimize the size or
importance of the difference between folk and IS notions, the soft
line acknowledges a substantial and significant divergence. To deal
with the problems this gap creates, the soft line proposes some
fiddling with the idealized notion of an intentional system. The
basic idea is that once we have an idealized theory of intentional
systems in hand, we can study an array of variations on the
idealized theme. We can construct theories about "imperfect
intentional systems" (the term is mine, not Dennett's) which have
specified deficiencies in memory, reasoning power, etc. And we can
attempt to determine empirically which imperfect intentional
system best predicts the behavior of a particular subject or species.
Rather than assuming the intentional stance toward an organism or
person, we may assume one of a range of "imperfect intentional
stances," from which it will make sense to ascribe a less than fully
rational set of beliefs and desires. From these various stances we can
give intentional descriptions of our cognitive shortcomings and
elaborate an empirical science which maps the inferential strengths
and weaknesses of humans and other creatures. We can also
legitimize our folk-psychological descriptions of
ourselves - protecting "personhood from the march of science" (2?,
p. 75) - by appeal to the imperfect-intentional-system theory which
best predicts our actual behavior. But genuine intentional-system
theory (sans phrase) would have a definite pride of place among these
theories of imperfect intentional systems. For all of the latter would
be variations on the basic IS framework.
Dennett, with his disconcerting penchant for working both sides
of the street, never flatly endorses the soft line, though it is clear that
he has pondered something like it:
Consider a set T of transformations that take beliefs into beliefs.
The problem is to determine the set Ts for each intentional system
S, so that if we know that S believes p, we will be able to determine
other things that S believes by seeing what the transformations of
p are for Ts. If S were ideally rational, every valid transformation
would be in Ts; S would believe every logical consequence of
every belief (and, ideally, S would have no false beliefs). Now we
know that no actual intentional system will be ideally rational; so
55
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
we must suppose any actual system will have a T with less in it.
But we also know that, to qualify as an intentional system at all, S
must have a T with some integrity; T cannot be empty. [BS, p. 21]
In the next few sentences, however, Dennett expresses qualms
about the soft line:
What rationale could we have, however, for fixing some set
between the extremes and calling it the set for belief (for S, for
earthlings, for ten-year-old-girls)? This is another way of asking
whether we could replace Hintikka's normative theory of belief
with an empirical theory of belief, and, if so, what evidence we
would use. "Actually," one is tempted to say, "people do believe
contradictions on occasion, as their utterances demonstrate; so
any adequate logic of belief or analysis of the concept of belief
must accommodate this fact." But any attempt to legitimize human
fallibility in a theory of belief by fixing a permissible level of error
would be like adding one more rule to chess: an Official Tolerance
Rule to the effect that any game of chess containing no more than
k moves that are illegal relative to the other rules of the game is a
legal game of chess. [BS, p. 21]
In a more recent paper, Dennett sounds more enthusiastic about the
soft line:
Of course we don't all sit in the dark in our studies like mad
Leibnizians rationalistically excogitating behavioral predictions
from pure, idealized concepts of our neighbors, nor do we derive all
our readiness to attribute desires to a careful generation of them
from the ultimate goal of survival. . . . Rationalistic generation of
attributions is augmented and even corrected on occasion by
empirical generalizations about belief and desire that guide our
attributions and are learned more or less inductively. . . .
I grant the existence of all this naturalistic generalization, and its role
in the normal calculation of folk psychologists - i.e., all of
us. ... I would insist , however , that all this empirically obtained lore is laid
over a fundamental generative and normative framework that has the
features I have described. [TK, pp. 14^-15, last emphasis added]
Whatever Dennett's considered view may be, I think the soft line
is clearly preferable to the hard line. Indeed, the soft line is similar to
a view that I have myself defended. 14 As a way of focusing in on my
misgivings about the soft line, let me quickly sketch my own view
and note how it differs from the view I am trying to foist on Dennett.
Mine is an effort squarely situated in what Dennett calls "the
56
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
anthropological quest" ( TK , p. 6). I want to describe as accurately as
possible just what we are up to when we engage in the "folk
practice" of ascribing beliefs to one another and dealing with one
another partly on the basis of these ascriptions. My theory is an
elaboration on Quine' s observation that in ascribing beliefs to others
"we project ourselves into what, from his remarks and other
indications, we imagine the speaker's state of mind to have been,
and then we say what, in our language, is natural and relevant for us
in the state thus feigned" ( Word & Object , p. 219). As I see it, when
we say S believes that p we are saying that S is in a certain sort of
functionally characterized psychological state, viz., a "belief state."
The role of the "content sentence," p, is to specify which belief state
it is. If we imagine that we ourselves were now to utter p in earnest,
the belief we are attributing to S is one similar (along specified
dimensions) to the belief which would cause our own imagined
assertion. One of the dimensions of similarity that figures in belief
ascription is the pattern of inference that the belief states in question
enter into. When the network of potential inferences surrounding a
subject's belief state differs substantially from the network
surrounding our own belief that p, we are reluctant to count the
subject's belief as a belief that p. Thus we will not have any
comfortable way of ascribing content to the belief states of a subject
whose inferential network is markedly different from ours. Since we
take ourselves to approximate rationality, this explains the fact,
noted by Dennett, that intentional description falters in the face of
egregious irrationality. It also explains the fact, missed by Dennett,
that familiar irrationality - the sort we know ourselves to be guilty
of - poses no problem for folk psychology.
A full elaboration of my theory would be a long story, out of place
here. What is important for our present purposes is to note the
differences between my account arid what I have been calling
Dennett's soft line. These differences are two. First, my story does
not portray folk psychology as an instrumentalist theory. Belief states
are functional states which can and do play a role in the causation of
behavior. Thus folk psychology is not immune from the advance of
science. If it turns out that the human brain does not have the sort of
functional organization assumed in our folk theory, then there are
no such things as beliefs and desires. Second, the notion of idealized
rationality plays no role at all in my account. In ascribing content to
belief states, we measure others not against an idealized standard
but against ourselves. It is in virtue of this Protagorean parochialism
that the exotic and the insane fall outside the reach of intentional
explanation.
57
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
So much for the difference between my view and Dennett's. Why
should mine be preferred? There are two answers. First, I think it is
simply wrong that we ordinarily conceive of beliefs and desires in
instrumentalist terms - as abstracta rather than illata. It is, however,
no easy task to take aim at Dennett's instrumentalism, since the
target refuses to stay still. Consider:
Folk psychology is instrumentalistic. . . . Beliefs and desires of folk
psychology ... are abstracta. [TK, p. 13]
It is not particularly to the point to argue against me that folk
psychology is in fact committed to beliefs and desires as
distinguishable, causally interacting illata; what must be shown is
that it ought to be. The latter claim I will deal with in due course.
The former claim I could concede without embarrassment to my
overall project, but I do not concede it, for it seems to me that the
evidence is quite strong that our ordinary notion of belief has next
to nothing of the concrete in it. [TK, p. 15]
The ordinary notion of belief no doubt does place beliefs
somewhere midway between being illata and being abstracta.
[TK, p. 16]
In arguing for his sometimes instrumentalism Dennett conjures
the sad tale of Pierre, shot dead by Jacques in Trafalgar Square.
Jacques
is apprehended on the spot by Sherlock; Tom reads about it in the
Times and Boris learns of it in Pravda. Now Jacques, Sherlock, Tom
and Boris have had remarkably different experiences - to say
nothing of their earlier biographies and future prospects - but
there is one thing they share: they all believe that a Frenchman
has committed a murder in Trafalgar Square. They did not all say
this, not even "to themselves"; that proposition did not, we can
suppose, "occur to" any of them, and even if it had, it would have
had entirely different import for Jacques, Sherlock, Tom and
Boris. [TK, p. 15]
Dennett's point is that while all four men believe that a Frenchman
committed a murder in Trafalgar Square, their histories, interests
and relations to the deed are so different that they could hardly be
thought to share a single, functionally characterizable state. This is
quite right, but it does not force us to view beliefs as abstracta. For if,
as my theory insists, there is a similarity claim embedded in belief
ascriptions, then we should expect these ascriptions to be both
vague and sensitive to pragmatic context. For Jacques and Boris
58
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
both to believe that a Frenchman committed a murder in Trafalgar
Square, they need not be in the very same functional state, but only
in states that are sufficiently similar for the communicative purposes
at hand.
As Dennett notes, one need not be crucially concerned with what
"folk psychology is in fact committed to." Since he aims to replace
folk psychology with intentional-system notions, it would suffice to
show that the instrumentalism of these latter notions is no
disadvantage. But here again I am skeptical. It is my hunch that our
concept of ourselves as moral agents simply will not sit comfortably
with the view that beliefs and desires are mere computational
conveniences that correspond in no interesting way to what goes on
inside the head. I cannot offer much of an argument for my hunch,
though I am encouraged by the fact that Dennett seems to share the
intuition lying behind it:
Stich accurately diagnoses and describes the strategic role I
envisage for the concept of an intentional system, permitting the
claim that human beings are genuine believers and desirers to
survive almost any imaginable discoveries in cognitive and
physiological psychology, thus making our status as moral agents
well neigh invulnerable to scientific disconfirmation. Not 'in
principle' invulnerable, for in a science-fiction mood we can
imagine startling discoveries (e.g., some 'people' are organic
puppets remotely controlled by Martians) that would upset any
particular home truths about believers and moral agenthood you
like
Now if our concept of moral agenthood were really compatible with
the intentional-system construal of beliefs and desires, it is hard to
see why the imagined discovery about Martians should be in the
least unsettling. For, controlled by Martians or not, organic puppets
are still intentional systems in perfectly good standing. So long as
their behavior is usefully predictable from the intentional stance,
the transceivers inside their heads sanction no skepticism about
whether they really have IS beliefs and IS desires. But Dennett is
right, of course. We would not count his organic puppets as
believers or moral agents. The reason, I submit, is that the morally
relevant concept of belief is not an instrumentalistic concept.
The second reason for preferring my line to Dennett's soft line is
that the idea of a normative theory of beliefs and desires, which is
central to Dennett's view, plays no role in mine. And this notion, I
would urge, is one we are best rid of. Recall that from the outset we
59
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
have been relying on rough and ready intuitions about what an
organism ought to believe, desire and do, and assuming that these
intuitions could be elaborated and systematized into a theory. But I
am inclined to think that this assumption is mistaken. Rather, it
would appear that the intuitions Dennett exploits are underlain by a
variety of different ideas about what an organism ought to believe or
desire, ideas which as often as not pull in quite different directions.
Sometimes it is an evolutionary story which motivates the intuition
that a belief or desire is the one a well-designed intentional system
should have. At other times intuitions are guided by appeal to logic
or decision theory. But as we have seen, the evolutionary account of
what an organism ought to believe and desire just will not do for
Dennett, since it presupposes an abundance of information about
the ecological niche and physiological workings of the organism.
Nor is there any serious prospect of elaborating logic and decision
theory into a suitably general account of what an organism ought to
believe and desire. Indeed, apart from a few special cases, I think
our intuitions about what an organism ought to believe and desire
are simply nonexistent. The problem is not merely that we lack a
worked-out normative theory of belief and desire; it runs much
deeper. For in general we have no idea what such a normative
theory would be telling us. We do not really know what it means to
say that an organism ought to have a given belief or desire. Consider
some examples:
Ought Descartes to have believed his theory of vortices?
Ought Nixon to have believed that he would not be impeached?
Ought William James to have believed in the existence of a
personal God?
Should all people have perfect memories, retaining for life all
beliefs save those for which they later acquire negative
evidence?
In each of these cases our grasp of what the question is supposed to
mean is at best tenuous. The prospects of a general theory capable of
answering all of them in a motivated way are surely very dim.
Worse still, the general theory of intentional systems that Dennett
would have us work toward must tell us not only what people in
various situations ought to believe, but also what other animals
ought to believe. Ought the frog to believe that there is an insect
flying off to the right? Or merely that there is some food there? Or
perhaps should it only have a conditional belief: if it flicks its tongue
in a certain way, something yummy will end up in its mouth?
60
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
Suppose the fly is of a species that causes frogs acute indigestion.
Ought the frog to believe this? Does it make a difference how many
fellow frogs he has seen come to grief after munching on similar
bugs? A normative theory of desire is, if anything, more
problematic. Should I want to father as many offspring as possible?
Should the frog?
To the extent that these questions are obscure, the notion of a
normative theory of belief and desire is obscure. And that obscurity
in turn infects much of what Dennett says about intentional systems
and the intentional stance. Perhaps Dennett can dispel some of the
mystery. But in the interim I am inclined to think that the
normatively appropriate attitude is the skepticism I urged in my
opening paragraph.15
NOTES
1. References to Dennett's writings will be identified in parentheses in the text. I
will use the following abbreviations:
BS = Daniel Dennett, Brainstorms (Montgomery, VT: Bradford Books, 1978).
TK = Daniel Dennett, 'Three Kinds of Intentional Psychology," forthcoming in a
Thyssen Philosophy Group volume edited by R. A. Healey. (The manuscript from
which I quote is dated November, 1978.)
R = Daniel Dennett, "Reply to Professor Stich," Philosophical Books, 21, 2 (April,
1980).
TB = Daniel Dennett, "True Believers: The Intentional Strategy and Why It
Works," forthcoming in a volume of 1979 Herbert Spencer Lectures on the topic of
Scientific Explanation (Oxford University Press).
2. For his arguments on this point, cf., "Brain Writing and Mind Reading," (BS,
pp. 39-50) and "A Cure for the Common Code," (BS, pp. 90-108).
3. This "anthropological quest," when pursued systematically is the business of
the cognitive simulator. Cf., for example, Roger Shank & Robert Abelson, Scripts,
Plans, Goals and Understanding (Hillsdale, NJ: Lawrence Erlbaum Associates, 1977);
also Aaron Sloman, The Computer Revolution in Philosophy (Atlantic Highlands, NJ:
Humanities Press, 1978), chapter 4.
4. Skinner often muddies the waters by claiming to offer "translations" of
common-sense mentalistic terms into the language of behaviorism. But, as Dennett
and others have noted, (BS, pp. 53-70) these "translations" generally utterly fail to
capture the meaning or even the extension of the common-sense term being
"translated."
5. For an elaboration of the point, cf. Thomas Nagel, "Armstrong on the Mind,"
Philosophical Review, 79 (1970) pp. 394r-403.
6. An entirely parallel strategy works for those other common-sense mental
phenomena which Dennett takes to be essential to our concept of ourselves as
persons - e.g., consciousness (BS, p. 269). If we can give an acceptable intentional
system Ersatz for the folk-psychological notion of consciousness, we need have no
fear that advances in science will threaten our personhood by showing that the
notion of consciousness is otiose in the causal explanation of our behavior.
61
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
7. For some qualms about Dennett's treatment of "program resistant" features of
mentality like pains, see my "Headaches," Philosophical Books, April, 1980.
8. Cf., P. C. Wason & P. N. Johnson-Laird, The Psychology of Human Reasoning :
Structure and Content (London: B. T. Batsford, Ltd., 1972).
9. For parallel passages, cf., TB, p. 19; R, p. 74; BS, p. 22.
10. For a detailed discussion of some examples and further references, cf., H. A.
Lewis/The Argument From Evolution," Proceedings of the Aristotelian Society,
Supplementary Vol. LIII, 1979, also my "Could Man Be an Irrational Animal?"
(forthcoming).
11. Cf., Richard Nisbett & Lee Ross, Human Inference (Englewood Cliffs, NJ:
Prentice-Hall, 1980).
12. E.g., Nisbett & Ross, op. cit., and Wason & Johnson-Laird, op. cit., along with
the many studies cited in these books.
13. Dennett appends the following footnote to the quoted sentence: "In practice
we predict lapses at the intentional level ('You watch! He'll forget all about your
knight after you move the queen') on the basis of loose-jointed inductive hypotheses
about individual or widespread human frailties. These hypotheses are expressed in
intentional terms, but if they were given rigorous support, they would in the process
be recast as predictions from the design or physical stance" (BS, p. 246). So the
scientific study of intentionally described inferential shortcomings can aspire to no
more than "loose-jointed hypotheses" in need of recasting. But cf. TK, pp. 11-12,
where Dennett pulls in his horns a bit.
14. In "On The Ascription of Content," to appear in A. Woodfield, ed. Thought and
Object, Oxford University Press.
15. I have learned a good deal from the helpful comments of Bo Dahlbom, Robert
Cummins, Philip Pettit and Robert Richardson.
62
This content downloaded from 128.112.200.107 on Sat, 04 Apr 2020 01:50:10 UTC
All use subject to https://about.jstor.org/terms
