Searle ConscUnconsc&Intentionality 1991 OCR PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 22
At a glance
Powered by AI
One of the key takeaways is that the author believes contemporary analytic philosophy and cognitive science have largely avoided seriously discussing the nature of consciousness. Another takeaway is that the author argues intentional states are either actually or potentially conscious states.

One of the most important reasons for the fear of consciousness according to the author is that the presuppositions and methods of contemporary philosophy are ill-equipped to deal with the subjectivity of conscious states.

The author believes that any intentional state is either actually or potentially a conscious intentional state, and cognitive science cannot avoid studying consciousness.

PHILOSOPHICAL ISSUES, 1

3 Consciousness, 1991

Consciousness,
Unconsciousness and
Intentionality

John R. Searle*

One of the most amazing things about the past half century
or so in analytic philosophy of mind is the scarcity of seri-
ous work on the nature of consciousness. Even works pur-
portedly about consciousness have very little to say about
the general structure of consciousness or about its special
features. Thus for example of three recent books contain-
ing "Consciousness"t in their titles not one contains even
an attempt to state a comprehensive account of the struc-
ture of conscious states, much less state a general theory
that will account for subjectivity, the stream of conscious-
ness, the qualitative character of conscious states, etc. In
each case consciousness is regarded not as a primary sub-
ject matter in the philosophy of mind but as a "problem",
a potential embarrassment to the author's theory, which is,

*© John R. Searle.
1 Churchland, Paul M. (1984), Jackendoff, Ray (1987) and Lycan,
William G. (1987).
46 JOHN R. SEARLE

in each case, some version of "functionalism" or "material-


ism" or "computationalism". What goes for the philosophy
of mind also goes for most -not all- mainstream philoso-
phy of language. You can test this for yourself: Which of
the many works on meaning by, say Carnap or Quine, has
given you the greatest insight into the special features of the
relationships between meaning and consciousness?
I think there are many deep reasons for this fear of con-
sciousness in contemporary analytic philosophy. One of the
most important of these is that the presuppositions and meth-
ods of contemporary philosophy are ill equipped to deal with
the subjectivity of conscious states. A similar reluctance to
accept the consequences of ontological subjectivity also af-
flicts psychology and cognitive science generally.
In order to account for the mind without consciousness, one
must postulate some other sorts of phenomena. The most ob-
vious solution is to postulate unconscious mental processes
to do the jobs that traditional philosophy, psychology, and
common sense ascribed to conscious mental processes. This
manoeuvre takes different forms, but the general tendency
in cognitive science has been to drive a wedge between, on
the one hand, conscious, subjective mental processes, which
are not regarded as a proper subject of scientific investiga-
tion; and, on the other hand, those which are regarded as the
genuine subject matter of cognitive science, and which, there-
fore, must be objective. The general theme is always that the
unconscious mental processes are more significant than the
conscious ones. Perhaps the earliest canonical statement is
in Lashley's claim,2 "No activitity of mind is ever conscious"
(Lashley's italics). Another extreme version of this approach
is to be found in Jackendoff's claim 3 that in fact there are two
"notions of mind" , the "computational mind" and the "phe-
nomenological mind". I believe that it is a profound mistake

2 Lashley, Karl (1956). I don't think Lashley means this literally.


I think he means that the processes by which the various features of
conscious states are produced are never conscious. But even that is an
overstatement; and the fact that he resorts to this sort of hyperbole is
revealing of the theme I am trying to identify.
3 Jackendoff, Ray (1987).
3. CONSCIOUSNESS/UNCONSCIOUSNESS 47

to try to describe and explain mental phenomena without ref-


erence to consciousness. In this article I will argue that any
intentional state is either actually or potentially a conscious
intentional state, and for that reason cognitive science cannot
avoid studying consciousness. The attribution of any inten-
tional phenomena to a system, whether "computational" or
otherwise, is dependent on a prior acceptance of our ordinary
notion of the mind, the conscious "phenomenological" mind.
In order to substantiate this claim, I am also going to
have to explore the notion of an unconscious mental state.
The course of this exploration is rather more complex than
I would like, and I need to begin by reminding the reader
of the distinctions between ontology, causation, and episte-
mology. For any phenomenon, but for biological phenomena
especially, we need to know:
1. What is its mode of existence? (ontology)
2. What does it do? (causation)
3. How do we find out about it ? (epistemology)
So, for example, if we were examining the heart, the an-
swer to our three questions is: the heart is a large piece
of muscle tissue located in the chest cavity (ontology); the
heart functions to pump blood throughout the body (cau-
sation); and we find out about the heart indirectly through
such methods as using stethoscopes, cardiograms and taking
pulse, and directly by opening up the chest cavity and looking
at the heart (epistemology) . Now, these distinctions apply to
both conscious and unconscious mental states. The history
of the philosophy of mind in recent decades is in large part
a series of confusions between these three questions. Thus,
behaviorism confuses the epistemology of the mental with
the ontology (we find out about mental states by observing
behavior, so mental states just consist in behavior and dispo-
sitions to behavior). And functionalism consists in confusing
the causation with the ontology (mental states have causal
relations to input stimuli, other mental states, and output
behavior, so mental states just consist in having these causal
relations) .
Our naive, pre-theoretical notion of an unconscious mental
state is the idea of a conscious mental state minus the con-
sciousness. But what exactly does that mean? How could
48 JOHN R. SEARLE

we subtract the consciousness from a mental state and still


have a mental state left over? Since Freud, we have grown
so used to talking about unconscious mental states that we
have lost sight of the fact that the answer to this question
is by no means obvious. Yet it is clear that we do think of
the unconscious on the model of the conscious. Our idea of
an unconscious state is the idea of a mental state that just
happens then and there to be unconscious; but we still un-
derstand it on the model of a conscious state in the sense
that we think of it as being just like a conscious state and
as one which in some sense could have been conscious. This
is clearly true, for example, in Freud, whose notions of both
what he calls "preconscious" and "unconscious" states are
built on a rather simple model of conscious states.4 Perhaps
at its most naive, our picture is something like this: uncon-
scious mental states in the mind are like fish deep in the sea.
The fish that we can't see underneath the surface have ex-
actly the same shape they have when they surface. The fish
don't lose their shapes by going under water. Another simile:
unconscious mental states are like objects stored in the dark
attic of the mind. These objects have their shapes all along,
even when you can't see them. We are tempted to smile at
these simple models, but I think something like these pictures
underlies our conception of unconscious mental states; and
it is important to try to see what is right and what wrong
about that conception.
In recent decades, the connection between consciousness
and intentionality is being gradually lost in theoretical writ-
ings in linguistics, cognitive science, and philosophy. There
has been an effort of varying degrees of explicitness to try to
separate the issues concerning intentionality from those con-
cerning consciousness. I think the underlying -and perhaps
unconscious- motivation for this urge to separate intention-
ality from consciousness, even among people who do not
share the ideology of the behaviorist-materialist tradition,
is that we do not know how to explain consciousness, and we
would like to get a theory of intentionality which will not be

4 Freud, S. (1949) pp. 19-25.


3. CONSCIOUSNESS/UNCONSCIOUSNESS 49

discredited by the fact that we do not have a theory of con-


sciousness. The idea is to treat intentionality "objectively",
to treat it as if the subjective features of consciousness did
not really matter to it. For example, many functionalists
will concede that functionalism can't "handle" consciousness
(this is called the problem of qualia), but they think that
this issue doesn't matter to their accounts of belief, desire,
etc., since these intentional states have no quale, no special
conscious qualities. They can be treated as if they were com-
pletely independent of consciousness. Similarly, both the idea
of some linguists that there are rules of syntax that are psy-
chologically real but totally inaccessible to consciousness and
the idea of some psychologists that there are complex infer-
ences in perception that are genuine psychological inferential
processes but inaccessible to consciousness, try to separate
intentionality from consciousness. The idea in both cases is
not that there are mental phenomena which just happen to
be unconscious, but somehow in some way they are in prin-
ciple inaccessible to consciousness. They are not the sort of
thing that could be or could ever have been conscious.
I think these recent developments are mistaken. There are
deep reasons having to do with the nature of mental phe-
nomena whereby our notion of an unconscious mental state
is parasitic on our notion of a conscious state. Of course, at
any given moment, a person may be unconscious; he or she
may be asleep, in a coma, etc.; and of course, many men-
tal states are never brought to consciousness. And no doubt
there are many which could not be brought to consciousness
for one reason or another -they may be too painful and
hence too deeply repressed for us to think of them, for ex-
ample. Nonetheless, not every state of an agent is a mental
state and not even every state of the brain which functions
essentially in the production of mental phenomena is itself
a mental phenomenon. So what makes something mental
when it is not conscious? For a state to be a mental state,
and a fortiori for it to be an intentional mental state, certain
conditions must be met: What are they?
To explore these questions let us first consider cases which
are clearly mental, though unconscious, and contrast them
with cases which are "unconscious" because not mental at
50 JOHN R. SEARLE

all. Think of the difference, for example, between my belief


(when I am not thinking about it) that the Eiffel Tower is in
Paris, and the myelination of the axons in my central nervous
system. There is a sense in which both are unconscious. But
there is a big difference in that the structural states of my
axons couldn't be themselves conscious states, because there
isn't anything mental about them at all. I assume for the
sake of this argument that myelination functions essentially
in the production of my mental states, but even if myeli-
nated axons were themselves objects of experiences, even if
I could feel inwardly the state of the myelin sheathes, still
the actual structures are not themselves mental states. Not
every unconscious state in my brain which (like myelination)
functions essentially in my mental life is itself a mental state.
But the belief that the Eiffel Tower is in Paris is a genuine
mental state. Even though it happens to be a mental state
that most of the time is not present to consciousness. So
here are two states in me, my belief and my axon myelina-
tion; both have something to do with my brain; and both are
unconscious. But only one is mental, and we need to get clear
about what makes it mental and the connection between that
feature -whatever it is- and consciousness.
There are at least two constraints on our conception of
intentionality which any theory of the unconscious must be
able to account for: It must be able to account for the dis-
tinction between phenomena which are genuinely intentional
and those which in some respects behave as if they were but
which are not in fact. This is the distinction I have discussed
elsewhere between intrinsic and as-ijforms of intentionality.5
And second it must be able to account for the fact that in-
tentional states represent their conditions of satisfaction only
under certain aspects and those aspects must matter to the
agent. My unconscious belief that the Eiffel Tower is in Paris
satisfies both of these conditions. My having that belief is a
matter of intrinsic intentionality, and not a matter of what
anybody else chooses to say about me or how I behave or
what sort of stance someone might adopt toward me. And

5 Searle, John R. (1980), Searle, John R. (1984a) and especially


Searle, John R. (1984b).
3. CONSCIOUSNESS/UNCONSCIOUSNESS 51

the belief that the Eiffel Tower is in Paris represents its con-
ditions of satisfaction under certain aspects and not others.
It is, for example, distinct from the belief that 'the tallest
iron structure built in France before 1900 is located in the
French capital', even assuming that the Eiffel Tower is identi-
cal with the tallest iron structure built in France before 1900,
and Paris is identical with the French capital. We might say
that every intentional state has a certain aspectual shape; and
this aspectual shape is part of its identity, part of what makes
it the state that it is.

1
These two features, the fact that an unconscious intentional
state must nonetheless be intrinsically mental and the fact
that it must have a certain aspectual shape, have important
consequences for our conception of the unconscious. Specif-
ically, we understand the notion of an unconscious mental
state only as a possible content of consciousness, only as
the sort of thing which, though not conscious, and perhaps
impossible to bring to consciousness for various reasons, is
nonetheless the sort of thing that could be or could have been
conscious. Someone might have mental states which are im-
possible to bring to consciousness -because of repression or
brain lesions or what have you- but if they are genuine un-
conscious mental states, they can't be the sort of thing which
in the nature of the case no one could ever have brought to
consciousness. To be mental at all, they must be at least
possible candidates for consciousness.
The argument for this thesis is a bit complex, but the cen-
tral idea behind it can be given a simple formulation: the
concept of an intrinsic intentional mental state is the concept
of something that has an aspectual shape. All representation
is under aspects. You can see this, if it is not obvious on its
face, by reminding yourself that mental contents are possible
or actual contents of thoughts or experiences. What you can
believe, you have to be able to think; and what you can per-
ceive, you have to be able to experience perceptually. But
the notions of thinking and experiencing are notions which
52 JOHN R. SEARLE

imply the presence of aspectual shapes and that in turn im-


plies accessibility to consciousness. The link, then, between
intentionality and consciousness lies in the notion of an as-
pectual shape. To be intentional, a state or process must be
thinkable or experienceable; and to be thinkable or experi-
enceable, it must have an aspectual shape under which it is
at least in principle, consciously thinkable or experienceable.
It must be the sort of thing that could be the content of a
conscious thought or experience.
I will now spell out this argument in more detail. For the
sake of clarity I will number the major steps in setting out the
argument, though I do not mean to imply that the argument
is a simple deduction from axioms. Sometimes, indeed, I will
be approaching the same goal from more than one path.
1. The first step is simply to remind ourselves that there
is a distinction between intrinsic intentionality and as-if in-
tentionality; only intrinsic intentionality is genuinely mental.
I have argued at some length for this rather obvious distinc-
tion in other writings (Searle 1980, Searle 1984a, and Searle,
1984b), and I will not repeat the arguments here. I believe
the distinction is obviously correct, but the price of giving it
up would be that everything then becomes mental, because
relative to some purpose or other anything can be treated
as-if it were mental. E.g., water flowing downhill behaves
as-if it had intentionality. It tries to get to the bottom of
the hill by ingeneously seeking the line of the least resistance,
it does information processing in order to calculate the size
of rocks, the angle of the slope, the pull of gravity, etc. But
if water is mental then everything is mental.
2. The second step is to point out that the distinction be-
tween the intrinsic and the as-if lies neither in the behavior
of the system in question nor in the fact that there are under-
lying causal mechanisms which cause the behavior nor in the
complexity of the system. These features are not sufficient
to account for the distinction, since they can be duplicated
equally in both intrinsic and as-if systems.
3. The distinction does lie, at least in part, in the fact
that the idea of an intentional state is closely connected to
the ideas of thinking and experiencing. To be a possible
belief something has to be a possible thought content: It is a
3. CONSCIOUSNESS/UNCONSCIOUSNESS 53

conceptual truth that anything that can be believed can be


thought. Similarly anything that can be a perception can be
the content of a perceptual experience. Of course there are
plently of unconscious thoughts but even if unconscious they
still have to be thoughts.
4. Thoughts and experiences and hence intrinsic inten-
tional states generally, have a certain sort of aspectual shape.
They represent their conditions of satisfaction under aspects.
For example, the desire for water can be a different desire
from the desire for H 2 0, even though there is no way to sat-
isfy the one without satisfying the other. (This aspectual
shape of intentional states explains the referential opacity of
ascriptions of intentional states. The ascriptions are opaque
because the states themselves are aspectual. 6 )
5. This aspectual feature must matter to the agent. It
must exist from his/her point of view.
It is, for example, from my point of view that there can
be a difference for me between between my wanting water
and my wanting H2 0, even though the external behavior that
corresponds to these desires may be identical in each case.
In the case of conscious thoughts, the way that the aspectual
shape matters is that it is constitutive of the way the agent
thinks about a subject matter: I can think about my thirst
for a drink of water without thinking at all about its chemical
composition. I can think of it as water without thinking of
it as H 2 0.
Its obvious how it works for conscious thoughts and expe-
riences, but how does it work for unconscious mental states?
One way to get at the main question of this discussion is to
ask, "What fact about an unconscious intentional state gives
it the particular aspectual shape that it has, Le. what fact
about it makes it the mental state that it is?"
6. The aspectual feature cannot be exhaustively or com-
pletely characterized solely in terms of third person, behav-
ioral, or even neurophysiological predicates. None of these
is sufficient to give an exhaustive account of em the way it
seems to the agent.

6See Searle, John R. (1983). Chapter 7, for an extended discussion


of this issue.
54 JOHN R. SEARLE

Behavioral evidence concerning the existence of mental


states, including even evidence concerning the causation of a
person's behavior, no matter how complete, always leaves the
aspectual character of intentional states underdetermined.
There will always be an inferential gulf between the evidence
for the presence of the aspect and the aspect itself. (This is
one feature of the other minds problem.)
It is less obvious that a similar condition holds for neuro-
physiological facts, but it does. Since the neurophysiologi-
cal facts are always causally sufficient for any set of mental
facts 7 someone with perfect causal knowledge might be able
to make the inference from the neurophysiological to the in-
tentional at least in those few cases where there is a law-like
connection between the facts specified in neural terms and
the facts specified in intentional terms. But even in these
cases, if there are any, there is still an inference. The specifi-
cation of the neurophysiological in neurophysiological terms
is not yet a specification of the intentional, even though the
neurophysiological is causally sufficient to fix the mental.
Any account that uses only a third person objective vocab-
ulary will leave their aspectual character underdetermined;
because no third person objective vocabulary, by itself, will
be sufficient to characterize all of the aspectual facts. Thus
to take an example of behavioral evidence, the fact that a
person consumes H 2 0 only if it is described to him as "wa-
ter" and not if it is described as "H 2 0" would reasonably lead
us to conclude that the person desires the substance under
the aspect, "water" and does not desire it under the aspect
"H 2 0". But this is still an inductive inference; the behavior
still underdetermines the aspect in that it is still in principle
possible for the person to exhibit exactly that behavior and
yet represent his own desires to himself in ways that are not
completely manifested to us in his behavior.

7 For these purposes I am contrasting "neurophysiological" and "men-


tal", but of course on my view of mind body relations, the mental simply
is neurophysiological at a higher level ( see Searle, (1984a)). Icontrast
mental and neurophysiological as one might contrast humans and ani-
mals without thereby implying that the first class is not included in the
second. There is no dualism implicit in my use of this contrast.
3. CONSCIOUSNESS/UNCONSCIOUSNESS 55

To take the more extreme case of neurophysiological evi-


dence, imagine that we had a perfect science of the brain that
enabled us to conclude with empirical certainty from the fact
that a man was in state S that he wanted a substance under
the aspect "water" and not under the aspect "H 2 0". All
the same the characterization "S" is still not yet a charac-
terization of those aspects. S will both cause and realize the
aspects but under that description it is still not constitutive
of the aspects.
This point is so important that it is worth digressing briefly
to explain some of its implications. No amount of purely
third person behavioral evidence is sufficient to entail all
of the aspectual features of intentional states. Such third
person evidence always leaves the aspectual features under-
determined. This is one of the weaknesses of behaviorism
as a thesis in the philosophy of mind. Behaviorism in the
philosophy of mind is now generally regarded as discredited,
however, it survives vestigially in the philosophy of language.
And you see it in an extreme form in Quine's writings. If you
think, as he does, that all of the facts that there are about
linguistic meaning are public third person facts, and if you
think, as again he does (and as again I believe is mistaken),
that the principle that language is public implies that all the
facts that there are about meanings are third person facts,
then you get not merely the underdetermination of aspectual
facts by the behavioral facts but you get an indeterminacy
of aspectual facts. Thus, Quine argues that within certain
limits there simply is no fact of the matter about under what
aspects a speaker represents a situation when he utters an
expression that we are inclined to translate as "rabbit". I
have argued elsewhere that this view is simply a reductio ad
absurdum of linguistic behaviorism. Since it is obvious from
my own case that there are determinate aspectual facts, ifbe-
haviorism leads to a denial of this point that can only show
that behaviorism is false. 8
This is not the place to go into detail, but I believe similar
remarks can be made about functionalist analyses of inten-

8 Searle, John R. (1987).


56 JOHN R. SEARLE

tional states, at least insofar as they try to give a third per-


son, objective, account of intentionality. Quine's inadvertent
refutation of behaviorism can be applied mutatis mutandis
to functionalism. Quine's argument shows that any account
of purely third person, objective phenomena given in func-
tional terms will always leave the precise details of aspectual
facts underdetermined. 9
Now, these six considerations lead to our first significant
conclusion:
7. There is a sense, then, in which intentional states, con-
scious or unconscious, are irreducibly subjective.
The aspectual character is irreducibly subjective in the
sense that no characterization in purely neutral third per-
son terms will ever be sufficient to express how the aspectual
character seems to the agent, but how it seems to the agent
is essential to its identity.
So far so good. We seem to have labored heavily to arrive
at the common sense conclusion that there is something sub-
jective about mental states even when they are unconscious.
But this leads to a very puzzling question: how could uncon-
scious intentional states be subjective if there is no subjective
feel to them, no "qualia", no what-it-feels-like-for-me to be
in that state? Indeed how could the ontology of the uncon-
scious be anything other than completely objective? To begin
to probe this question, let us ask what is the ontology of or-
dinary mental states, such as my belief that the Eiffel Tower
is in Paris, when I am not thinking about such beliefs. And
to avoid muddying the waters, let's leave out halfconscious
thoughts, peripheral consciousness, nagging but suppressed
conscious states and other shadowy phenomena. What is
going on, ontologically speaking, when I have a belief that
is totally and completely unconscious but nonetheless there?
If there are no conscious neurophysiological processes going
on in my brain, then the only other processes would be un-
conscious neurophysiological processes, that is; the ontology

9 For this reason functionalism, like the behaviorism that preceded it,
remains programmatic. To my knowledge, no functionalist has so far
given anything like a plausible analysis of even one intentional state.
3. CONSCIOUSNESS/UNCONSCIOUSNESS 57

of the unconscious when unconscious is entirely neurophysio-


logical. The difference, for example, between my unconscious
belief about the Eiffel Tower and the unconscious myelinated
condition of my axons is not that the one is a feature of my
brain in some mental form and the other is purely neuro-
physiological; they are both purely neurophysiological. You
can see this by imagining that the person is totally uncon-
scious, e.g. in a sound dreamless sleep. Now lots of mental
attributions are still true of that person, but the only men-
tal ontology is neurophysiological. Indeed this point about
unconscious states can be put in the form of a general prin-
ciple:
8. The ontology of unconscious mental states, at the time
they are unconscious, can only consist in the existence of
purely neurophysiological phenomena. At the time the states
are totally unconscious there simply is nothing else going on
except neurophysiological processes.
But now we seem to have a contradiction: the ontology
of unconscious intentionality is entirely describable in third
person, objective neurophysiological terms, but all the same
the states are irreducibly subjective. How can this be? I
believe there is only one solution to this puzzle. The apparent
contradiction is resolved by pointing out that:
9. The notion of an unconscious intentional state is the
notion of a state which is a possible conscious thought or ex-
perience. There are plenty of unconscious phenomena, but to
the extent that they are genuinely mental they must in some
sense preserve their aspectual shape even when unconscious,
but the only sense that we can give to the notion that they
preserve their aspectual shape when unconscious is that they
are possible contents of consciousness.
This is the main conclusion of this article. But this an-
swer to our first question immediately gives rise to another
question: What is meant by "possible" in the previous sen-
tence? After all, I grant that it might be quite impossible
for the state to occur consciously, because of brain lesion,
repression, or other causes. So, in what sense exactly must it
be a possible content of a thought or experience? This ques-
tion leads to our next conclusion, which is really a further
explanation of 9, and is implied by 8 and 9 together:
58 JOHN R. SEARLE

10. The ontology of the unconscious consists in objective


features of the brain capable of causing subjective conscious
thoughts. Hence when we describe something as an uncon-
scious intentional state we are characterizing an objective
ontology in virtue of its causal capacity to produce subjectiv-
ity. But the existence of these causal features is consistent
with the fact that in any given case their causal powers may
be blocked by some other interfering causes -such as psy-
chological repression or brain damage.
The possibility of interference by various forms of pathology
does not alter the fact that any unconscious intentional state
is the sort of thing that is in principle accessible to conscious-
ness. It may be unconscious not only in the sense that it does
not happen to be conscious then and there, but also in the
sense that for one reason or another the agent simply could
not bring it to consciousness, but it must be the sort of thing
that can be brought to consciousness because it must be the
sort of thing that can be the content of a thought or experi-
ence.
To summarize: So far I have tried to establish that in-
tentional states, conscious or unconscious, essentially have
an aspectual character and this aspectual character is essen-
tially subjective, in the sense that it cannot be exhaustively
accounted for in third person "objective" terms. But since
unconscious mental states consist in nothing but objective
neurophysiological features of the brain how can they have
an intrinsic subjective character? I have suggested that the
only explanation for this fact is that unconscious intentional
states while not conscious are at least potentionally so in
the sense that they are the sort of things which could be
conscious. When we characterize an unconscious intentional
state in terms of its aspectual character, we are character-
izing a present brain state in terms of its causal capacity to
produce a conscious thought or experience.
Paradoxically, the naive mentalism of my view of the mind
leads to a kind of dispositional analysis of unconscious mental
phenomena; only it is not a disposition to behavior, but a
'disposition' -if that is really the right word- to conscious
thoughts.
So the overall picture that emerges is this. There is nothing
going on in my brain but neurophysiological processes. Those
3. CONSCIOUSNESS/UNCONSCIOUSNESS 59

processes are capable of generating conscious states, (which


are, of course, higher-level features of the neurophysiological
systems and hence neurophysiological themselves). But of
the unconscious neurophysiological features, some are mental
and some are not. The difference is not in consciousness, for
they are both, by hypothesis, unconscious. The difference is
that the mental ones are candidates for consciousness. That's
all. There isn't any aspectual shape at the level of neurons
and synapses.
In my skull, there is just the brain with all its intricacy.
All my mental life is lodged in the brain. But what in my
brain is my "mental life"? Just two things: conscious states
(of course, caused by neurophysiological processes and real-
ized in the structures of the brain) and those neurophysio-
logical states and processes that -given the right attendant
circumstances- are capable of generating conscious states.
There isn't anything else to the ontology of the unconscious.

2
I want to illustrate these points further by imagining a case
in which we would have a use for the notion of "unconscious
pain". We don't normally think of unconscious pains, and
many people, I believe, would accept the Cartesian notion
that in order for something to be a genuine pain, it has to
be conscious. But I think it is easy to invoke contrary intu-
itions. Consider the following: it is a very common occurence
for people who suffer from chronic pains, say, chronic back
pains, that sometimes the pain makes it difficult for them to
go to sleep. And indeed, once they have fallen asleep, there
sometimes are occasions during the night when their con-
dition causes them to wake up. Now, how exactly shall we
describe these cases? Shall we say that during sleep there re-
ally was no pain, but that the pain began when they woke up
and that they were awakened by neurophysiological processes
which normally would cause pain, but didn't cause pains be-
cause at the time they were asleep? Or shall we say, on the
other hand, that the pain, i.e. the pain itself, continued both
before, during and after their sleep, but that they were not
60 JOHN R. SEARLE

consciously aware of the pain while they were asleep? My


intuitions find the second just as natural, indeed probably
more natural, than the first. However, the important thing
is to see that there is no substantive issue involved. We are
simply adopting an alternative vocabulary for describing the
same sets of facts. But now consider the second vocabulary:
on this vocabulary, we say that the pain was for a while con-
scious, then it was unconscious, then it was conscious again.
Same pain; different states of consciousness of that one and
the same pain. We might increase our urge to speak this way
if we found that the person, though completely unconscious,
made bodily movements during sleep which served to protect
the painful portion of his body.
Now what exactly is the ontology of the pain when it is
unconscious? Well, the answer seems to me quite obvious.
What inclines us to say that the pain continued to exist even
though unconscious is that there was an underlying neuro-
physiological process that was capable of generating a con-
scious state and capable of generating behavior appropriate
to someone who had that conscious state. And in the exam-
ple as described, that is exactly what happened.
But now if I am right about this then it is hard to see
how there could be any factual substance to the old disputes
about whether unconscious mental states really exist. If you
grant my argument so far, then I am unable to see how it
could be other than a purely verbal terminological matter,
different only in complexity from the issue about the exis-
tence of unconscious pains as I just described it. One side
insisted that there really are unconscious mental states; the
other insisted that if they were really mental, why then, they
must be conscious. But what facts in the world are supposed
to correspond to these two different claims?
The evidence that the Freudians adduced involved causal
histories, behavior, and conscious admissions by the agent
-all of which seemed only interpretable on the assumption
of an unconscious mental state, which was just like a con-
scious state except for being unconscious. Consider a typical
sort of case. A man under hypnosis is given a posthypnotic
suggestion to the effect that he must crawl around on the
floor after coming out of the hypnotic trance. Later, when
3. CONSCIOUSNESS/UNCONSCIOUSNESS 61

conscious, he gives some completely extraneous reason for his


behavior. He says, e.g. "I think 1 may have lost my watch
on this floor somewhere", whereupon he proceeds to crawl
around on the floor. Now we suppose, with good reason I
believe, that he is unconsciously obeying the order, that he
unconsciously intends to crawl around on the floor because
he was told to by the hypnotist; and that the reason he gives
for his behavior is not the real reason at all.
But assuming that he is totally unconscious of his real mo-
tives, what is the ontology of the unconscious, right then
and there, supposed to be? To repeat our earlier question,
what fact corresponds to the attribution of the unconscious
mental state at the time the agent is acting for a reason of
which he is totally unconscious? If the state really is to-
tally unconscious, then the only facts are the existence of
neurophysiological states capable of giving rise to conscious
thoughts and to the sort of behavior appropriate for someone
having those thoughts.
Sometimes there may be several inferential steps between
the latent unconscious mental state and the manifest con-
scious intentionality. Thus, we are told, the adolescent boy
who revolts against the authority of the school is uncon-
sciously motivated by hatred of his father. The school sym-
bolizes the father. But again, as in the hypnosis case we have
to ask, what is the ontology of the unconscious supposed to
be when unconscious? And in this case, as in the hypnosis
case, the identification of a specific aspectual shape to the
unconscious must imply that there is in the neurophysiology
a capacity to produce a conscious thought with that very
aspectual shape.
Once you see that the description of a mental state as "un-
conscious" is the description of a neurophysiological ontology
in terms of its causal capacity to produce conscious thoughts
and behavior, then it seems there could not be any factual
substance to the ontological question: Do unconscious men-
tal states really exist? All that question can mean is: Are
there unconscious neurophysiological states of the brain ca-
pable of giving rise to conscious thoughts and to the sorts of
behavior appropriate for someone having those thoughts? Of
course neither side thought of the issue this way, but perhaps
62 JOHN R. SEARLE

part of the intensity of the dispute derived from the fact that
what looked like a straight ontological issue -do unconscious
states exist?- was really not an ontological issue at all.
I am not sure I am right about this, but it does seem at least
prima facie that the old Freudian arguments -involving all
that evidence from hypnotism, neuroses, etc.- are not so
much conclusive or inconclusive as they are factually empty.
The issue is not less important for being conceptual or ter-
minological, but it is important to understand that it is not
a factual issue about the existence of mental entities which
are neither physiological nor conscious.

3
This account of the unconscious has a useful consequence
that I want to call attention to immediately. An old puzzle
about intentional states has to do with the absence of any
clear principle of individuation, and this problem is especially
acute for unconscious beliefs. How many unconscious beliefs
do I have? We don't know how to get started answering
that question. Earlier, I said glibly that I had a belief that
was unconscious most of the time to the effect that the Eiffel
Tower is in Paris. But do I also believe the following?
Station wagons are inedible.
or
Doctors wear underwear.
If someone asked me whether doctors wear underwear or
whether station wagons are inedible, I would have no diffi-
culty in anwering; but it seems funny to think of them as
unconscious beliefs that I have had all along. Whereas it
doesn't seem quite as funny to think of the belief about the
Eiffel tower as an unconscious belief. Why the difference?
I can so confidently answer questions about doctors, station
wagons and Paris because I have a set of capacities realized in
my brain that enable me to generate conscious thoughts and
hence generate answers to questions in which I express my
conscious thoughts. As long as my thoughts are unconscious
they consist only in a neuroanatomy and a neurophysiology
3. CONSCIOUSNESS/UNCONSCIOUSNESS 63

that has capacities for generating conscious thoughts and be-


havior.
The difference between the case of the Eiffel Tower and
the other two cases is that I have already had the conscious
thought that the Eiffel Tower is in Paris. I learned it at
school, I have seen the Eiffel Tower, I climbed to the top,
etc. But the other cases are simply manifestations of the
capacity of my brain to generate an indefinitely large number
of different conscious thoughts. It is hard but not impossible
to count speech acts and even to count one's own conscious
thoughts insofar as one can individuate them by content.
But when it comes to one's unconscious beliefs, the question
of counting is of a different sort altogether: I can't count
my unconscious beliefs because there isn't anything there to
count except in terms of what I have already consciously
thought or in terms of what I could consciously think. But
the latter class is indefinitely large and does not reflect a
fixed preexisting set of mental representations in the brain.
Our ordinary ways of thinking and talking incline us to
think of memory as like a big filing cabinet in which we store
a whole lot of information in some language or other -as if
written on my brain were a whole lot of sentences in English
(or in the "language of thought"). And certain researchers in
cognitive science have encouraged this conception by postu-
lating that in the brain there are a whole lot of unconscious
"mental representations". But as an ontological claim (and
how else are we supposed to take it?) that picture is ob-
viously wrong. What we have instead are a whole lot of
capacities in the brain for generating conscious thoughts.

4
Let us return to the question that I asked at the beginning
of this article: Can we really think of unconscious states as
being like submerged fish or like furniture in the dark attic of
the mind? I think these pictures are inadequate in principle
because they are based on the idea of a reality which appears
and then disappears. But in the case of consciousness, the
only reality is the appearance. The submerged belief, unlike
the submerged fish, can't keep its conscious shape even when
64 JOHN R. SEARLE

unconscious; for the only reality of that shape is the shape


of conscious thoughts. To repeat, the ontology of the uncon-
scious is strictly the ontology of a neurophysiology capable of
generating the conscious.

5
Now oddly enough, this connection between consciousness
and intentionality is lost in discussions of the unconscious
mental processes in contemporary linguistics, philosophy and
cognitive science. Many of the phenomena which are cited as
explanatory psychological features simply could not have any
psychological reality because they are not the sort of things
that could be mental states.
This is a central feature of much contemporary cognitive
science, and it is disguised from us by the vocabulary. Some
of the key terms are, in effect, a set of puns: "information pro-
cessing", "intelligent behavior", "rule following", and "cog-
nition" are all used in two quite distinct senses, only one
of which is genuinely mental. The reason for this is that
the authors in question want a third-person objective sci-
ence but they also want it to be about a mental reality. It
is literally impossible to have both of these features, so they
disguise their failure by using a vocabulary that looks men-
tal (what could be more mental than engaging in 'intelligent
behavior'?) but which has been stripped of any mental con-
tent. And they can get away with this because they can
claim that the mental reality they claim to be discussing
is all "unconscious"; but that expression now becomes the
biggest pun of all because there are two completely different
sorts of phenomena called "unconscious", unconscious men-
tal phenomena and unconscious phenomena which have no
mental reality at all.

6
But someone might object: "Well, why does it matter? Why
don't we just scrap the old time folk psychological vocabulary
3. CONSCIOUSNESS/UNCONSCIOUSNESS 65

once and for all and get on with the genuine science of cogni-
tion? Why does it matter whether or not we use a mentalis-
tic vocabulary to des.cribe those brain processes which cause
genuine mental phenomena or whether we confine ourselves
to a purely neutral physiological vocabulary?" The short an-
swer is that it is crucial to understanding the character of
the processes involved that we have a clear distinction be-
tween those which are mental ( hence also physiological) and
those which are only physiological. This can perhaps be il-
lustrated by the example of the famous studies of what the
frog's eye tells the frog's brain. 1o The retina in the frog's
eye filters out most of the stimulus before transmitting the
rest of the signal to the frog's brain. The processes in the
retina do not literally involve any rule following, nor cogni-
tion, nor intelligence. They are simply brute physiological
processes. However, their effect is crucial for the frog's con-
scious intelligent behavior. By making it possible for the frog
to have certain visual experiences and not others, they make
it possible for the frog to eat and survive.
I give the example of the frog, because here I take it the
facts are obvious. But the same thing should be equally ob-
vious about those neurophysiological processes in the human
brain which enable us to have genuine intelligence, cognition,
and rule following, but which are not themselves cases of in-
telligence, cognition, or rule following. Unless we are clear
about this distinction, we have no hope of understanding
how the neurophysiology produces its crucial mental conse-
quences; and indeed, many of the disappointments of cog-
nitive science derive from its failure to pose the question in
the appropriate terms. We need, in short, to turn Lashley's
claim upside down: Roughly speaking, all genuinely mental
activity is either conscious or potentially so. All of the other
activities of the brain are simply non-mental, physiological
processes, some of which produce conscious and unconscious
mental processes.

lOLettvin,J.Y.; Maturana, H.R.; et al. (1959).


66 JOHN R. SEARLE

7 REFERENCES

Churchland, Paul, M. (1984). Matter and consciousness:


A Contemporary Introduction to the philosophy of mind.
Cambridge, Mass.: MIT Press.
Freud, Sigmund (1949). Outline of Psycho-Analysis.
Translated by James Strachey. The Hogarth Press: Lon-
don.
Jackendoff, Ray (1987). Consciousness and the Computa-
tional Mind. Cambridge: MIT Press, A Bradford Book.
Lashley, K. (1956). "Cerebral Organization and Behavior."
In R. Solomon, S. Cobb, and W. Penfield (eds.). The Brain
and Human Behauiour, pp. 1-18, Baltimore: Williams and
Wilkins.
Lettvin, J.Y.; Maturana, H.R.; McCulloch, W.S.; and
Pitts, W.R. (1959). "What the Frog's Eye Tells the Frog's
Brain", Proceedings of the I.R.E., 47: pp. 1940-51.
Lycan, William, G. (1987). Consciousness. Cambridge,
Mass.: MIT Press. Searle, John R. (1980). "Intrinsic In-
tentionality". Reply to criticisms of "Minds, Brains and
Programs," The Behavioral and Brains Sciences. Vol. 3,
pp.450456.
Searle, J. R. (1983). Intentionality. An Essay in the Phi-
losophy of Mind. Cambridge University Press. Cambridge.
Searle, John R. (1984a). Minds, Brains and Science. Har-
vard University Press. Cambridge, Massachusetts.
Searle, John R. (1984b). "Intentionality and its Place in
Nature", Synthese, Vol. 61, D. Reidel Publishing Company.
pp. 3-16.
Searle, J .R. (1987). "Indeterminacy, Empiricism and the
First Person". Journal of Philosophy. February.

You might also like