A Critical Examination Of D. M. Armstrong'S Materialistic Theory Of Mind

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1.The sign or “target" for pages apparently lacking from the document
photographed is "Missing Paga(s)". If it was possible to obtain the mining
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific p ages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed a s
received.
Xerox University Microfilms
300 North Zoob Road
Ann Arbor, Michigan 46100
WALKER, Kendrick Wayne, 1943-
A CRITICAL EXAMINATION OF D. M.
ARMSTRONG'S MATERIALISTIC THEORY OF
MIND.
University of Southern California, Ph.D., 1974
Philosophy
Xerox University Microfilms , Ann Arbor, Michigan 48106
r,
I
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED.
A CRITICAL EXAMINATION
OF D. M. ARMSTRONG'S
MATERIALISTIC THEORY OF MIND
by
Kendrick Walker
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Philosophy)
August 1974
UNIVERSITY OF SOUTHERN CALIFORNIA
TH E G RADUATE SCHOOL
U N IV E R S ITY PARK
LOS AN G ELES. C A LIF O R N IA 8 0 0 0 7
This dissertation, written by
/ ( i n t / r t c A • d O t f M t r
under the direction of h.Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
1 -
Dttn
Date.
ANALYTICAL TABLE OF CONTENTS
CHAPTER I
INTRODUCTION: THE TASK OF THIS ESSAY
Section 1
A. The mind-body problem arises over the attempt
to answer two fundamental questions: (1) what
kinds of things are there? and (2) how are
these things related?
B. D. M. Armstrong attempts to answer these ques
tions and does so by answering the second one
first. His answer to the second question
entails an answer to the first, however.
C. Armstrong divides his work into two sections.
He first attempts to discredit Dualism in its
various forms. This is done to make way for
his second aim, the presentation of Central-
state Materialism.
D. What I hope to show in this paper is that Arm
strong does no serious damage to Dualism nor
does he succeed in making his version of
Materialism plausible.
Section 2
A. A brief history of the mind-body problem since
Descartes is presented. The history is organ
ized around whether there is a legitimate dis
tinction between mind and body and if so, whether
there are entities corresponding to each or
either term and finally, if there are such enti
ties, what the relationship between them is.
B. Depending on the answers to these questions, the
views on the issue are divided into two main
categories: Dualism and Monism.
ii
Pages
1-2
2-3
3
4
4-5
C. Descartes, Malebrance, Leibniz, Spinoza, Huxley
are discussed as representative of the various
forms dualism may take while Berkeley is cited
for his monist view. This leads us to a dis
cussion of materialism, specifically one version
of it, which is the topic of this paper.
CHAPTER II
A CRITIQUE OF ARMSTRONG'S CRITIQUE OF DUALISM
Section 1
A. In order to establish Central-state Materialism
Armstrong thinks he must first discredit
Dualism.
B. Nevertheless, he disavows interest in the truth
or falsity of either Dualism or Central-state
Materialism; the truth or falsity of this rival
position will be decided by science. Arm
strong's concern is purely conceptual.
C. Hence, Armstrong hopes to show, not that Dualism
is false, but that it is "implausible." He pre
sents five major arguments to this effect and
they form the subject of Chapter One of this
essay.
Section 2
A. In his first argument, Armstrong charges Carte
sian Dualism with being "formal and empty."
B. This follows from our supposed lack of "positive
knowledge" of mental substance.
Section 3
A. But this requires an analysis of what Armstrong
means by "positive knowledge."
B. The first indication is that Armstrong thinks we
do have "positive knowledge" of physical
substance.
C. Further, what Armstrong might mean is clarified
in his discussion of a principle of individua
tion for non-physical objects.
Pages
D. In that discussion "positive knowledge" is seen
to rest on "concrete acquaintance," and this is
embodied in Armstrong's second argument. 28-30
Section 4
A. But we are no better off concerning "positive
knowledge" until we know what Armstrong means by
"concrete acquaintance." It is argued here that
Armstrong means nothing by "concrete acquaint
ance" but sense experience. 30-31
B. This follows not only from the language Arm
strong uses to characterize "concrete acquaint
ance" but from his Materialism as well. 31-33
C. But if Armstrong equates "concrete Acquaintance"
with sense experience, his second argument needs
to be reformulated to express this. 34
D. But the reformulated argument is subject to
criticism on three counts. 34
E. First, it requires mental substance to be access
ible to sense experience, something it cannot, by
definition, provide. And, if Armstrong insists
on this requirement, he has begged a crucial ques
tion against the Dualist 35
F. Second, this argument undercuts Armstrong's claim
that we do have "positive knowledge" about
physical substance. 36
G. But this brings to light Armstrong's confused
discussion of physical substance 36
H. His official position is that "substance" refers
to what is "logically capable of independent
existence." 36-38
I. But questions immediate arise. Independent of
what? What does Armstrong mean by "logically
independent?" How can physical substances be
logically related? Further, none of Armstrong's
examples meet his definition of substance. 38-42
J. Armstrong's working definition of physical sub
stance approximates the notion of substance as
substratum. 42-44
iv
S l
Pages
K. But this position on physical substance opens
Armstrong to the very criticisms he levels
against Cartesian mental substance. 44-46
Section 5
A. The principle embodied in Armstrong's revised
second argument also undermines Armstrong's
major claim about self-awareness. That is, this
principle requires that we be "concretely
acquainted" with dispositions. 46
B. But dispositions, either physical or mental, are
not given to sense experience. 4 7-48
C. It becomes clear that Armstrong cannot maintain
his empiricist principle concurrently with his
views on physical substance and the disposi
tional nature of our mental life. But if he
relaxes his principle his criticisms of mental
substance are vitiated. 48-52
Section 6
A. Armstrong's third argument against Dualism
charges that Dualism fails to account for the
"unity" of man. 53-54
B. But "unity" does not necessarily rule out being
composed of parts. To insist that "unity"
requires being composed of only one "part" surely
begs the question against the Dualist. 54-57
Section 7
It is contended that Armstrong is not simply
trying to show Dualism "implausible"; he is
really concerned with showing it to be false.
This is evidence by the relationship between
Armstrong's Materialism and Cartesian Dualism.
If the one is true the other is false. Also,
Armstrong's claim that "there are no good
philosophical reasons for denying that man is
nothing but a material object," under one mean
ing of "good reason," clearly asserts that
Dualism is false. 57-61
Section 8
A. Armstrong's fourth argument concludes that the
Dualist cannot provide a principle of individ
uation for mental entities.
B. This is contrasted with physical substance where
the spatial relation provides a clear principle
of individuation. But this is assumed by Arm
strong, never argued for. He neglects the argu
ment that spatial relations presuppose numerical
difference.
C. Again, Armstrong's major complaint against any
individuating principle for mental entities is
that we have had no "concrete acquaintance"
with such a principle.
D. Again, Armstrong might deny that he means sense
experience by "concrete acquaintance." Yet it
is hard to see what else he could mean.
Section 9
A. Armstrong's last two complaints against Dualism
are treated together. In effect, he wants to
know exactly where and when the mind acts on
the body, and vice versa.
B. Armstrong first wants to know when the mind
emerges in the history of the individual. He
allows the Dualist three possible answers none
of which is acceptable since none accord with
modern physiology in particular and with the
"workings of the world" in general.
C. Second, Armstrong wants to pinpoint the where of
mind-body interaction. He rejects any Dualist
answer because he claims neurophysiology has
failed to find any event in any neural series
that could be the required mental event.
D. But it is clear that not all neurophysiologists
fully agree with Armstrong.
E. In addition, there is no a priori reason why
something spatial cannot causally affect some
thing non-spatial, and vice versa.
vi
Pages
61
62-63
63-66
66-67
68
69-72
72-74
74-77
77-78
Pages
F. Finally, Armstrong demands the "mechanism" of
mind-body interaction when he is never clear
about the "mechanism" of body-body causation.
In fact, Armstrong never presents a clear dis
cussion about what he takes the causal relation
to be. Further, given the consideration of Hume
and Ducasse, for example, it appears that the
questions Armstrong asks about mind-body causa
tion may equally be asked about physical
causation. 78-85
Section 10
Summary of key points. 85-86
CHAPTER III
CENTRAL-STATE MATERIALISM STATED
Section 1
A. The core of Central-state Materialism is the
identity asserted between mind and brain. 87-89
B. Since the identity claim is contingent, Armstrong
must be able to give "logically independent"
descriptions of both terms of the identity.
Apparent difference, then, is the starting point
for the proposed identity. Armstrong sees this
as his "great problem."
C. With "logically independent" descriptions in
mind, Armstrong defines the brain ostensively.
The definition of mind presents him with dif
ficulties. Mind must be defined without intro
ducing uniquely mental properties and without
mention of the brain or Central Nervous System.
D. Armstrong's solution is topic neutrality. If
mind can be defined without mentioning any
properties whatsoever, then he has avoided
introducing any uniquely mental properties
or of describing mind in physical terms.
Section 2
A. Armstrong sees the similarity of his definition
of mind to that of Place's and Smarts's. He is
90-91
91-93
vii
Pages
careful to distinguish his from theirs. The
distinction is that Armstrong describes mind in
terms of causes and effects, whereas Place and
Smart speak solely in terms of effects. 93-95
B. Armstrong believes a fuller account of our
mental life may be given by talking not only
about stimuli affecting us, but about our
response to that stimuli as well. 95-96
Section 3
A. But speaking in terms of stimuli and response
looks like Behaviorism. So, Armstrong is careful
to distinguish between his view and Behaviorism.
The two views part over the nature of
dispositions. 96
B. For Armstrong, dispositions may be actual states
of the objects they are truly predicated of.
But this requires a mental disposition to be an
"inner" state of the person (the brain). But
since Behaviorism denies that there are any
"inner" states of man's mental life, he must
deny what Armstrong asserts. Given this, we
argue that the distinction Armstrong has in
mind between his and the Behaviorist's account
is not clearly drawn, since the criticisms he
levels against the Behaviorist apply to his 97-
position as well. 103
C. By way of analogy and summary, Armstrong thinks
the claim "The gene is the DNA molecule" is a 103-
good model for his identity claim. 105
CHAPTER IV
ARMSTRONG'S ANALYSIS OF SELF-AWARENESS
Section 1
A. Armstrong claims that we know nothing about the
"intrinsic nature" of our mental states. But
he hedges by claiming that we do know some
thing about them. We may be directly aware of
our mental states either as actual causes,
dispositions to cause or as resembling some 106-
other mental state. 107
viii
Pages
B. We hope to show that the only reasons Armstrong
offers to support these claims is an analogy
with touch perception and that this analogy
falls apart upon inspection.
Section 2
Armstrong's view that we are directly aware
of causes, dispositions to cause and resem
blances in our mental life is presented.
Section 3
Armstrong's comparison between "inner sense"
and sense experience is outlined. By way of
analogy, Armstrong argues that in sense per
ception we are directly aware of the same
components we are aware of in our mental
life, i.e., causes, dispositions to cause and
resemblances. It is noted that even if we are
aware of these same components in sense percep
tion, this does not show that the same holds
true for "inner sense.”
Section 4
A. Armstrong's analogy is presented. He claims
that when we come into contact with some
object about which we know nothing, what is
given to the sense of touch is the causal
relation, "pressure."
3. But, to complete his analogy between such
instances of touch perception and "inner
sense," Armstrong must convince us that in
the case he constructs we not only are
directly aware of the causal relation but
of dispositions and resemblances as well.
C. But there seems no necessity to reify "pres
sure". We resist doing so with the other
ways an object may come into contact with us.
D. Of course, there is no a priori reason why we
cannot feel a causal relation. The issue
should be decided by experience and observa
tion. And with this in mind, we argue that
when we come into contact with an object,
what we feel is the object. "Pressing" is
just one (of many) ways of describing the
total event.
107-
108
108-
111
111-
113
113-
114
114
114-
116
11 fi
ll?
E. But even if we do "feel" an actual causal
relation, Armstrong still has to show clearly
how we sense (feel?) a disposition. It is
argued here that given what dispositions are,
we cannot be aware of them in the manner Arm
strong requires.
F. Even if Armstrong is speaking of a disposi
tion of the causal relation itself, the same
criticisms apply.
G. And, if we do not immediately perceive causes
and dispositions, then it can hardly be argued
that we immediately perceive resemblances
among them.
H. Summary of Chapter IV.
CHAPTER V
THREE IMPLICATIONS OF ARMSTRONG'S IDENTITY CLAIM
Section 1
A. That sensations, or "raw feels," appear to be
radically different than any physical events
is held by respected philosophers and physi
ologists. Armstrong must show that this dif
ference is mere appearance.
B. But rather them show that such sensations are
actually identical with some brain state, we
argue that Armstrong defines sensations out of
existence. Hence, Armstrong is arguing that
sensations are identical with some brain state,
though there really are no sensations. Arm
strong does this by reducing sensations to
perceptions and perceptions to acquiring of
beliefs. No room is left for the original
"raw feels."
C. Armstrong's two possible rejoinders are either
tautologous or circular.
Section 2
A. Further, since we are ignorant of our mental
Pages
117-
120
120-
121
121-
123
123-
125
12 6- *
129
129-
133
133-
135
x
Pages
states, Armstrong's identity claim becomes
one wherein A is identical with B though we
are ignorant of A.
B. The question arises about the possibility of
discovering that Armstrong's claim is true.
How do we discover that A = B when we are
ignorant of A? Armstrong's reply to this
would probably be in terms of the discovered
connections between forms of consciousness
and cranial activity.
C. This introduces Armstrong's conviction that
science will discover that his identity claim
is true, But it is argued here that in this
respect, Armstrong expects too much from
science.
D. The situation is remarkably close to Leibniz'
description. Even though we may inspect the
brain more thoroughly them Leibniz, there is
still nothing cibout mental activity that
points to the brain nor is there anything
about the brain that points to mental activ
ity. Without responses by subjects, there
would never be correlations drawn between the
two. Further, the oddity of inferring B from
A though A and B are identical is noted.
E. In addition, it is pointed out, as Jaegwon Kim
argues, there is no set of experiments that
would establish identity without also estab
lishing a correlation.
Section 3
A. There is a third set of considerations center
ing on introspective awareness and its
objects. What, if anything, are we directly
aware of when we are directly aware of our
mental states? Some think that it is at
this point that Armstrong's identity claim
breaks down.
B. Those who think so argue from the substitutivity
of identicals. Since mental states are identi
cal with brain states, it would follow that we
must be directly aware of our brains wherever
we were directly aware of some mental state.
xi
135
135-
136
136-
138
138-
141
141-
142
142
But, it is argued, this conclusion is not
true. Thus, Armstrong's identity claim is
false.
C. Armstrong might reply to this charge in one
or two ways. First, he might accept the
implication. That is to say, he might argue
that we are directly aware of the brain. We
just are not aware of it as the brain. He
could employ here his topic neutral analysis
of mental states.
D. Second, he might reject the implication by
arguing that "is aware of" generates opaque
contexts. From one's awareness of m, and
m's identity with b, it does not follow that
one must be aware of b.
E. Summation of Chapter V.
CHAPTER VI
CONCLUSION: SUMMARY OF THE ARGUMENT
Section 1
A. Summation of the key points of the essay.
Pages
143
144-
146
146-
151
151-
153
154-
158
xii
CHAPTER I
INTRODUCTION: THE TASK OF THIS ESSAY
1
The mind-body problem, with which this essay deals, is
generated from two of the most fundamental questions in
philosophy, questions which G. E. Moore thinks are at the
heart of the philosophical inquiry and which might be
expressed as: (1) What kinds of things are there in the
universe? and (2) In what relationship do these things stand
to one another?1 Clearly, these two questions stand at the
core of the mind-body question since the controversy over
minds and bodies centers on whether such things as a mind or
a body exist, and, if they do, what the relationship between
the two is. And, though there might be some questions of
existence and relation which might be approached disinter
estedly as matters of simple curiosity about the ultimate
constituents of the universe, this surely is not the case
where minds and bodies are concerned. Once understood, the
questions surrounding these two entities command as much
genuine human interest and importance, to most reflective
persons, as any question discussed in philosophy. This is
1G. E. Moore, Some Main Problems of Philosophy (New
York: The MacMillan Company), p. 1.
1
2
due, partially, to our fascination with ourselves and our
natural inquisitiveness about what kinds of things we are,
but also to the many related and important issues affected
by the outcome and resolution of the mind-body problem.
One possible outcome is offered by D. M. Armstrong in
his book, A Materialistic Theory of the Mind, and his spe
cific solution is the subject matter of this paper. It is a
feature of his solution that it is arrived at by answering
Moore's second question first. By giving the kind of answer
he does give to the second question, Armstrong ipso facto
gives an answer to the first. By arguing that the relation
ship between mind and body is one of strict identity, Arm
strong in effect gives a partial answer to Moore's first
question, even though his answer is negative. Armstrong
simply denies that there are minds, if minds are taken to be
non-physical substances serving as the seat of consciousness
and all mental activity. He does this by arguing that minds
just are brains, or that mental states just are brain
states. (He does not consider that if minds and brains are
identical, then a "mentalist" position might be just as
plausible as a "physicalist" position.)
But this is not all Armstrong attempts to do in A Mate
rialist Theory of the Mind. In addition to answering
Moore's two questions, Armstrong thinks it necessary to dis
prove competing theories of mind, the most important of
which he takes to be Dualist theories. Thus, the scope of
his book includes not only a positive aspect, but a negative
one as well. Not only does he offer a possible solution to
the mind-body problem— his Central State Theory, or Central
State Materialism— but he argues against this theory's chief
rival. Consequently, Armstrong splits his efforts between
arguing against Dualism (mainly in its Cartesian variety)
and arguing for his version of the Identity Thesis. This
strategy provides a convenient format for this paper as
well, which, accordingly is divided into two parts. The
first half will concentrate on Armstrong’s arguments against
Dualism while the second half will consider some of the cru
cial claims of his Central State Theory.
Needless to say, with a topic so variagated, much of
genuine philosophical interest, will of necessity, be
omitted, though hopefully, what is included is of more
importance than what is left untouched. For instance, Arm
strong discusses "The Will," "Knowledge and Inference,"
"Perception and Belief," etc. These topics will be only
considered indirectly, or not at all. The justification for
such omission resides in the fact that they deal only with
implications from Armstrong's work on the Central State
Theory: their truth or plausibility is largely a function
of the truth or plausibility of the latter. Hence, the
important work needs to be done on the Central State Theory
itself.
Hopefully, what will emerge from what is discussed here
4
will be twofold: (1) an analysis of Armstrong's arguments
against Dualism with the intent to show that these arguments
do not imply the falsity or even the "unsatisfactoriness" of
Dualism, and (2) a review of Armstrong's arguments in favor
of his version of the Central State Theory with the aim to
show that if that theory is true, it is not true for the
reasons Armstrong advances. It is the thesis of this essay
that Armstrong succeeds neither in rendering Dualism
implausible, nor in rendering his version of materialistic
monism plausible.
2
Before directly considering Armstrong's work, it might
be helpful briefly to survey some of the classic treatments
of the mind-body problem to see where Armstrong fits into
the broad spectrum of views on the topic. These views
split, it seems, over answers to at least three questions.
First, is there a legitimate distinction between minds and
bodies? Second, if there is a legitimate distinction, is
there anything that exists which corresponds to either or
both of these terms? Third, if entities are admitted which
correspond to each term, what is the relationship between
them? These questions do not exhaust what may be asked
about this issue, but they provide an easy framework for
this brief history, which does not intend to be either
extensively critical or complete. Our aim is simply to give
a rough backdrop against which Armstrong's position might be
seen more clearly.
Depending on the answers given to the above question,
views on this topic roughly divide into two broad cate
gories: dualism and monism. As the terms imply, the divid
ing issue between these positions is over the number of
entities each is to allow. One may argue that both minds
and bodies exist— dualism— or that just minds or just bodies
exist but not both— monism. The monist view that only
bodies, or the physical, exists is further classified as
materialism, while the view that only minds exist is classi
fied as materialism, while the view that only minds exist is
classified as idealism. Dualism, in turn, is divided into
five major variants, which we now consider.
Dualism's first modern defender is Rene Descartes who
held not only that minds and bodies are distinct substances,
but that they causally interact with one another. There is
little question that Descartes thought that the distinction
between minds and bodies was legitimate, and that there were
two entities corresponding to the two terms. As to the
nature of mind and body, Descartes first defines body in
terms of extension.
By "body" I understand all that can be bounded by some
figure; that can be located in some place and occupy space
in such a way that every other body is excluded from it;
that can be perceived by touch or sight or hearing or
taste or smell; that can be moved in various ways, not by
itself but by some other object by which it is touched and
from which it receives an impulse.2
And, taking thought to be the essence of mind, and the
essence of thought to be non-spatial, Descartes draws a
clear distinction between mind and body.
In the second, the mind, which in its intrinsic freedom
supposes that everything which is open to the least doubt
is nonexistent, recognizes that it is nevertheless abso
lutely impossible that it does not itself exist. This is
also of the highest utility, since by this means the mind
can easily distinguish between those qualities which
belong to it— that is to say, to its intellectual nature—
and those which belong to the body.*
Obviously, by drawing such a clear and definite dis
tinction between the mental and the physical, Descartes must
have given an account of the relationship between the two.
That there is a relation, indeed a very close one, Descartes
did not doubt. He took the view that many philosophers
think closest to common sense: he held that mind and body
causally interact with one another. Descartes thought that,
under normal conditions when I decide to move my arm (mental
event), my decision causes my arm to move (physical event).
And conversely, when I stub my toe (physical event), the
disturbance of tissue causes me to feel a sharp, throbbing
pain (mental event). These and like "facts" Descartes took
at face value. The mind and the body causally interact,
1Descartes, Discourse on Method and Meditations, trans
lated by Laurence J. Lafleur (The Library of Liberal Arts;
Indianapolis, New York: The Bobbs-Merrill Company, Inc.,
1960), p. 83.
2Ibid., p. 71.
hence such a position is called dualist interactionism.
Naturally, question immediately arises about how,
exactly, something non-physical can causally affect some
thing physical and conversely, how can something physical be
causally related to something non-physical. How is it that
such radically different entities stand in a causal relation
with one another?
Descartes' apparent solution was to posit a small
gland, the "pineal" gland, midway in the brain which serves
as the causal mediator between mind and brain.
But, in examining the matter with care, it seems as though
I had clearly ascertained that the part of the body in
which the soul exercises its functions immediately is in
nowise the heart, nor the whole of the brain, but merely
the most inward of all its parts, to wit, a certain very
small gland which is situated in the middle of its sub
stance and so suspended above the duct whereby the animal
spirits in its anterior cavities have communication with
those in the posterior, that the slightest movements which
take place in it may alter very greatly the course of
these spirits; and reciprocally that the smallest changes
which occur in the course of the spirits may do much to
change the movements of this gland.1 *
Thus, we can see that Descartes clearly thought that
the distinction between mind and body was legitimate and
that the terms involved referred to two distinct yet caus
ally related entities. Whether this is the correct analysis
of the issue or not, it is clear, from contributions subse
quent to Descartes, that he established the framework within
which the problem has been discussed. Dualists reacted
l *Descartes, Passions of the Soul, as found in Introduc
tory Philosophy, Edited by Tillman, Berofsky & O'Connor; 2nd
ed. (New York: Harper and Row, 1971), p. 21.
8
either against Descartes' notions of mind or body or against
his assessment of the relation between them.
For instance, Malebrance argues, in effect, that
Descartes had so defined mind and body that it was impos
sible for them to stand in a causal relation with one
another. They were so radically distinct for Descartes, and
for Malebrance as well, that it took God to "occasion" the
apparent interaction between them. Contrary to appearances,
these two substances do not causally interact according to
Malebrance. Its just that when one decides to move one's
foot, that decision is the "occasion” for God to cause one's
foot to move. God, then, becomes the causal link which
relates mind to body. Of course, God also "occasions" the
mental events that are apparently caused by previous physi
cal events. When I stub my toe, this physical event is the
"occasion" for God to cause me to feel pain.
But, at least, it is evident that a body, that extension,
a purely passive substance, cannot operate by its own
activity upon a mind, upon a being of another nature and
infinitely more excellent than it. Thus, it is clear
that, in the union of soul and body, there is no other
bond than the efficacy of divine and immutable decrees,
and efficacy never without its effects. God has then
willed, and wills without ceasing, that the various dis
turbances of the brain shall always be followed by the
various thoughts of the mind with which it is in union.
And it is this constant and efficient will of the Creator
which causes the union of these two substances.5
sNicolas Malebrance, Dialogues on Metaphysics and on
Religion, as found in The Philosophy of the loth and 17th
Centuries, Richard Popkin, ed; Series editors Richard Popkin
and PaulEdwards (New York: The Free Press; London: Col-
lier-MacMillan), p. 295.
One can, of course, do away with having to explain the
causal relationship between radically different substances
simply by denying that such a relationship exists. If this
can be carried out, then not only will one be able to main
tain the existence of two substances, but the problem of
interaction will be solved. No longer would have to bring
in a third entity to act as the causal mediator. Nor will
one have to subsequently explain how this third entity (God
or the pineal gland) could produce effects radically differ
ent from itself. One such attempt is called parallelism,
since it views minds and bodies as both fundamentally real
but with no causal interaction between them. They simply
run like trains on parallel tracks. They each form causal
sequences within themselves, but they are causally distinct
from each other.
Leibniz, for example, developed a variant of this view.
The apparent reason he opts for such a position is that he
%
despaires of understanding how mind and body could causally
interact
After establishing these things, I thought I had arrived
in port; but when I began to meditate on the union of the
soul with the body, I was cast back, as it were, into the
open sea. For I found no way of explaining how the body
causes something to happen in the soul, or vice versa; nor
how a substance can communicate with another created
substance.6
6Gottfried Wilhelm von Leibniz, New System of Nature
and of the Communication of Substance, as Well as of the
Union of Soul and Body, The Philosophy of the 16th and 17th
Centuries, Richard Popkin, ed., p. 328.
10
But Leibniz also rejects the notion of Malebrance that
God constantly intervened and created "miracles" whenever
two objects causally affect one another.
. . . but in order to solve problems it is not enough to
employ the general cause and to invoke what is called Deus
ex Machina. For when that is done without any other
explanation drawn from the order of secondary causes,
recourse is being taken to miracle properly speaking.7
Consequently, Leibniz argues that God so ordered and
synchronized "souls" and bodies that they perfectly re
flected each other's activity. What happens in one's mind
perfectly "conforms" to activity in the external world sim
ply because God has so arranged the history of each to per
fectly coincide in a "pre-arranged harmony."
Therefore, though I was obliged to agree that it is impos
sible for the soul, or any other true substance, to
receive any influence from the outside except through
divine omnipotence, I was gradually led to a thought which
surprised me but seems to me inevitable and indeed has
very great advantages and a very considerable attraction.
That is, we must say that God has from the first created
the soul or any other real unity in such a way that every
thing arises in it from its own internal nature through a
perfect spontaneity relatively to itself, and yet with a
perfect conformity to external things. Thus our internal
thought, that is, those in the soul itself and not in the
brain nor in the subtle parts of the body . . . , these
perceptions internal to the soul itself, must happen to it
through its own original constitution, that is to say,
through its representative nature . . . given to it since
its creation and constituting its individual character.
And that is what makes each one of these substances repre
sent, each exactly in its own way, the whole universe from
a certain point of view . . . There will be a perfect har
mony among all these substances which produces the same
effect that would be noticed if they communicated mutually
through that propagation of species or of qualities imag
ined by the common run of philosophers.*
7Ibid., p. 329. 'ibid.
I
11
Another attempt to eliminate the inherited problem of
causal interaction between mind and body is to so analyze
mind and body that they simply become two "aspects" of the
same thing. For instance, Spinoza held such a view. For
him, the mental and the physical are two "aspects" of some
more primitive underlying reality. Thus, man may be consid
ered alternatively as an extended thing or as a thinking
thing, though these different "descriptions" are nothing but
two different manners of conceiving the same thing. Hence,
there is no question of two separate substances and the
ensuing relation between them.
Before we go any farther, we must here recall to our mem
ory what we have already demonstrated, that everything
which can be perceived by the infinite intellect as con
stituting the essence of substance pertains entirely to
the one sole substance only, and consequently that sub
stance thinking and substance extended are one and the
same substance, which is now comprehended under this
attribute and now under that.9
Of course, questions naturally arose about both the
nature of the underlying "stuff" and the two "aspects" as
well. Needless to say, this "double aspect" theory has much
in common with many modern, linguistic, materialistic treat
ments of the issue. Under these views, mental language and
physical language are merely two manners of talking about
the same subject, the brain. Another version of this view
has it that the different "aspects" are simply "ways of
9Benedictus De Spinoza, Ethics, as found in Spinoza
Selections, ed. by John Wild (New York: Charles Scribner's
Sons, 1$30), p. 149.
12
appearing," as if we were seeing the same object from dif
ferent points of view or vantage points.
There is at least one other major attempt to resolve
the problems Descartes bequeathed, and that is epiphenomen-
alism. This point of view regates the mental to a second
ary, derived status. The mental is merely a by-product, an
"epiphenomena" of the more basic underlying reality, the
physical. Mental activity is merely something incidental to
the organism in which it occurs, and has no causal influence
on it. Thus, physical events may cause mental events but
not vice versa; there is only one way— causation.
Further, thoughts do not even cause other thoughts;
that would be superfluous under this view. Each mental
state results directly from some corresponding brain state.
Nor do mental events cause or produce bodily actions, even
though it might appear that they do. This again would be
unnecessary since each physical state is thought to be
caused solely by some preceding physical state. Thus, the
mental is without causal efficacy of any kind. As T. H.
Huxley, the central modern defender of this view writes,
Consciousness would appear to be related to the mechanism
of the body simply as a collateral product of its working,
and to be as completely without any power of modifying
that working as the steam-whistle which accompanies the
work of a locomotive engine is without influence upon its
machinery.10
19T. H. Huxley, "On the Hypothesis That Animals Are
Automata and Its History," as found in A Modern Introduction
to Philosophy, edited by Paul Edwards and Arthur Pap; 3rd
ed. (New York: The Free Press, 1973), p. 228.
13
Thus, according to epiphenomenalism, the mental is
wholly an inactive accompaniment of a completely enclosed
causal system. In this view, the mental is always an effect
and never a cause, and it is an effect which costs the phys
ical nothing.
All states of consciousness . . . are immediately caused
by molecular changes of the brain substance . . . There is
no proof that any state of consciousness is the cause of
change in the motion of the matter of the organism . . .
It follows that our mental conditions are simply the sym
bols in consciousness of the changes which take place
automatically in the organism; and that, to take an ex
treme illustration, the feeling we call volition is not
the cause of a voluntary act, but the symbol of that state
of the brain which is the immediate cause of that act. We
are conscious automata.11
We can see, then, if not in detail at least in purpose,
that the dualists have struggled over relating the mental to
the physical in a plausible way. Since they clearly see a
legitimate difference between the two, either in terms of
substances or events, etc., they are faced with showing how
they may be intelligibly related. The monists do not have
to face such a serious problem in this respect. Since they
deny that the distinction between mind and body necessarily
implies distinct entities, they are not forced to describe a
real relation between them.
For instance, George Berkeley argues that minds only
exist, and matter is reducible to the perception of some
mind. All that exists are minds and their "ideas." There
are no physical things, no bodies that exist.
xlIbid., p. 229.
14
As far as physical objects are concerned, independent
of some mind. To be is simply to be perceived.
That neither our thoughts, nor passions, nor ideas formed
by the imagination, exist without the mind, is what every
body will allow. And it seems evident that the various
sensations or ideas imprinted on the sense, however blend
ed or combined together (that is, whatever objects they
compose), cannot exist otherwise than in a mind perceiving
them. I think an intuitive knowledge may be obtained of
this, by any one who shall attend to what is meant by the
term exist, when applied to sensible things. The table I
write on, I say, exists; that is, I see and feel it; and
if I were out of my study I should say it existed, meaning
thereby that if I was in my study I might perceive it, or
that some other spirit actually does perceive it. There
was an odour, that is, it was smelled; there was a sound,
that is, it was heard; a colour or figure, and it was per
ceived by sight or touch. This is all that I can under
stand by these and the like expressions. For as to what
is said of the absolute existence of unthinking things
without any relation to their being perceived, that seems
perfectly unintelligible. Their esse is percipi, nor is
it possible they should have any existence, out of the
minds of thinking things which perceive them.12
One of the obvious advantages of this view, an advan
tage shared by materialism, is that if it is true, there is
no need to give an analysis of the relation between two
separate and distinct entities. But despite this advantage,
Berkeley's view has been challenged on the grounds that
statements about what is occurring in the next room, or
about what exists in the next room, cannot be translated
into statements about what God perceives.
Materialism, the other monism, holds in general that
only the physical exists. There is no legitimate distinc-
l2George Berkeley, On the Principles of Human Knowl
edge, as found in Philosophic Classics, Vol. II; 2d ed.
(Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1968),
p. 248.
tion, in the sense of different substances, events, pro
cesses, etc., between the mental and the physical.
Thoughts, wishes, sensations, paradigm cases of mental
events do occur; its just that they are physical in nature.
There is no radical break in "nature" between them. Need
less to say, even though this theory "simplifies" the prob
lem by reducing the entities to be considered by one, the
theory itself is by no means simple. Jerome Shaffer1* cites
four major variants of the thesis that have been developed
in the twentieth century, not all of which are compatible
with one another. It is with one of these variants, the
identity theory, that Armstrong defends in his book, A Mate
rialistic Theory of Mind and which is the subject of this
paper, to which we now turn.
1 *Jerome Shaffer, Philosophy of Mind (Englewood Cliffs,
New Jersey: Prentice-Hall, Inc., 1968), p. 46.
CHAPTER II
A CRITIQUE OF ARMSTRONG’S CRITIQUE OF DUALISM
1
D. M. Armstrong has a twofold purpose in writing A
Materialist Theory of Mind: (1) he aims to discredit com
peting theories of mind, and (2) he aims to present a com
pelling case for his brand of Materialism. The first under
taking, which takes roughly the first third of his book, is
dedicated to criticizing theories of mind standing in con
tradiction to Materialism. And, since he thinks that, in
our times, Dualist theories provide the only "important
alternatives”1 to Materialism, Armstrong limits his criti
cisms to various forms of the Dualist position. If he is
successful in his criticism, he thinks the road will then be
clear to present his view, the view "on the right track,"2
namely, Central State Materialism. This latter effort forms
the "positive" thrust of the book and its end is the accom
plishment of (2) above. It should be noted that neither in
the first section, where he argues against Dualism, nor in
the second section, where he argues for Central State Mate
rialism, does Armstrong openly argue for the truth or
XD. M. Armstrong, A Materialist Theory of the Mind
(London: Roatledge & Kegan Paul, 1969; New York: Humani
ties Press), p. 2.
2Ibid.
17
falsity of either position. The truth or falsity of his
position, and by implication, the truth or falsity of Dual
ism, will be decided by those "scientists . . . most
directly concerned with the problem, viz., the psychologists
. . . "3 Armstrong sees himself as simply silencing some
"philosophical" objections that might be made against what
he thinks most neurophysiologists are confident of: the
identity of mind and body.* As Armstrong puts it,5 neuro
physiologists might be slightly "harassed" by some philo
sophical objections to identifying mind with brain, and his
job is to aid the neurophysiologists by "carting away the
rubbish"6 that makes up these objections.
One might describe, then, Armstrong's work as concep
tual or logical, in the sense that he wants to dispel the
conviction, held by some, that there might be logical or
conceptual inconsistencies in his Central State Theory.
Once such convictions are put to rest, Armstrong thinks the
scientist will be free to go about discontinuing or confirm
ing his theory, just as the scientist does with any other
testable hypothesis. According to Armstrong, it is not his,
nor any other philosopher's task to discover the truth about
the mind-body relation; all the philosopher can do is deter
mine whether there are any logical or conceptual reasons for
rejecting a proposed analysis of that reaction. (Though if
this is a correct reading of Armstrong's position, it would
3Ibid., p. 74. I > lbid., p. 2. 5Ibid.
seem that we would sometimes have good reasons for thinking
some position false, for example, if it contained logical
inconsistencies. There would be room, then, for the philos
opher to rule out as false, some proposed analyses of the
mind-body problem.) Hence, one will not find, in either of
the book's two major divisions, an argument which has as its
conclusion that "Dualism (Materialism) is therefore false
(true)." What Armstrong substitutes for "true" or "false"
is either "implausible," "unsatisfactory," "untenable," or
some close variant. It is my aim in this chapter to examine
Armstrong's five major arguments, which have as their con
clusion, "Dualism is therefore implausible," or some close
cognate thereof. Thus, this chapter will be limited to
those arguments Armstrong utilizes in accomplishing the
first aim of the book. (The five arguments to be examined
are stated below on pages 23, 29, 53, 71, and 74 respec
tively.)
My criticisms of these arguments will generally take
one of four forms: (1) I will argue that Armstrong utilizes
a questionable epistemological principle which precludes the
Dualist alternative, (2) that Armstrong sometimes demands of
Dualism what is not even provided by his Central State The
ory, (3) that some of Armstrong's arguments are question-
begging, and (4) that Armstrong often depicts mental sub
stance, mental properties, or mental relations in such a way
that they turn out to be physical entities. What I hope to
show by such criticisms is not that Dualism is true or that
19
Armstrong's version of Materialism is false, but that Arm
strong has not accomplished what he set out to do under his
first aim: he has not "discredited" the Dualist position.
No attempt will be made here to demonstrate the truth or
even the "plausibility" of Dualism. The aim is simply to
discover whether Armstrong offers any sound arguments
against Dualism, and to accomplish this, I will look at the
five arguments Armstrong employs against Dualism, following
his order of presentation, and beginning with his argument
against the Cartesian form of Dualism.
2
Armstrong begins his section on Cartesian Dualism by
stating:
Nevertheless, there is something curiously formal and
empty about the Cartesian solution. What is this spiri
tual substance which has these mental affections? One
seems able to say nothing positive about it except that it
has the mental affections. One can say negative things
about it, that it lacks spatial properties, for instance,
but its positive nature remains a mystery. By contrast,
we can easily describe faces apart from their grins, or
physical substances apart from their soporific virtues.
Spiritual substance looks like something invented to be
the solution of Dualist problems.7
Further on Armstrong concludes,
Cartesian Dualism, then, may have formal advantages over
"Bundle Dualism," but it is a doctrine of a very empty
sort.8
(It should be noted that even though Armstrong uses "Carte
sian," he is not strictly referring to the view as Descartes
7Ibid.. p. 23. 8Ibid.. p. 24.
20
developed it. He has in mind "any view that holds that a
person's mind is a single, continuing, non-material sub
stance in some way related to the body."* So, it is against
this view that Armstrong's first argument is directed. And
this is Armstrong's central, if not his only, argument spe
cifically addressed to the Cartesian position. He does have
arguments which he thinks apply to any form of Dualism, but
this argument serves specifically to "disprove" the Carte
sian variety as he describes it.)
As indicated in the two quotations, Armstrong intends
to conclude two points from his first argument: (1) that
the notion of mental substance is "formal and empty," and
(2) that the notion of mental substance looks to be "in
vented to be the solution of Dualist problems." Since Arm
strong does not put his argument in a formal manner, one
might dispute whether these are the conclusions he has in
mind. I justify my selection by pointing to what appear to
be reasons given for them in the passages, and also by show
ing these conclusions to be consistent with what Armstrong
argues for in other sections of the book. Finally, what I
attempt to show regarding this first argument is that Arm
strong grounds the argument in an epistemological principle
that begs the issue as to what minds might be. In addition,
I will argue that the criticisms Armstrong levels against
Cartesian mental substance apply as well to his notion of
*Ibid., p. 6.
21
physical substance.
Textual evidence supports my reading of this first
argument. For instance, the first sentence quoted above—
"Nevertheless, there is something curiously formal and empty
about the Cartesian solution"— seems to stand in contrast
and be in reaction to whatever advantages such a solution
might have. This interpretation is supported both by his
introductory word, "nevertheless," and by the content of the
paragraph immediately preceding the one quoted. In the pre
ceding paragraph, Armstrong cites what might be considered
the advantages of Cartesian Dualism, for example, it circum
vents the "logical difficulties" of a Humean "Bundle" theory
by providing a something, a "background" which has the men
tal experiences. Then, after noting these advantages, Arm
strong launches into what he thinks are the obvious draw
backs of the Cartesian position and thus, the opening sen
tence, beginning as it does with "nevertheless," takes on
the appearance of something he would like to conclude or
convince his readers about. He seems to be saying, "even
though it might appear that Cartesian Dualism has logical
advantages over some forms of Dualism, it is a thoroughly
empty and formal theory." If this reading is a natural one,
then not only does Armstrong's conclusion become clear, but
the sentences immediately following fall into place as Arm
strong's reasons for charging the Cartesian position with
being contrived and vacuous.
The first of these reasons also constitutes Armstrong's
main complaint against Cartesian Dualism, and it is given in
various forms throughout the book. The thrust of this com
plaint is that we know nothing about this mental substance
which Descartes posits to be the repository of all mental
properties. Thus, Armstrong writes in sentences three and
four of the quoted passage that we can "say" only negative
"things" about mental substance and consequently, its "posi
tive nature remains a mystery." I interpret Armstrong as
meaning by "say" that we can never "know" anything about the
"positive nature" of mental substance. That is, in this
section at least, Armstrong is using "saying" for "knowing."
What he probably has in mind by "say" is "rightfully say,"
or "be in a position to say," or "justified in saying," each
being an equivalent of "know." My reasons for this inter
pretation are twofold: first, Armstrong offers no argument
to show that knowing and saying are identical activities,
and second, knowing and saying seem clearly to be separate,
though many times related, activities. Thus, what I take
Armstrong to be questioning is whether we can have any
knowledge about the "positive nature” of mental substance.
Now sentences three and four might stand as conclusions
for some other arguments Armstrong has in mind, but in this
instance I think he wants them to serve as reasons for con
cluding that Cartesian Dualism is both "formal and empty."
The emptiness would stem from our inability to say (know)
23
anything "positive" about mental substance, and the formal
ity would derive from Armstrong's conviction that mental
substance is purely an arbitrary solution to a thorny prob
lem. It does not seem that Armstrong is arguing from the
emptiness and formality of the Cartesian theory to our lack
of knowledge about mental substance, but from our lack of
knowledge about mental substance to its emptiness and
formality as an explanation. To buttress this conclusion,
Armstrong refers to physical substance about which he thinks
we have "positive" knowledge. "By contrast, we can easily
describe faces apart from their grins, or physical sub
stances apart from their soporific virtues."18 That is,
our knowledge about physical substance prevents it from be
ing an "empty" posit, and this is seen in direct contrast
with what he wants to conclude about mental substance. The
argument, then, is straightforward. Armstrong wants to con
clude that the Cartesian theory is "formal and empty," and
this is supposed to follow from our lack of "positive"
knowledge about Cartesian mental substance. We might formu
late Armstrong's argument in this manner.
Argument #1
1. All theories about an entity about which we know
nothing positive are empty and formal theories.
2. Cartesian Dualism is a theory about an entity about
which we know nothing positive.
Therefore, Cartesian Dualism is an empty and formal
24
theory.
As formulated the argument is valid, but the question
remains whether there is any reason to believe that the
premises are true. And, before this latter question can be
answered, one must be clear about the meanings of the key
terms. One would expect Armstrong to do the clarifying;
unfortunately however, he never mentions what he means by
"positive knowledge," "emptiness," or "formality." Conse
quently, the clarifying activity is left to the reader.
Thus, my strategy will be to begin clarifying what Armstrong
might mean by "positive knowledge." This, in turn, will
lead to a discussion concerning Armstrong's fundamental
epistemological principle beclouded in his term "concrete
acquaintance." I will argue that by "concrete acquaintance"
Armstrong means nothing more than sense experience such that
only if we have had sense experience of something may we
have "positive knowledge" about it. Finally, I will try to
show that if the foregoing is true, then Armstrong's episte-
mological principle runs counter to both his claims about
physical substance and, derivatively, his claims about self-
knowledge .
3
There is a trivial interpretation of "positive" which,
when combined with "knowledge," yields a somewhat redundant
phrase. When one is "positive" about something, in this
sense, one is fully assured or confident; and many times
25
"positive" is substituted for "know," e.g., when one says
"I am positive he is the one who did it." This could indi
cate a close relationship between "positive" and "know" such
that if this is the sense of "positive" which Armstrong has
in mind, then to say we have "positive" knowledge about
physical substance, or that we lack such knowledge about
mental substance would only be a way of saying: "We have
knowledge about physical substance, whereas we do not have
any knowledge about mental substance." Under this interpre
tation "positive" has no peculiar force and simply drops out
of the translation since it is so close in meaning to
"know." However, this trivial sense does not have to be
what Armstrong has in mind by "positive"; and, from what he
says in this and other passages, I think it is not.
What Armstrong says here, and reinforces in other sec
tions, is that we are unable to "say anything positive"
about mental substance whereas we are able to say some posi
tive "things"- about physical substance. On this point one
can look again at the third and fourth sentences of the
quoted paragraph, or one can read his section on the indi
viduation of mental objects. In that section Armstrong
writes, "The inability of the Dualist to say anything at all
about his spiritual principle of individuation strongly sug
gests that the only way he can understand his notion is a
negative way as a 'non-spatial principle of individ-
26
uation.'"11 The contrast here is also between either saying
something "positive" or saying something "negative" about
some kind of entity. It seems that what Armstrong has in
mind are two classes of knowledge claims, those expressed
"positively," for instance, "I know X is Y," and those
expressed "negatively," as when one says, "I know X is not
Y," or "I know X is non-Y." For Armstrong, the first is the
privileged knowledge claim. It has a higher status than the
latter, and it is this latter kind of cognitive claim that
would lead someone to conclude, according to Armstrong, that
any theory wholly based on such claims is both "formal and
empty."
But his claim is yet more precise than this. He wants
to argue that for any object, X, if all we can know about X
is that it is not Y, and that it is not Z, etc., then we
really do not know anything about the "nature" of X. We do
not know anything about what X is, only what it is not.
Compared to some other object, S, about which we know that
it is A and that it is B, we can see that our knowledge
about the two objects is not on a par. We know something
about what S is but we only know what S is not. For
instance, if I had an object and you wanted to know what it
was and I told you, "It is not green," what would you know
about it? By contrast, if I were to tell you, "It is red,"
you would have some knowledge about it; you would know some-
11Ibid., p. 29.
thing about what it is, not what it is not. Of course, one
could point out that "It is red" can be translated without
less of truth value into "It is not non-red," but this would
not alter Armstrong's point. The crucial case, he would
argue, is that "It is not green" cannot be translated into
some positive equivalent. What we would get is "It is non
green," and Armstrong's point would remain intact: We would
still know nothing about what the object is, for "non-green"
refers to everything in the universe other than being green.
Be this as it may, it does not seem that simply because
a proposition is expressed in a negative sentence the propo
sition is not expressing some bit of knowledge, e.g., "No
physical object can be in two places at the same time," "No
person with more than five citations will be issued a
driver's license," or "Numbers are not in space," etc.
There are countless propositions of this kind from all areas
of life which, even though expressed negatively, would still
be considered as expressing important bits of knowledge.
This leads me to believe that this is not all that Armstrong
means when he talks about not having any "positive" knowl
edge about mental substance, while at the same time having
"positive knowledge" about physical substance. And a hint
about what else he could mean is given when he talks about a
principle of individuation for "spiritual" objects.
Nevertheless, it does seem likely that the only "prin
ciple of individuation" with which we have any concrete
acquaintance is that of being in different times and
places. The inability of the Dualist to say anything at
all about his spiritual "principle of individuation"
strongly suggests that the only way he can understand his
notion is a negative way as a "non-spatial principle of
individuation." It seems that he is not in fact ac
quainted with such a principle, nor is there any reason to
postulate it.12
For our purposes, that is for determining what Arm
strong means by "positive knowledge," the key term is "con
crete acquaintance"; for Armstrong's inference here seems to
be that because we have no "concrete acquaintance" with a
principle of individuation for spiritual objects, we can
only say what they are not, never what they are. It would
seem that the third sentence above is given as a reason for
concluding that a Dualist can never say anything "positive"
about the individuating principle for "spiritual" objects.
If the Dualist could be "concretely acquainted" with such a
principle, then it would follow that he could talk about the
"nature" of the principle and it would thus no longer
remain, as does mental substance, a "mystery." Implicit,
then, in Armstrong's first argument is a connection between
"concrete acquaintance" and "positive knowledge," such that
without the former we would never have the latter. The one
stands at least as a necessary condition for the other, and
this is vital to Armstrong's criticisms.
This reliance of "positive knowledge" on "concrete
acquaintance" fills in, I think, the missing component in
Armstrong's conception of "positive knowledge." He is not
12Ibid., p. 29.
29
simply pointing out that "positive knowledge" is expressed
in affirmative sentences, but more importantly that such
knowledge claims are supported by "concrete acquaintance"
concerning whatever the assertion is about. We can formu
late then a general line of reasoning which supports Arm
strong's first argument.
Argument #2
1. All entities of which we have no concrete acquaint
ance are entities about which we have no positive knowledge.
2. The principle of individuation for spiritual ob
jects (mental substance) is an entity about which we have no
concrete acquaintance.
Therefore, the principle of individuation for spiritual
objects (mental substance) is an entity about which we have
no positive knowledge.
The conclusion of this argument would then form the
first premise of Argument #1, with the conclusion of the
then-formulated soritie being that Cartesian Dualism is both
empty and formal. (My suspicion is that another inference
is involved here, to the effect that whatever we have no
"positive knowledge" of, or whatever we are not "concretely
acquainted" with, fails to exist. But such an inference
would bring with it questions of the truth or falsity of
Dualism and the Central State Theory, and Armstrong is not
prepared to discuss that issue yet.) Nevertheless, before
we discuss the conclusion of Argument #2, some comments are
30
in order about "concrete acquaintance," as it is used in the
premises. Hopefully, such a discussion will aid us in
becoming clear about the meaning of "positive knowledge" as
Armstrong uses the term.
4
Armstrong gives no clear clues about the meaning of
"concrete acquaintance," so, we are left to conjecture; and
one of the best methods of conjecture is to see what would
be consistent with what he says elsewhere. Such an examina
tion leaves one with the conviction that "concrete acquaint
ance" is closely related to, if not identical in meaning
with, sense experience. There are at least two reasons for
this conclusion: First, this reading would be consistent
with Armstrong's Materialism. Second, Armstrong uses sense
experience language to summarize the section wherein he
first introduces the term "concrete acquaintance."
As to the first reason, Armstrong believes that if any
thing exists it must exist in time and space, and by most
accounts, the only established way of becoming aware of the
objects which exist in time and space is through our senses.
It seems that if there are only objects that take up space,
then the only kind of "acquaintance" we could have of them
would have to be mediated through our senses. Of course,
Armstrong might have some other method in mind, but if he
does he has failed to mention it. To be consistent with his
Materialism then, "concrete acquaintance" could only mean
31
acquaintance by and through the senses. If this reading is
correct, Armstrong is using some Lockean/Humean type princi
ple to the effect that for every "idea" there must in some
way be an anterior sense "impression."
Second, this interpretation is bolstered by what Arm
strong writes at the conclusion of Chapter Two. There he
summarizes what he considers problems for any Dualist the
ory.
If the Cartesian is prepared to postulate, or claim to
observe, a spiritual substance; if he is prepared to pos-
tulate, or claim to observe, a unique relationship that
ties particular spiritual substances to particular bodies
. . . (italics mine)13
Here Armstrong stipulates that a Dualist of the Cartesian
bent must be at least prepared to discuss the mental sub
stance he "observes" along with the relationship this sub
stance has to some body. He also requires this same Dualist
to be able to discuss the principle of differentiation among
"spiritual" objects in terms of what he "observes" about
this principle. This seems to support my interpretation of
"concrete acquaintance" as meaning some kind of sense expe
rience, since both "observe" and "acquaint" are usually used
to refer to some kind of sensory activity. It is true that
"acquaintance" has other uses and meanings, but in this con
text it is apparent that Armstrong is using it in a cogni
tive sense; and, coupled with the fact that for Armstrong
there are only physical objects, events, processes, etc.,
13Ibid., p. 36.
32
what else could "acquaintance" mean but sense experience?
Acquaintance I take to mean being presented with something
in such a manner that one is directly, non-inferentially
aware of whatever is being presented. One is reminded of
Russell's definition of acquaintance when he spoke of the
distinction between knowledge by description and knowledge
by acquaintance: "We shall say that we have acquaintance
with anything of which we are directly aware, without the
intermediary of any process of inference or any knowledge of
truths."11* Russell goes on to say that among the things
given in acquaintance are the sense data which "make-up"
physical objects, and which are given "just as they are."
(We also have acquaintance, according to Russell, of our own
thoughts, awarenesses, etc., but this is not an issue here.)
"Acquaintance," then, has a history of being used with rela
tion to something being presented to the senses, and so does
"observe." "Observe" is defined in terms of perceptual
terms like "see," "watch," "perceive," etc., such that there
is very little question that it is forthrightly a term of
sense perception. And, the most obvious connection for Arm
strong between the two words is that the section wherein he
utilizes "observe" is precisely the summary of the section
wherein he first introduces "concrete acquaintance." The
one is used to summarize the other. This would indicate
^Bertrand Russell, The Problems of Philosophy (Oxford
University Press: London, 19^0), p. 46.
33
that Armstrong has the same kind of activity in mind when he
speaks of observation as when he speaks of "concrete
acquaintance."
What "concrete" adds to "acquaintance" one can only
guess. It could mean "tangible," "actual," "real," "sub
stantial," or any number of closely related terms. But
regardless of the meaning of "concrete," Armstrong's point
seems to be that by "concrete acquaintance" he has in mind
sense experience. This I have argued follows, first, from
his Materialism. What else could he mean if all that exists
exists in space and time, and all knowledge of what exists
in space and time must ultimately rest on sense experience?
Second, this follows from Armstrong's use of "observe," a
perceptual term which he uses to summarize the section in
which "concrete acquaintance" first appears. Of course,
Armstrong might mean something else by "concrete acquaint
ance," but since he is mute of the subject, and since this
is the most reasonable interpretation of the passages in
which the phrase occurs, it seems fair to assume that he
does not have any other meaning in mind.
If this interpretation is satisfactory, then the first
premise of Argument #2 will be revised to read, "All enti
ties of which we have no sensory experiences are entities
about which we have no positive knowledge." This will
necessitate revising Argument #2.
34
Argument #2'
1. All entities of which we have no sensory experience
are entities about which we have no positive knowledge.
2. A principle of individuation for spiritual objects
(mental substance) is an entity about which we have no posi
tive knowledge.
Much might be said about the first premise of this
argument, but I will limit my comments to three points:
(1) this premise requires mental entities to be what they by
definition cannot be— sensible particulars, (2) this premise
rules out Armstrong's notion of physical substance, and
(3) this premise undercuts Armstrong's prime contention
about our mental life, to whit, that we can be knowledgeable
concerning our dispositions to cause future behavior. That
is, if these criticisms are sound, they will make two
points. The first criticism will show that if Armstrong
maintains the principle embodied in premise one, then he is
demanding from mental substance what it cannot possibly pro
vide. And it becomes a matter of definition that premise
one is true. Second, if the remaining two criticisms are
sound, they will force a dilemma upon Armstrong, a dilemma
issuing from the tension between his principle and his ren
dering both of material substance and of the nature of our
mental life. What I hope to show is that if he maintains
his principle, then he will have to alter his notion of
physical substance and his dispositional (and causal)
35
account of our mental lives. But if he foregoes his princi
ple, then his criticisms of mental substance are vitiated.
Yet he must either abandon his principle or alter his notion
of physical substance and his analysis of mental states;
otherwise, he allows his epistemology seriously to conflict
with his metaphysics.
There seems little question concerning the possibility
of sensory experience of mental entities. We cannot pos
sibly have had such experience. Given what mental substance
is supposed to be, there can be little serious concern about
whether we have had any sense experience of it. Neverthe
less, as far as I can tell, this can be no point against
Dualism. If Armstrong is requiring that mental substance be
"observable" in the same sense that a chair might be observ
able— and he seems to be doing this— then he is requiring of
mental substance something it cannot provide. By hypothe
sis, mental substance is non-spatial— and hence non-observ
able— and to demand that it be spatial is to demand that it
be something it cannot. To dismiss mental substance simply
on this account smacks of begging the question as to what
kind of thing mental substance is. Of course, Armstrong
might argue that the only kinds of things that do exist are
spatial things and this could be a necessary first step to
concluding that mental substance could not exist. But Arm
strong never attempts this kind of argument. He simply
demands of mental substance that it meet the requirements of
F
36
physical objects and then throws it out because it cannot.
Since the non-observability of mental substance follows from
the kind of thing it is, Armstrong's complaint is analytic
ally true. However, this truth should be of no worry to a
Dualist.
It was noted earlier that Armstrong argues that we
have, in direct contrast to mental substance, much "positive
knowledge" about physical substance. This would indicate
that we must have had some sensory experience of physical
substance, that is, if premise one of Argument #2' is true.
But before such judgment can be passed, it would be helpful
to try and determine exactly what Armstrong has in mind by
physical substance. Once this is done, I will argue two
points against Armstrong's account: first, that Armstrong
presents no consistent view of substance and second, that
the descriptions of substance he does present open him to
the same criticisms he levels against Cartesian mental sub
stance.
This would be made easier were Armstrong to present a
clear account of what he takes substance to be. As it is,
he offers only one sentence to clear the matter up: "I
understand by a substance nothing more than a thing that is
logically capable of independent existence."15 Since this
is all Armstrong directly says on the matter, one has to
look to his examples to fill his definition out. For
1'Armstrong, p. 7.
37
instance, when discussing Hume's "bundle" Dualism, Armstrong
thinks it a telling point that Hume's position allows mental
experiences to exist independently from all other experi
ences. This, Armstrong thinks, is mistaken; it would lead
to the embarrassing view that a "twinge of pain" could exist
independently from all else, not recognizing that a "twinge
of pain needs some further background, something to have the
twinge of pain."16 This something else (about which more
will be said later) which has the "twinge of pain," of
course, goes beyond the strictures of Hume's theory and for
this reason, Armstrong concludes that Hume cannot be right.
There can be faces without grins but there cannot be grins
without faces. Grins are not capable of independent
existence. The universe could not consist of a grin, and
absolutely nothing else. A grin demands a background,
something to have the grin. There are substances that
have soporific virtue, that is to say, the power of put
ting people to sleep. But there cannot be soporific vir
tues that inhere in no substance. Soporific virtues are
not capable of independent existence. The universe could
not consist of an instance of soporific virtue, and abso
lutely nothing else. The soporific virtue demands a back
ground, a thing to have the soporific virtue. Mental
experience, I am arguing, like grins or soporific virtues,
require something further to have them.17
Armstrong expresses his view of substance here in at
least two ways: first, he offers as a criterion as to
whether a thing is a substance or not, its ability to be the
only existing thing in the universe. If the universe could
consist of just one thing, then that thing would be a sub
stance; if not, then that thing would not be a substance.
16Ibid., p. 22. 17Ibid.
38
Second, Armstrong expresses this point derivatively when he
asks, in effect, "Does the thing in question need any other
thing or set of conditions in order to exist? Does it have
a set of necessary conditions which must prevail for it to
exist? Does it need something to be the 'background' to
make it possible for it to exist?" If it does, then it will
not qualify as a substance, but if it does not, then it
would be the kind of thing that could exist by itself, and
would thus, be a substance. This is all expressed in terms
of "independent existence," and this is in line with his
official doctrine of substance.
But not only must a thing be capable of "independent
existence," in which case it would supposedly be capable of
being the only thing in the universe. It must be "logically
capable" of independent existence. This would be all well
and good if Armstrong had made it clear what he means by
"logically capable of independent existence." But this
phrase remains unexplained so that it is left to the reader
to decipher its meaning. Unfortunately, this deciphering
activity raises more questions than it answers.
For instance, with regard to "independence," one might
ask, "independent of what?" Does Armstrong have in mind
some kind of empirical independence such as that for a thing
to be a substance there would have to be no set of empirical
conditions necessary for its existence? That this is what
Armstrong means is not at all clear, especially in light of
39
his examples of substance, since none of his examples meet
this requirement. For example, Armstrong cites human beings
as substances,18 though it would seem impossible for human
beings, as we know them, to exist without some conditions
which are necessary for their existence, e.g., oxygen.
Further, since all of Armstrong's examples are physical
entities (they would have to be), one may wonder if any of
them will pass this test. Can any physical entity exist
independent of all conditions? Could there be a universe
which consisted only of a human being, a "poisoned pill," a
"sleeping pill," etc.?1* If Armstrong does want to make
this claim, one might ask whether this is an empirical
claim. Does Armstrong want to claim that "some physical
objects can exist independently of all conditions," is some
how empirically testable? If he does consider this as an
empirical matter, one might wonder where the evidence for
this claim is. As it is, Armstrong offers none. But not
only does he not produce such evidence, he does not take
this claim to be the conclusions of some sound argument. He
simply offers a few examples of physical substance, none of
which appear obviously capable of independent existence.
This might explain why Armstrong inserts "logical" when
giving his definition of physical substance. If physical
substance is not capable of "empirical" independence, it
might be capable of "logical" independence. However, it is
18Jhid., p. 38. 1*Ihld., pp. 22, 38.
40
not clear how this insertion helps matters since it is never
clear in what sense Armstrong is using the term "logically."
For instance, "logical independence" refers to a relation
between propositions such that if two propositions are
logically independent of one another, neither one can in any
way affect the truth value of the other. If this under
standing is correct, then logical relations such as contra
diction, contrariety, implication, sub-contrariety, etc.,
cannot rightly be thought of as the relation in question,
since any two propositions related in such ways obviously
affect each other's truth value. Further, the relation of
logical independence holds between propositions, not physi
cal objects, and is understood in the former setting. To
speak of physical objects as existing "logically" indepen
dent of one another is, at best, to take a well-understood
term and use it in a context wherein there is no obvious
application. Physical objects do not bear logical relations
to one another; they do not affect the truth value of one
another simply because they have no truth value. They can
be causally related, or even ontologically related, if you
will, but to speak of them as logically related seems at
best to be misleading. In fact, Armstrong himself does not
conceive of the world as connected by any logical relations.
To him, the universe is a causally connected, contingent
universe where "mere logic" is incapable of providing any
41
information.2 0
However, though Armstrong's terminology might be mis
leading, he might be arguing along this line. He might
intend to divide existing things into at least two classes:
first, there are those things that depend (the nature of
this dependence is not now in question) for their existence
on some other thing or set of conditions. These, obviously,
are dependent entities. Second, there are those things
which have no such dependence but which are the things the
first class depends upon. This latter class will be sub
stances whereas the former class will be attributes, proper
ties, qualities, predicates, relations, or any such term
signifying some qualification of substance. For instance,
to take a case offered by Armstrong,21 "soporific virtue" is
not the kind of thing that can exist unsubstantiated. There
must be something which "has" the soporific virtue or it
would not be. (The terminology of "having" is Armstrong's.)
It would then qualify as a dependent existent, a qualifica
tion of substance; and the thing upon which the "soporific
virtue" depends would then be a substance. This seems to be
the distinction Armstrong has in mind when he speaks of
"soporific" virtue demanding a background, a "thing to have
the soporific virtue."22 And on this account, then, a
sleeping pill (again Armstrong's example) will be a sub
stance. It will be the thing which "has" the soporific
20Ibid., p. 29. 21Ibid., p. 22. 22Ibid.
k
42
virtue, the thing that "stands behind" it and gives it a
"backgroundi"
But is this account of substance consistent with the
earlier account, in which the test for a substance was
whether a thing could exist as the only thing in the uni
verse? If this account is consistent, it would follow that
a sleeping pill should be able to exist independently (in
some sense of independent) from all other objects and condi
tions. But could a sleeping pill pass this test? Could the
universe consist of just a sleeping pill? If this is Arm
strong's claim, we are entitled to some justification for
it, either in the form of empirical evidence or in the form
of some sound argument. And further, on first analysis,
there seems no reason to suppose that Armstrong is right.
There seems no reason to suppose that a sleeping pill is
the kind of thing that could exist independently of all con
ditions. For sure, none has ever been known to do so.
But this is not the only complication in Armstrong's
account of substance. His account is dangerously close, for
him at least, to the traditional account of substance as
substratum, as the thing that "stands under" its qualities,
that supports them and in a real sense "has" them. This is
a "dangerous" position for Armstrong, for were he to admit
this as the best account of substance, he would be subject
to the same complaints and criticisms he levels against Car
tesian mental substance. Physical substance, as well as
t
43
mental substance, would be an entity of which we would have
no sensory experience (concrete acquaintance) and thus we
would have no "positive knowledge" of it either. If Arm
strong does take this view of substance, then his position
would be similar, if not identical, with the notion of sub
stance made notorious by Locke as the "thing I know not
what." But given what Armstrong says about the sleeping
pill and the poisoned pill, it is hard to see how he avoids
the parallel.
For instance, if we examine the sleeping pill it will
be seen to be made up of a lot of qualities similar to
soporific virtue, in that they, too, could not exist without
some other thing "having" them. We could take its shape,
its weight, its color, its density, all its dispositional
properties and so on, and according to Armstrong, we could
properly ask, "What is the thing which has all these proper
ties?" Surely, the color of the pill does not "float"
around without inhering in something, nor does its shape,
etc. Further, these properties are not had by each other.
They require something else, the pill, to have them. What,
then, is the pill on this account? Following Armstrong's
examples, it is the substance which has all these qualities.
The pill is not simply a collection of all its properties,
but that collection plus the something else which has the
collection. This lands Armstrong squarely in a discussion I
am sure he would like to avoid, to whit, what exactly is the
44
nature of that "something else" which "has" all the proper
ties of the pill? This would lead him into some of the most
intractable problems in philosophy with a position which, by
all indications, he would not like to defend.
But more importantly for our purposes, if this is Arm
strong's view of substance, and that it is implied by his
examples, then he opens himself to the very criticisms he
brings to bear against Cartesian mental substance. That is,
since we have had no "concrete acquaintance"— sense experi
ence— of this entity, not only will we have no "positive
knowledge" about it, but all theories concerning it will be
both "formal and empty." So it would seem that Armstrong
has to give up either his attack on mental substance, at
least as this attack is presented in his first two argu
ments, or he has to alter his notion of physical substance.
Of course, Armstrong may have another view of sub
stance, or I may have misinterpreted what his position is.
As to the first possibility, this is, to say the least, not
apparent in the text. Obviously, what is needed is a clear
presentation of substance, though it would seem that the
view I have outlined is implicit in his examples. It is in
those examples that he brings up the relation of one thing
"having" another, as standing "behind" the other, or as pro
viding a "background" for the thing had. It is in those
examples, that he brings up the notion of "logical indepen
dence," of the possibility of a thing being the only consti
45
tuent in the universe. And given the nature of the things
that here are "had," one can only wonder what Armstrong has
in mind by the thing that has all these properties. It
would appear that that thing is quite similar, if not iden
tical, with the view of substance as substratum, a view Arm
strong explicitly criticizes when it is used to describe
mental substance.
Finally, by looking at one last example, one can fur
ther substantiate the claim that Armstrong does take a view
of physical substance that opens him to the same charges he
levels against mental substance. Armstrong describes a
"grin" as something needing some "background" in order to
exist.2* A grin, he argues, is a kind of thing that cannot
exist by itself but requires some other thing to have it,
and in this sense it is a dependent thing. Under this
interpretation, then, whatever has the grin would presumably
be a substance. But what are the possibilities here? Do we
take the face to be a substance? Do we take a face to be
"logically capable? of independent existence? If not the
face then the head? If not the head then the entire body?
If Armstrong means any one of the latter possibilities, then
in what sense, exactly, is any one of the latter capable of
"independent existence?" One wonders how Armstrong could
explain this without invoking the distinction he previously
introduced, that between dependent entities— properties,
2 *Ibid.
I
I
46
relations, qualities, etc., and independent entities, i.e.,
substance as substratum. But if he does the latter, then
the force of Arguments One and Two apply to his view of
physical substance as well. Needless to say, much of this
confusion could have been avoided had Armstrong, at the out
set, given a clear and consistent account of what surely is
one of his most crucial terms, "substance." Since he wants
to discredit mental substance by comparing it to physical
substance, it is only fair that he give a decently thorough
account of what he takes substance to be. But given the
account he does provide, it is evident that his view runs
head on into the epistemological principle embodied in his
first two arguments.
5
That principle— viz., that we cannot have any positive
knowledge about an entity of which we have had no sensory
experience— also subverts Armstrong's major claim about
self-knowledge. That is, fundamental to Armstrong's account
of what is given to us in acts of awareness directed toward
our own mental states, is the belief that in such acts we
are directly aware of dispositions to cause future behav
ior.21* Armstrong defines mental states in dispositional
terms and since we can be, and often are, directly aware of
our mental states, it will follow that we can be, and often
2l *Ibid., p. 82.
47
are, directly aware of our mental states, it will follow
that we can be, and often are, directly aware of disposition
to behave. But the premise in question, the premise that
expresses Armstrong's epistemological principle, is couched
in sense experience terms, and it is clear that we do not
have sensory experience of our mental states. To allow for
this, Armstrong argues that we have a faculty of "inner
sense" by which we become aware of the happenings in our
minds.25 But even if we substitute "inner sense" for "sense
experience," it does not seem the case that we ever have
direct experience of dispositions, and this follows from
what dispositions are.
To this end, three points need to be made about dispo
sitions: (1) dispositional properties or states are not
occurrent states of objects. They are properties of things
that become apparent only under presently unactualized con
ditions. For instance, the maleability of gold is not some
thing given to onlookers upon initial inspection through
ordinary perception. We could inspect gold thoroughly, and
unless we did the right things to it, we would never dis
cover its maleability. Maleability is simply not one of the
properties disclosed to the senses. And, of course, this
would naturally preclude it from being "immediately" given
to the senses. (2) Even when the maleability of the gold is
actualized, for instance when we strike it with a hammer, we
2 5Ibid., p. 95.
48
do not directly witness the maleability. We witness an
event, the depression of the surface of the gold, or a
series of events, none of which is identical with the gold's
maleability. We simply do not witness the dispositional
property. We witness its manifestations which, of course,
are not identical with the disposition itself. (3) Knowl
edge of dispositions is always counterfactual. It is always
knowledge about what would happen if, even if this knowledge
is cast in the form of a natural law. When I know that gold
is maleable, I know what would happen to gold if certain
conditions were to arise. This knowledge could be knowledge
about how gold always behaves and could be couched in terms
of a regularity of naturei Nevertheless, we are not aware
of some further property of gold as disclosed to the senses;
we know a conditional statement such that if the antecedent
is fulfilled, the consequent will follow.
The same is true of our mental lives: (1) When I admit
to having a quick temper, this is not to admit that I am
presently and directly aware of a tendency for future behav
ior. To say that I have a quick temper seems to consist of
two assertions. First, there is the implied generalization
that in the past I have often lost my temper at the slight
est provocation and, second, there is the prediction that
under like circumstances I am prone to lose my temper again.
But this is not to say that while admitting this I am
presently aware of some mental entity that is "picked out"
as a proclivity or a disposition. I simply know some coun
ter factual truth. I know what would happen if conditions
presently not prevailing were to arise. But this does not
entail, nor does it require, that I be directly aware of a
disposition. (2) Neither am I directly aware of the dispo
sition during any actual loss of temper. When the anger is
raging, I might be aware of a violent burst of uncontrolled
"feeling" accompanied by flushing cheeks, trembling hands,
unrestrained voice, etc. But at no time am I aware,
directly, of a disposition to cause future behavior. If I
reflect at all during the outburst, I would be aware of the
surge of "feeling" along with its various manifestations,
but not the disposition itself. (3) Further, at the time of
the outburst, I would be aware of something current, some
thing on-going, some present event or series of events, and
not some counterfactual state of affairs. I would not be
directly aware of unfulfilled possibilities.
I may be mistaken in arguing that we are never, either
in the "external world," or in our mental lives, directly
aware of dispositions. But if I am right, and if I am right
about Armstrong's working notion of material substance, then
there is an obvious tension between his epistemological
principle as embodied in Arguments One, Two, and Two prime
on the one hand, and his analysis of substance and his dis
positional account of mind on the other.
In this regard, I have been arguing that both his
50
analysis of substance and his dispositional account of our
mental lives run counter to the principle he utilizes in
criticizing mental substance. That is, if I am right in
arguing that Armstrong relies on the principle that to have
"positive knowledge" of a thing we must first have "concrete
acquaintance" of that thing, then it will follow that Arm
strong will have to alter or abandon his analysis of physi
cal substance along with his dispositional rendering of our
mental lives. Either this, or he must abandon or alter his
principle as he employs it. Under either alternative, how
ever, Armstrong must do severe damage to his overall posi
tion. And, in this regard, his options seem to be two.
First, he may maintain his principle, in which case he
will have to alter his notion of physical substance and his
dispositional account of our mental lives. Thus, he could
drop his notion of substance as the thing "logically capa
ble of independent existence," or as the thing that stands
"behind" or "under" properties, and opt for an analysis of
substance that allows it to be sensibly perceived. By so
doing, he might escape the force of his own criticisms of
mental substance. But, be this as it may, if he were to
keep his principle intact, he would have to alter his dis
positional analysis of mental states, and this would surely
undercut his entire theory of mind. As we will see, funda
mental to Armstrong's analysis of mind is his contention
that mental states can be accounted for dispositionally (and
51
causally). Thus, were he to abandon this account in favor
of his principle of "concrete acquaintance," he would have
to abandon his most important assertions. This seems less
acceptable than his other alternative.
Second, Armstrong could drop his principle of knowledge
and maintain his analysis both of our mental states and of
physical substance. But if he does this, and the most
obvious way of so doing would be by relaxing the principle
to allow for knowledge of physical substance and disposi
tions as he describes the latter two, then he will have to
abandon his criticisms of Cartesian mental substance. That
is, I have argued that Armstrong's prime contention against
mental substance is that we can have no "positive knowledge"
of it simply because we have no "concrete acquaintance" of
it. Yet this principle would also rule out our knowledge of
dispositions and physical substance (and I will later argue
of causes as well). Thus, to maintain the possibility of
knowing physical substance, as he describes physical sub
stance, and of knowing the dispositional content of our men
tal lives in the manner he requires— directly, non-inferen-
tially, immediately— Armstrong will have to relax his prin
ciple and allow for such cases. But, if he does so, the
Cartesian may do so as well (if he did, in fact, think Arm
strong's criticisms were telling), thus vitiating Arm
strong's original criticisms of the Cartesian position.
That is, the Cartesian may claim to know something about
mental substance on the same grounds that Armstrong would
52
claim to know something about that which "stands behind"
physical properties. If this is not open to the Dualist,
but is open to Armstrong, we would surely have to be shown
why.
Thus, the flow of the argument is this: Armstrong
explicitly argues that the Cartesian notion of mental sub
stance is both "formal and empty," and this purportedly fol
lows from the fact that we can have no "positive knowledge"
about it. Further, Armstrong implicitly argues that for one
to have "positive knowledge" about a thing, one must first
be "concretely acquainted" with that thing. Thus, since we
are not "concretely acquainted" with mental substance, we
can have no "positive knowledge" of it, and consequently,
mental substance stands as a completely vacuous notion.
This only begs a crucial question against the Cartesian
Dualist. Further, what I have argued is that we could sub
stitute in the place of mental substance, either material
substance as Armstrong describes it, or his dispositions of
material substances. In either case, the outcome would be
the same. Physical substance and/or dispositions thereof
will turn out to be entities of which we have no "concrete
acquaintance"; hence we would have no "positive knowledge"
of them and, consequently, they would remain "empty and
formal" notions. But this result would be disastrous for
Armstrong's positions as he presents them.
Armstrong's third major argument against Dualism cen
ters on the "unity" of the individual and with this in mind
he writes,
Nevertheless, there is an underlying point to this objec
tion to Dualism which is not so easy to deal with. The
difficulty is this: does a Dualist theory provide for a
sufficiently close connection between the spiritual and
the physical components of man? We ordinarily think of
the connection between the mind and the body as very close
indeed. Man is a unity. Dualism is unsatisfactory
because it breaks up that unity.26
This argument is straightforward, the thrust of the reason
ing being that man is a "unity," and hence, any theory about
man's ultimate "nature" which fails to explain, describe, or
in some way fully account for this "unity" is "unsatisfac
tory." Dualism is such a theory; it fails to give an ade
quate account of this unity, and consequently, it is "un
satisfactory." For the sake of clarity, we could formulate
the argument in this manner.
Argument #3
1. Man is a unity.
2. Any theory which fails to "fully account" for this
unity is unsatisfactory.
3. Dualism, in any form, fails to account for this
unity.
Therefore, any form of Dualism is unsatisfactory. As
in the previous two arguments, one wishes Armstrong had been
26Ibid., p. 25.
54
clearer about his key terms, especially "unity," "unsatis
factory," and "account for." Again, whatever work is done
for clarity's sake has to be done by the reader, and in
doing it, I will restrict my comments to the first two of
those terms, for it seems the soundness of the argument
rests with them.
Prima facie, the first promise seems clearly true. Man
is a "unity" in the sense that he has many aspects which
make up one thing. But this is ambiguous since there are at
least two senses of "unity" that could be at work here, one
of which is consistent with the truth of Dualism while the
other begs the question at issue. In the first instance,
"unity" could mean "one," but in a sense that does not pre
clude being made of parts. For instance, my car is "one"
thing, yet its "oneness" does not prevent it from having
many parts. Would Armstrong want to argue that because my
car has many parts it cannot be understood to be "one" thing
or it cannot be "unified?" Surely if he does, he must take
"unity" in a sense other than this commonly accepted mean
ing. Again, one can think of various objects which consist
of parts though no one would want to deny that they are
still "one" thing. Of course, the relationship among the
various parts of those objects will vary. The relationship
between the limbs and trunk of a tree might be that of
dependence or causality whereas the relationship among the
various states and a Federal Government might be legal.
55
Nevertheless, each can rightly be considered as "one" thing,
though "one" thing composed of many parts. Thus, if Arm
strong means "unity" as consistent with being composed of
parts, then this argument easily allows for the truth of
Dualism.
Two other points follow from this first interpretation
of "unity." First, under this reading, the first premise in
Argument #3, though true, does not imply the falsity of
Dualism; second, if "unity" is taken in this first sense,
the third promise in the argument could well be false and
consequently the soundness of the argument would be ruled
out. As to the first point, not only does this initial
interpretation not imply the falsity of Dualism, but it is
also obviously consistent with the truth of Dualism. Man
could be "one" thing, yet composed of two parts, his mind
and his body. With regard to the second point, if "unity"
is taken to be consistent with being divisible, then it is
not obviously true that "Dualism, in any form, fails to
account for this unity." Further, if "unity" is taken to be
consistent with "one thing made of parts," then not only
will it not obviously be false that Dualism fails to give an
account of this unity, but Materialism could just as easily
be criticized for failing to do justice to man's unified
nature since it does not deny that man has parts. This, of
course, would partly depend on the kind of parts man is com
posed of, and this first interpretation of "unity" surely
56
leaves this question open. So, if Armstrong has this mean
ing of "unity" in mind when he criticizes Dualism for fail
ing to account for man's "unity," then he surely begs the
question as to the nature of man's "parts."
Of course, Armstrong might mean by "unity" something
not consistent with being divisible. He could mean that if
anything is "one," then it is a "simple,” a thing not com
posed of parts and which resists analysis into any further,
more basic or fundamental components. Just as G. E. Moore
argued that good was a simple, indefinable quality which
could not be broken down into any more primary or elemental
constituent parts, so Armstrong might wish to argue the same
for man. Man, then, would be similar to one of Descartes'
"simple natures," something cognized so clearly and dis
tinctly" that he could not be "analyzed by the mind into
other (elements) more distinctly known." But if Armstrong
means this, then he is clearly begging the question, since
it is the "dual" nature of man that is at stake. If he
begins by assuming in the first premise of this argument
that man cannot be composed of "parts" and the concludes
that any theory that conceives man as so composed is "un
satisfactory," then he has begged the question at issue by
assuming the truth of his conclusion in his premises.
As to this first point, then, it is possible for the
first premise to be true in at least two senses. If it is
true in the sense consistent with Dualism, then not only is
doubt cast on the truth of the third premise, but the truth
of Dualism remains unchallenged. Second, if "unity" is used
in the sense that implies "simplicity," then the premise
begs the issue. Consequently, unless Armstrong employs a
sense of "unity" that neither begs the question nor allows
for the division of the thing in question into parts, his
third argument fails to do what he wants it to do, i.e.,
show Dualism to be "unsatisfactory."
7
Finally, a word needs be said about the conclusion of
this and the other arguments Armstrong utilizes to discredit
Dualism. As we have already noted, in these arguments Arm
strong has claimedonly that Dualism is either "unsatisfac
tory," "implausible," or both "formal and empty." In none
of the arguments wherein he makes these claims does he
openly argue for the truth or falsity of Dualism. He is
content, for some reason, to remain within the hazy bound
aries of such vague terms as "unsatisfactory." Not only
does he not once mention the truth or falsity of his subject
matter, but he never attempts to clarify what standards he
is appealing to when he charges Dualism with "emptiness,”
"formality," or whatever form of deficiency he has in mind.
(What he might mean will be discussed later.) This is dis
concerting in itself, but it is even more so in light of the
obvious logical tension between Dualism and the Central
State Materialism for which Armstrong argues in the latter
half of his book. That is, if what he argues for in the
core of the book is true, then Dualism is undeniably false.
This relation of contradiction between Central State Mate
rialism and Dualism seems inescapable. If man is composed
of only one kind of stuff as Armstrong claims, then man is
not composed of two kinds of stuff as the Dualist claims.
If Armstrong is right, then Descartes is wrong. It is not
just that Descartes offers an "unsatisfactory" account, it
is that he offers a false account. (Or the account is "un
satisfactory" because it is false.) Of course, Armstrong
could mean by "unsatisfactory" and his other similar terms
that whatever is such is false. He could then have as his
standard or criterion of appraisal some theory of truth or
falsity so that what appears vague in his writing would
really be quite clear. He could then argue straightfor
wardly that Dualism is false; to which my initial reply
would be that his first arguments fail to show this. But
what I would like to argue here is that Armstrong really is
arguing, implicitly at least, against the truth of Dualism
and I will cite as evidence for this claim the strategy of
his book.
With this in mind, it is important to point out that
the first sections of his book are dedicated to what Arm
strong calls "removing the rubbish," that is, to discredit
ing rival theories, so that the way will be clear to present
the one theory, that is "on the right track."
59
Like John Locke, I conceive my task negatively. I am an
underlabourer carting away rubbish from the path along
which I conceive, or guess, that scientific progress lies
. . . The first chapters consider and criticize what I
take to be the important alternatives to an identification
of mind and brain. I regard the difficulties that beset
these alternatives as a most compelling reason for believ
ing that my own theory is on the right track.27
The clear inference here is that the "difficulties" Arm
strong uncovers in the Dualist position will provide a
sweeping and devastating critique of that position, and open
the way for his alternative, the alternative which is "on
the right track." But are Armstrong's devastating critiques
simply to issue in the conclusion that Dualism is "on the
wrong track?" This, of course, would follow from the logi
cal relation between Dualism and Materialism. But is this
anything more or less than calling Dualism a false account
and Materialism the true account? (If not, then it is at
least a claim that Materialism is a better account.) Arm
strong wants the reader to reject Dualism, and to this end
he describes the Dualist alternative as "rubbish." He sup
ports this claim with a series of arguments and charges
replete with accusations of "implausibility," "unsatisfac
toriness," "formality," and "emptiness." This would indi
cate that he thinks it impossible for Dualism to be true.
Nevertheless, Armstrong disavows any interest in the
truth or falsity of the issue. The truth of Central State
Materialism or of Dualism is not to be decided by him or by
2 7Ibid., p. 2.
60
any other philosopher. He relinquishes the field to the
neurophysiologists. Armstrong's object is simply to demon
strate that there are no "conceptual," "philosophical," or
"logical" reasons for rejecting his brand of Materialism.
Its [the book's] object is to show that there are no good
philosophical reasons for denying that mental processes a
are purely physical processes in the central nervous sys
tem and so, by implication, that there are no good philo
sophical reasons for denying that man is nothing but a
material object.
It does not attempt to prove the truth of this physi-
calist thesis about the mind. The proof must come, if it
does come, from science: from neurophysiology in particu
lar.2*
But a case can be made that, even in this disclaimer,
Armstrong implicitly admits that he takes part of his task
to be the demonstration that Dualism is false. The case
rests on Armstrong's use of "good reason" while describing
the aim of his book. Under one sense of "good reason," a
good reason is one that shows the truth of its conclusion.
In this sense, then, a sound argument consists of good rea
sons offered for its conclusion. Consequently, when Arm
strong states as one of his aims the demonstration that
there are no good reasons standing in the way of identifying
mind with brain, he is stating that there are no good rea
sons implying that Materialism is false. Now one reason
that would imply the falsity of Materialism would be the
truth of Dualism. That is, if Dualism is true, then Mate
rialism is false. And since Armstrong thinks he has shown
28Ibid.
61
that there are no good reasons for supposing that Material
ism is false, he must think he has shown that Dualism is
false. This reasoning, of course, rests on a specific
interpretation of "good reason" and would be invalidated if
Armstrong has some other notion in mind. That he does have
some other notion is never indicated in the book. Thus,
since the truth of Dualism would be a good reason— in the
sense indicated— for the falsity of Central State Material
ism, and since Armstrong thinks he has shown that there are
no good reasons for supposing such Materialism false, he
must, contrary to his disavowals, assume he has shown Dual
ism false.
In sum, then, not only does Argument #3 fail to demon
strate its conclusion, but in light of the logical relation
ship between Armstrong's Materialism and Dualism, and in
view of both the language Armstrong uses to characterize
Dualism and taking into account his use of "good reason," it
appears that more is packed into the conclusion than simply
"unsatisfactoriness." I suspect that what Armstrong is
really arguing for, and wholeheartedly believes, is that
Central State Materialism is the true account of the mind/
body issue, and by implication, that Dualism could not pos
sibly be true.
8
Armstrong offers as the fourth of his five "difficul
ties" for Dualism the problem of individuating "spiritual"
62
objects.
Another serious difficulty for any Dualist theory of mind
is the question whether it can provide for the numerical
difference of spiritual objects. If we consider two phys
ical objects that exist at the same time/ we can say that
what makes them two, that is to say makes them numerically
different, is that they are in different places. However
alike they may be, their difference of place makes them
distinct from each other. Difference of place individ
uates them. If they are in exactly the same place at the
same time, they are not two objects but only one.2*
Fundamental to this criticism of "spiritual" objects or
substances is the comparison of such objects to physical
substance. In the case of physical substance, individuation
is taken to be clearly a case of space "making" one object
different or "other" than another. This is not argued for.
It is simply assumed to be the case. There is no mention
here of impressive reasons for the view that it cannot be
spatial (or other) relations that individuate. As Edwin
Allaire argues, "Relations— I'll stick with spatial one—
presuppose numerical difference; they do not account for it.
The thisness and the thatness of things is presupposed in
saying that the one is to the left of the other."3 0 This
topic, whether relations individuate or not, will not be
discussed here for the answer to it does not affect the man
ner in which Armstrong argues to the conclusion that "there
can be no acceptable principle of individuation for mental
substances." This consideration of Allaire's is brought up
2 * Ibid., p. 27.
30Edwin B. Allaire, Philosophic Studies, Vol. XVI,
1963.
63
only to indicate that more needs to be done by Armstrong
before he can justifiably assert that the principle of indi
viduation for physical objects is clearly obvious.
For Armstrong, it is simply assumed that there is no
principle which does for mental substance what the spatial
relation (allegedly) does for physical substance, even
though he presents no clear account of what he takes physi
cal substance to be. Nevertheless, Armstrong's demand for
such a principle rears a considerable problem for Dualism.
But it is not my purpose here to defend or to offer any
solution to the problem he raises. Rather, it is my aim to
analyze the arguments Armstrong presents to establish that
not only is individuation, a "difficulty" for any Dualist
theory, but that it is a "difficulty" which renders Dualism
"implausible," "unsatisfactory" and the like. What I will
argue for is that the central argument Armstrong utilizes
has as a premise a criterion of acceptability for individ
uating principles that alone effectively rules out any pos
sible principle offered by a Dualist, and that insofar as
this is done, Armstrong has closed the issue before discus
sion has ensued. In arguing for this, there will be mention
of considerations used in criticizing the second argument 1
simply because in the second argument Armstrong employs some
of the same principles as found in this fourth argument.
Armstrong's discussion of the "difficulty" of providing
a principle of individuation for "spiritual" objects is
64
divided into two sections: the first is a presentation of
the various forms which the criticism might take, and the
second centers on the assertion that even if an individuat
ing principle could be found, it would neither be on a par
with the principle that individuates physical objects, nor
would it be a principle of which we could have "concrete
acquaintance." Thus, the first section is expressed in a
series of questions, e.g., What individuates spiritual ob
jects which are identical in nature but only for a limited
span of time? What individuates them when disembodied?
This is followed by the second section wherein he writes,
As usual, the argument might be met by postulation. One
could postulate a "principle of individuation" of a non-
spatial nature to ensure the numerical difference of spir
itual objects. And, indeed, I think that the existence of
such a principle of distinction is an intelligible concep
tion. 31
Armstrong concedes that this is an "intelligible conception"
because he wants to avoid the conclusion that logic is cap
able of providing any information about the world. That is,
he wants to deny that the assertion that any two distinct
objects which exist at the same time are in different places
is a "logically necessary truth." This expresses a factual,
not a logical, feature of the world; and because of this,
Armstrong concludes that: "We cannot rule out the existence
of a non-spatial or even a non-temporal 'principle of indi
viduation' a priori." Given that such a notion is an "in-
31Armstrong, p. 28.
telligible conception," and that Armstrong does not want to
dismiss it a priori, for what reasons does he reject it? He
rejects it for the same reasons he rejects mental substance
in Arguments One and Two: we have no "concrete acquaint
ance" with such a principle. Again we are faced with the
implicit assertion and equation of "concrete acquaintance"
and sense experience. Consequently, his complaint becomes,
in effect, that "since we have had no sense experience of a
principle of individuation for mental objects, this fact
poses a difficulty for any Dualist theory." This is sup
ported by a passage used in the first argument, but since it
embodies Armstrong's position so clearly, it bears repeat
ing.
Nevertheless, it does seem likely that the only "principle
of individuation" with which we have any concrete
acquaintance is that of being in different times and
places. The inability of the Dualist to say anything at
all about his spiritual "principle of individuation"
strongly suggests that the only way he can understand his
notion is a negative way as a "nonspatial principle of
individuation." It seems that he is not in fact acquaint
ed with such a principle, nor is there any reason to
postulate it.32
This passage was used in the first section to help
establish a connection between "concrete acquaintance" and
"positive knowledge,” such that the latter was seen to be
dependent upon the former. In this context, two points may
be made about it. First, we can see that what Armstrong
wants from a Dualist with regard to the problem of individ-
32Ibid.
66
uation is some "positive knowledge" about whatever principle
he proposes. Needless to say, this would require the Dual
ist to be "concretely acquainted" with such a principle.
But this is to require the Dualist to have some sense expe
rience of this "non-spatial" principle, and this seems an
unfair request of him, in the sense that it would be impos
sible, by hypothesis, to fulfill. One must, rather, argue
with the hypothesis itself. Instead, Armstrong expects the
Dualist to produce a non-spatial principle of individuation
which is nevertheless open to sensory inspection. Armstrong
is here using a criterion of acceptability which presupposes
the falsity of whatever principle the Dualist has in mind,
instead of refuting the Dualist principle itself. Second,
Armstrong asserts that there is no reason to postulate a
non-spatial principle of individuation. But this surely
begs the question. There would be no reason to postulate
such a principle only if there are no mental objects. In
that case, there would be no reason to postulate a principle
of distinction among them. But the existence of such
objects is what is in question here, so that one cannot
assume an answer to this question in order to refute an
answer about individuation.
But here, as in the first two arguments, Armstrong
might balk. He might reply that he does not mean "sense
experience" when he uses "concrete acquaintance." Neverthe
less, it is hard to imagine what else he could mean and
(
still be consistent with what he says elsewhere and with
his thorough Materialism. That is, as in the first argu
ment, Armstrong summarizes this section on the problem of
individuation with the perceptual term "observe," and this
indicates that what would satisfy him would be some princi
ple that disclosed itself to the senses. As we saw this
insistence seems odd in the face of his treatment of physi
cal substance and his dispositional account of the mental
life. In neither case are we aware of the objects in ques
tion via the senses. In addition, if Armstrong is to main
tain his Materialism, then the most compatible epistemology
with it would be one grounded in the senses. If no other
things exist than those that take up space, and if we are to
be aware of these things, then it would seem that such
awareness would have to come by way of the senses. And
given this framework, it is understandable why Armstrong is
so "dissatisfied" with the Dualist theory. Consequently,
one has to conclude that Armstrong must have some strong
correlation, if not strict identity, between "concretely
acquainted" and "sense experience"; and if this is the case,
his demand that the Dualist present a principle of individ
uation for mental objects of which we could be "concretely
acquainted" presumes an answer which, as we have noted, pre
cludes the Dualist alternatiave. If Armstrong does mean
something else by "concrete acquaintance" it is surely
incumbent upon him to clearly say so.
68
9
The last two arguments Armstrong brings to bear against
Dualism center on the where and the when of the Dualist
position. That is, Armstrong wants to know exactly where
the mind acts on the body, if in fact it does so, and when,
exactly, in the history of the body the mind emerges. Arm
strong argues that all attempts to answer these questions
conflict with what we know about the "workings of the
world," specifically physiology, and consequently his con
clusion is again that Dualism is "unsatisfactory," "unten
able," etc. The latter of these two questions seems to pre
sent the greatest difficulty for a Dualist theory, for
implicit in Armstrong's question is a demand for a clear
explanation of how interaction (omitting for the moment the
Parallelist account) actually takes place. Even if Arm
strong's arguments do not show that interaction cannot take
place, this does not absolve the Dualist from attempting an
account, consistent with what appears to be the facts both
of physiology and introspection, of how the mind and body
are related.
The first argument, concerning "when the mind emerges
in the growth of men and animals" issues from a view of the
mind-body relationship that Armstrong thinks a Dualist must
take.
It seems that the Dualist must conceive of the emergence
of mind in the following way. At some time after concep
tion, when the nervous system of man and the higher ani
mals reaches a certain level of physiological complexity,
69
a completely new, non-spatial entity is brought into
existence in a certain completely new sort of relation to
the body. The emergence of this new existence could not
have been predicted from laws that deal with the physical
properties of physical things.3*
It is a new individual existent, not new properties that
would surprise Armstrong. He accepts as "plausible" the
notion that the body, more specifically the central nervous
system, could at some time in its history produce some non
physical properties. What he finds difficult is the notion
of the body producing some new entity, both separate from
itself and of an entirely different nature.
It is not a particularly difficuLt notion that, when the
nervous system reaches a certain level of complexity, it
should develop new properties. Nor would there be any
thing particularly difficult in the notion that when the
nervous system reaches a certain level of complexity it
should affect something that was already in existence in a
new way. But it is a quite different matter to hold that
the nervous system should have the power to create some
thing else, of a quite different nature from itself, and
create it out of no materials.3*
Armstrong allows, then, the Dualist three alternatives.
First would be the development, by the central nervous sys
tem, of some new entity, second would the creation by the
central nervous system of a new set of properties, and the
third would be the affecting, again by the central nervous
system, of some already existing entity. Armstrong quickly
rules out the third alternative simply because "There is
very little evidence for believing that the mind pre-exists
its body."35 But all three alternatives exist as empirical
33Ibid., p. 30. 3 *Ibid. 35Ibid.
70
possibilities. It is possible, Armstrong concedes, for the
central nervous system to affect some pre-existing entity in
some new way, and it is possible for the central nervous
system to develop new properties; it is even possible for
the central nervous system to bring into existence something
distinct and separate from itself. "Admittedly, there is no
contradiction in the notion that the nervous system should
have these powers, but what we know of the working of the
world makes the hypothesis a very unlikely one."*6 It would
appear, then, that the core of Armstrong's complaint does
not consist in pointing out that these alternatives are
highly unlikely— though he does think they are. The core of
his complaint seems to center on the timing of these events.
When, exactly, does all this occur? If the Dualist wants to
argue that the body in some way produces some new proper
ties, for instance, exactly when does this take place?
Should there not be some alteration in the bodily operations
that would indicate the precise moment when the "creation"
took place? Given that the new items are so radically dif
ferent from the body, and given that such an event is rather
momentous, Armstrong thinks the Dualist must be able to
point to some time wherein the physiological changes would
indicate that something new had just been brought into
being.
At what point in the development of the organism shall we
S6Ibid.
71
say that such a momentous event occurs? The difficulty
is that there seems to be nothing in the physiological
development of the organism to suggest any point of sharp
break . . . But because the Dualist sets up so sharp a gap
between the material and the mental/ he must find a
definite point when the mental comes into existence.37
Since Armstrong has already ruled out the alternative of
some pre-existing entity, the "definite point" he refers to
has only to do with the remaining two possibilities. Both
of these possibilities, Armstrong thinks, will require some
flurry of appropriate physiological activity to signal their
arrival. And, since no flurry has been found, Armstrong
concludes that Dualism does not jell with modern physiology,
and hence, cannot be "plausible" or "satisfactory." Arm
strong's argument could be formulated in the following
manner.
Argument #4
1. Any theory of mind which does not account for the
emergence of mind in accord with modern physiology is
implausible.
2. Dualism fails to account for the emergence of mind
in terms of modern physiology.
Therefore, Dualism is implausible.
Now, since Armstrong's final argument is based on the
same considerations as this argument, and since generally
the same criticisms will apply to both arguments, it is
simpler to go ahead and develop the final argument and then
*7Ibid., p. 31.
72
comment on them both concurrently.
Whereas Armstrong's first argument in this section has
to do with the "when" of the mind/body relationship, the
last argument concerns the "where" of the relationship. The
Dualist assertion is that mind and body causally interact
with one another, and Armstrong thinks that if this is the
case, there must be some exact spot where the mind causally
affects the body, and he thinks the natural place for this
to occur would be in the brain.
Difficulties arise when we try to think out an Interac-
tionist theory in a concrete way. It seems that the mind
will have to act on the body by acting on the brain. Now,
to a physiologist, the brain is an enormously complex and
highly organized system to which physical stimuli of
various sorts are applied, and which in turn has certain
physical effects on the rest of the body. Inside the
brain one physical event is followed by, and is the cause
of, another physical event. If the Dualist is to be an
Interactionist, two conditions must be satisfied. In the
first place, there must be last physical events in the
brain which are followed by, and are the causes of, mental
events. (Fulfillment of just this condition does not mark
off Interactionism from Parallelism.) In the second
place, there must be last mental events which are followed
by, and are the causes of, events in the brain.*8
Armstrong, then, pictures the activity of mind and brain as
a series of events, the last event being a mental event
which is both preceded on the "way up" by a physical event
and followed on the "way down" by another physical event.
Thus, if I am burned by a flame, stimulation is sent "up" to
the brain, then there is the mental event of feeling the
pain, and then there are stimulations sent "down" so that I
3 8
Ibid.. p. 32,
will wring my hand or perform whatever activity is appropri
ate. The events leading up to the "feeling" are all physi
cal as are the events leading away from it so that we have a
causal sequence consisting of both mental and physical
events.
On the "way up" there must be a last physical event in the
brain before the mental events ensue. The mental events
must then bring about a first physical event in the brain
on the "way down."3*
Further, were we able to isolate such a series of events, we
should be able to discover a "gap" in it, exactly where the
mental event would take place. The series would continue on
its "way up," there would be a brief pause, and then the
series would begin its travel "down." "This means that
there will be, as it were, a "gap" between the state of the
brain before the mental event has had its effect and the
state of the brain after the mental event has had its
effect."1 *0 But since physiologists have never found such a
"gap," Armstrong is reluctant to think there is one. "Now
with the gradual advance of knowledge of the operation of
the brain and nervous system, physiologists are becoming
increasingly unwilling to think that there is any such
gap."1 *1
Armstrong's argument, then, seems to be this. If the
mind really did act on the body there would have to be some
break or "gap" in the neural series of any sensation, and
38 Ibid..
“ • i b i d .
**1Ibid.. p. 33.
74
even though physiologists have not fully isolated such a
series, Armstrong does not think any of the physiologists
are willing to admit that such a break does exist. This
indicates, to Armstrong, that the mind does not act on the
body, and this is a death blow to interactionism. Again,
the argument can be formulated as a two-premise argument but
of the form modus tollens.
Argument #5
1. If the mind does act on the body, there would have
to be a "gap" in the neural series of any sensation.
2. No "gap" has been discovered.
Therefore, mind does not act on body.
Three points may be made about these last two argu
ments, all of which indicate that these are not the best
kinds of arguments to disprove Dualism. First, not all
physiologists are as unanimous in their opinion as Armstrong
intimates. Some recognize an irreducible and fundamental
distinction between mind and body and are seemingly not
daunted by the considerations brought to bear here by Arm
strong. Second, there is no a priori reason why mind and
body, where mind is taken to be a non-spatial entity, cannot
interact; or, in general why something without spatial loca
tion cannot be causally related to something which instances
no spatial relations at all. Third, Armstrong demands that
the Dualist fully describe the "mechanism" or the "how" of
the causal interaction of mind and body, even though the
75
"how" or the "mechanism" of straightforward physical causa
tion eludes our scrutiny. That is Armstrong's very ques
tions may be raised about clear cases of physical causation,
and if Hume is right, we will end up with events which defy
full explanation. It might be possible that Armstrong's
problem with the mechanism of mind-body interaction is more
a problem with the causal relation in general than with
mind-body interaction in particular.
In the first instance, it seems clear that not all
physiologists share Armstrong's convictions. For instance,
Wilder Penfield thinks that there exists a fundamental chasm
between mental activity and brain activity.
. . . for the neurophysiologist there is a working bound
ary (between the brain and the mind) that does exist.
Physiological methods bring him nearer and nearer to it.
But he comes to an impasse, and beyond that impasse no
present-day method can take him.1 *2
In fact, Penfield and other physiologists assume some form
of Dualism and feel justified in doing so by the nature of
the experiments they perform on patients. For instance, the
neurophysiologist stimulates the brain electrically or chem
ically and then asks the patient what he remembers, thinks,
sees, feels, etc. Since the thinking, feeling, seeing, and
remembering seem clearly to be caused by the stimulation, it
is hard to resist the temptation to conclude that two dif
ferent processes are involved. First there is the stimula
tion, and then there is the experience as reported by the
patient. Here are typical experiments as reported by
1 *2Wilder Penfield and Lamar Roberts, Speech and Brain
Mechanism (Princeton University Press, 1959), p. ST
76
Penfield.
There is an area of the surface of the human brain where
local electrical stimulation can call back a sequence of
past experience . . . The sights and sounds, and the
thoughts, of a former day pass through the man's mind
again.* 3
Occasionally . . . gentle electrical stimulation in this
temporal area, right or left, has caused the conscious
patient to be aware of some previous experience. The
experience comes back to him in great detail.‘ >l*
Many more such examples could be cited, but the point is
clear. It is evident to some physiologists, pace Armstrong,
that there is a boundary between neural events and "experi
ence," though the two may be, and most undoubtedly are,
intimately connected. Of course, if this is the case, then
pointing out the difficulty of pinpointing that intimate
connection does little to show that such a connection fails
to exist. Penfield might reply, "So much the worse for our
knowledge." In fact, Penfield cites the assumption of the
mind-body distinction as necessary for the workings of
neurophysiology.
Let us consider the brain-mind relationship briefly . . .
it is a boundary which, as some philosophers explain it,
does not exist at all. But for the neurophysiologists
there is a working boundary that does exist. Physiologi
cal methods bring him nearer and nearer to it. But he
comes to an impasse, and beyond that impasse no present-
day method can take him. If he should state that nerve
impulses moving in certain patterns are one and the same
thing as mind, he accomplishes little for his future work
except to deprive himself of a useful working terminology.
Any man who adopts the dualistic terminology speaks of
two elements in a living conscious human being: a body
and a soul, a brain and a mind, electrical energy con-
**Ibid. ""Ibid.
77
ducted through the integrating patheways of the cerebral
hemispheres and conscious thought, a living machine and a
spirit. However it is expressed, he must think either of
a parallelism or a back and forth relationship. **5
Second, there is no a priori reason why mind and body
(non-spatial and spatial entities) cannot interact. This
must be brought up because it seems implicit in Armstrong's
writing that there could be no causal relation between mind
and body simply because the two are so "radically differ
ent. It is true that Armstrong does not explicitly include
in his reasoning the supposition that any two entities in
different "categories" cannot be causally related. Never
theless, there is enough mention of the difference of mind
and body that this might well be an assumption on Arm
strong's writing that there could be no causal relation
between mind and body simply because the two are so "rad
ically different." It is true that Armstrong does not
explicitly include in his reasoning the supposition that any
two entities in different "categories" cannot be causally
related. Nevertheless, there is enough mention of the dif
ference of mind and body that this might well be an assump
tion on Armstrong's part. For instance, he speaks of the
body producing something "new" or of entering in "special"
relations with this "new" object, all seemingly indicating
that he finds this activity terribly strange; and it may be
that its strangeness for Armstrong rests on his conviction
1,5 Ibid.
78
that since the two things in question are not of the same
"kind," they cannot be causally related. What else might it
rest on? But a number of philosophers, including C. D.
Broad, have pointed out that the causal relation exists
among items that, at least on the surface, do not appear to
be at all similar.1 *6 How similar is a draft on a chilly
night and a cold the next morning, an insult and a tear, a
missed tennis shot and anger? Such items may be causally
related though their similarity is not evident. It may be
that mind and body do not interact (Though on Armstrong's
own view they would have to. Parts of the brain and central
nervous system would be causally related to one another and
since the mind is the brain, the mind and brain would inter
act causally), but it does not seem that the reason that
they do not would have to be some a priori reason about
their dissimilarity.
Finally, Armstrong insists on the "how" of interaction-
ism. If mind acts on body, or if body "creates" mind, then
Armstrong wants the Dualist to pinpoint the spot where the
mind affects the body, more specifically where the mind
affects the brain, and he would like the Dualist to display
the "mechanism" of this causal relation. Further, if body
is supposed to "create" mind (Argument #4), then Armstrong
1 *6C. D. Broad, The Mind and Its Place in Nature as
found in A Modern Introduction to Philosophy, 3rd ed., Paul
Edwards and Arthur Pap, editors (New York: The Free Press,
1973), p. 203.
would like a full account of how this is accomplished. Pro
viding that the Dualist cannot produce a clear account of
the "how" or the "mechanism," Armstrong will conclude that
this inability on the part of the Dualist is good reason for
denying the mind, as something distinct from the body,
causally interacts with the body. What I will argue is that
the inability of the Dualist on this score is paralleled by
the inability of anyone to give the "how" or the "mechanism"
of purely physical causation. In both instances, whether it
be mind-body interaction or whether it be body on body, the
"how" is kept from our scrutiny. And if this is the case,
the force of Armstrong's demand is vitiated.
At the outset, one would expect Armstrong to give a
minimal account of what he takes the causal relation to be
since in many respects, this is the most important term he
employs. Not only does it play a vital role in his criti
cisms of interactionism, but it is crucial to his disposi
tional-causal theory of mind. In spite of its importance,
however, one never gets a clear description of the causal
relation, though some hints as to what it must be are con
tained by Armstrong in his criticisms of Cartesian Dualism.
Armstrong implicitly claims in those criticisms that if we
are to have any "positive" knowledge of an entity, then we
must first have "concrete acquaintance" of that entity. I
have argued that by "concrete acquaintance," Armstrong means
some form of sense perception. As applied to the causal
80
relation, this would require us to have some sense percep
tion of the causal relation in order to have any knowledge
of it. The problem here is that this requirement forces the
causal relation to be something no one thinks it is, i.e.,
something open to inspection by some kind of sense percep
tion. No philosopher I know of, especially since Hume,
wants to support this kind of interpretation. Nevertheless,
this is exactly what Armstrong wants to do. He wants to
assert that in touch we are directly aware of the causal
relation. Armstrong has to so argue in order to maintain
his criticisms of Cartesian Dualism and to support his con
tention that in "inner sense" we are directly aware of
actual causes. His other alternative is to accept an analy
sis of the causal relation which does not make it access
ible to the senses, in which case we could have knowledge of
something without being "concretely acquainted" with it.
But since this would materially affect his criticisms of
Cartesian Dualism, Armstrong avoids such an analysis.
But regardless of which analysis of the causal relation
Armstrong does opt for, he, too, can be required to give an
account of the "how" of it. One is immediately reminded of
Hume.
Hence we may discover the reason why no philosopher who is
rational and modest has ever pretended to assign the ulti
mate cause of any natural operation, or to show distinctly
the action of that power which produces any single effect
in the universe. It is confessed that the utmost effort
of human reason is to reduce the principles productive of
natural phenomena to a greater simplicity, and to resolve
the many particular effects into a few general causes, by
81
means of reasoning from analogy, experience, and observa
tion. But as to the causes of these general causes, we
should in vain attempt their discovery, nor shall we ever
be able to satisfy ourselves by any particular explication
of them. These ultimate springs and principles are
totally shut up from human curiosity and inquiry. Elas
ticity, gravity, cohesion of parts, communication of
motion by impulse— these are probably the ultimate causes
and principles which we shall ever discover in nature; and
we may esteem ourselves sufficiently happy if, by accurate
inquiry and reasoning, we can trace up the particular phe
nomena to, or near to, these general principles. The most
perfect philosophy of the natural kind only staves off our
ignorance a little longer, as perhaps the most perfect
philosophy of the moral or metaphysical kind serves only
to discover larger portion of it. Thus the observation of
human blindness and weakness is the result of all philoso
phy, and meets us, at every turn, in spite of our endeav
ors to elude or avoid it.
Hume points here to the difficulty (for him, an impossibil
ity) of explaining the ultimate nature of standard physical
causation. Hume argues that if we take a supposed instance
of physical causation, we might at best be able to describe
the event under some highly general principles of nature;
but if we try to explain how it is that these principles
yield the effect they do, Hume thinks we will have come to
the end of our investigation. We will have pushed our
knowledge to its limits, only to make evident our ignorance.
We will have come, for Hume, to a brute correlation of
events which resists any further attempt to analyze or
describe the "mechanism" involved. The how and the why of
one event bringing about another, even if it invariably does
so, is simply beyond our discovery; we simply have to
**7David Hume, An Inquiry Concerning Human Understanding
(Indianapolis; New Yorks Bobbs-Merrill Company, Inc.),
p. 45.
acquiesce in our "blindness and weakness . . . in spite of
our endeavors to elude or avoid it."
Thus, Hume offers a challenge to anyone who not only
thinks there is a causal connection between what are norm
ally thought to be causes and effects, but he also thinks
that such a connection can be described. And one would
think by his demands that Armstrong falls into this cate
gory. That is, Armstrong demands from the Dualist just what
Hume thinks it is impossible to give. Armstrong wants in
mental terms what he has yet to provide in physical terms,
i.e., the "mechanism" of straightforward physical causation.
This is not to indicate that his questions are not fair
ones. They are. But the conclusion Armstrong draws from
his dissatisfaction with what he takes to be the Dualist
reply, is not altogether fair, unless he is willing to
accept the same kind of answer with regard to physical
causation. His own position lives by failing to apply to
it questions used to reject Dualism.
Regarding physical causation, it has been notoriously
difficult to provide answers to the very causal questions
Armstrong raises about mind-body interaction. C. J. Ducasse
reiterates Hume's point when he argues that causal explana
tion ceases with "proximate" causes. In such cases, we sim
ply have brute facts.
Moreover, the objection that we cannot understand how a
psychical event could cause a physical one (or vice versa)
has no basis other than blindness to the fact that the
"how" of causation is capable at all of being either
mysterious or understood only in cases of remote causa
tion, never in cases of proximate causation. For the
question as to the "how" of causation of a given event by
a given other event never has any other sense than through
what intermediary causal steps does one cause the other. 8
Armstrong never attempts to answer Ducasse. In fact,
Armstrong never clarifies what he takes the causal relation
to be.
So, regardless of the wide-ranging attempts, from regu
larity to necessary and sufficient conditions to "manipula
tory handles," etc., one may still ask, "How does A bring
about B?" Even if A is a set of necessary and sufficient
conditions, how does A "produce" B? Armstrong never
attempts to answer such questions. He never offers an
account of what the causal relation is or how it is that A
brings B about. Thus, the "mechanism of the simplest kind
of physical causation is not displayed. And it would seem
by the historical attempts made to answer Hume, that Arm
strong might have a difficult time doing so. But if this is
the case, then Armstrong's criticisms cut both ways. The
result is that our ignorance, even about physical causation
is reasserted. Armstrong has never shown us in a convincing
manner that mind-body causation is unintelligible or absurd
for reasons which do not likewise show body-body causation
to be unintelligible or absurd.
Further, Armstrong appears to assimilate mind-body
**8C. J. Ducasse, "In Defense of Dualism," in Dimensions
of Mind, S. Hook, ed. (New York: Collier Books, 1961),
p. 68.
84
interaction with standard physical causation. He pictures
the former as a linear series of events with the mental
event being the result of the "upward" sequence and the
cause of the "downward" sequence. Thus, mind-body interac
tion is made to be similar to simple physical causation: it
is seen as flowing through temporally and spatially distinct
yet contigious events with the lone exception that one men
tal event is "inserted" about midway. Armstrong's model,
then, is standard physical causation and it appears that he
rejects mind-body interaction because it does not seem to
him to be in accord with this model.
Leaving aside, for the moment, the question of whether
mind-body interaction does in fact accord with Armstrong's
model, we might consider whether it is reasonable to assume
that it should accord with it. After all, might it not be
mere surprising that it did so accord than if it did not?
For it seems clear, that if the Dualist interactionist is
right, there will be two radically different kinds of enti
ties standing in causal relations with one another. In such
a situation, is it not just as reasonable to expect that the
causal relation between such entities would be quite differ
ent than the causal relation between objects of the same
kind? In such an instance, one might expect the mind-body
relation to be dissimilar to the causal relation existing
between two events in the physical realm. And, since Arm
strong does hold mind-body interaction deficient simply
85
because it fails to comport with his idea of body-body cau
sation, one might question his model and the subsequent
criticism based on that model. Analogously, when one com
pares universals to physical objects, one does not reject
the former simply because they do not resemble the latter.
In conclusion, then, it does not appear that in these
last arguments that Armstrong has presented insuperable dif
ficulties for Dualism, and this for three reasons. First,
not all physiologists, contrary to Armstrong's assertions,
agree with Armstrong. Some find there to be a legitimate
boundary between mind and brain, even though the two be
intimately connected in some yet unspecified way. Second,
there is no a priori reason why mind and body may not inter
act. What causally interacts with what can be determined
only by experience and observation. Their dissimilarity is
not of itself sufficient reason to prevent causal interac
tion. And, third, the very question Armstrong raises with
regard to mind-body interaction may be raised to physical
causation such that it appears no complaint against Dualism
that a fully satisfactory answer has not yet been given.
9
This concludes our canvasing of the arguments Armstrong
uses to "discredit" Dualism. His aim was to remove the
"rubbish" of Dualism, though I hope it has been shown that
he was not entirely successful. Dualism surely may be
false, but it doesn't appear to be false (or "unsatisfac
86
tory") for the reasons Armstrong presents. It is time now
to look at Armstrong's positive account of the mind-body
relationship.
CHAPTER III
CENTRAL-STATE MATERIALISM STATED
Chapter 6 is the backbone of Armstrong's A Materialist
Theory of the Mind, since it is in that chapter that Arm
strong expounds the basis of his Central State theory.
Though the chapter is relatively short, it contains his most
important claims and arguments; indeed, the balance of the
book may be seen as the working out of what is argued for in
this chapter. It would be helpful then, in advance of any
possible criticisms of Armstrong's position, to reconstruct
clearly the main outlines of the view he takes to be "on the
right track" about the mind-body relationship. This will
not necessitate that his entire view be presented; but just
those features which are crucial and which I explicitly
examine in later chapters. With this end in mind, I will
divide the present discussion into three sections: (1) the
problems Armstrong considers as they arise from the contin
gency of his identity claim, (2) his divergence from the
identity claims of U. T. Place and J. J. C. Smart, and
(3) his disavowal of Behaviorism.
1
The core of Armstrong's view is quite simply stated.
87
But now we must examine the second form of Materialism,
the view which identifies mental states with purely physi
cal states of the central nervous system. If the mind is
thought of as "that which has mental states," then we may
say that, on this theory, the mind is simply the central
nervous system, or, less accurately but more epigram-
matically, the mind is simply the brain.1
The identity asserted between the mind and the brain is the
central claim Armstrong makes, and upon stating it, he
immediately considers what he takes to be "the great prob
lem, or, at any rate, one great problem faced by the Central
S tate theory."2
This problem stems from the contingency of the identity
asserted between mind and brain. Armstrong is not, nor are
any of the Identity Theorists, arguing that "The mind is the
brain" is a logically necessary truth. Surely, there is no
contradiction in denying that "The mind is the brain," as
there would be were the identity claim logically necessary.
Further, it would be strange were one to mean by "I am
thinking now" that some specific events were presently
occurring in one's brain, when one knew nothing about the
workings of the latter. Finally, if "The mind is the Brain"
is a logically necessary truth, then the empirical nature of
Armstrong's claim, as expressed by the fact that he believes
science will either vindicate or repudiate his position,
would be obviated. So it is clear, both from the nature of
the claim in question and by Armstrong's clear statements
about the issue, that Armstrong's identity claim is nothing
Armstrong, p. 73. 2Ibid., p. 78.
89
if not contingent.
"The mind is the brain" is not a logically necessary
truth. When Aristotle said that the brain was nothing but
an organ for keeping the body cool, he was certainly not
guilty of denying a necessary truth. His mistake was an
empirical one.3
But since his claim is contingent, strictures are
placed upon it which lead directly to "the great problem" as
Armstrong describes it. Armstrong realizes, as other Iden
tity Theorists before him have, that because his claim is
contingent, he must be able to give what he calls "logically
independent descriptions" of both terms of the identity.
That is, the two terms of the identity must not initially
appear to be the same; they must appear as two. Otherwise,
there would be no issue to be settled, and even though such
difference will vanish after conceptual analysis and empir
ical discovery, the apparent difference is the crucial
starting point. In this sense, then, Armstrong's identity
claim presupposes difference. J. J. C. Smart recognizes this
when he writes that a difference of properties is a pre
requisite for identifying the Evening star with the Morning
star.
For suppose we identify the Morning Star with the Evening
Star. Then there must be some properties which logically
imply that of being the Morning Star, and quite distinct
properties which entail that of being the Evening Star.1 *
"Logically independent descriptions" then, seem to be
3Ibid., p. 77.
‘ ‘ J. J. C. Smart, "Sensations and Brain Processes" in
The Philosophy of Mind. V. C. Chappell, ed., 1962.
90
features of any contingent identity claim, and Armstrong
clearly recognizes this.
But if "The mind is the brain" is a contingent statement,
then it follows that it must be possible to give logically
independent explanations (or, alternatively, "ostensive
definitions") of the meaning of the two words "mind" and
"brain." For consider, "The morning star is the evening
star" is a contingent statement. We can explain the mean
ing of the phrase "the evening star" thus: it is the very
bright star seen in the sky on certain evenings of the
year. We can explain the meaning of the phrase "the morn
ing star" thus: it is the very bright star seen in the
sky on certain mornings of the year.5
Since this is true of any contingent identity claim, it will
thus be true of the identity asserted between mind and
brain. "Now if it is meaningful to say that "The mind is
the brain," it must be possible to treat the words "mind"
and "brain" in the same way.6 Armstrong must, then, give
his "logically independent descriptions" of the two terms to
be identified, "mind" and "brain." He thinks the brain the
easier of the two to define since it can be done so osten-
sively. It is the description of the mind that presents a
problem for Armstrong, and it is this problem that consti
tutes "the great problem" for the Identity Theorists. Arm
strong's problem is this: how can he define mind without
either (1) begging any important questions as to what kind
of thing mind might be, or (2) attributing properties to
mind which prevent it from being identified with the brain.
The first requirement is not unique to this issue; indeed,
one would hope it met in any discussion. But the second
5Armstrong, p. 77. 6 Ibid.
91
requirement is peculiar to contingent identity claims, in
that if one term of the asserted identity is discovered to
have properties not predicable of the other term, then the
identity cannot be made. Thus, if mind turns out to have
properties not predicable of physical objects, then mind and
brain cannot be identical. And prima facie, even Armstrong
has misgivings about the endeavor in this regard.
Put the problem another way. Central-state Materialism
holds that when we are aware of our mental states what we
are aware of are mere physical states of our brain. But
we are certainly not aware of the mental states as states
of the brain. What then are we aware of mental states as?
Are we not aware of them as states of a quite peculiar,
mental sort?
The problem has so daunted on physicalist, Paul Feyera-
bend, that he has suggested that the materialist ought
simply to recognize that his world-view does not allow
statements that assert or imply existence of mind. . . .7
So, part of Armstrong's problem consists in so describing
mind that no uniquely mental properties are interjected; for
if they are, minds can not possibly be identified with
brains. And this must be done without assuming what minds
must be at the outset.
Armstrong proposes his Causal Theory of the mind par
tially as a solution to this problem. The crux of this
attempt rests in his denial that we are aware of any proper
ties "intrinsic" to minds. That is, his analysis of mind
leaves open the possibility that minds are brains simply
because his analysis is neutral about the properties given
to us whenever we are introspectively aware of any of our
7Ihid., p. 78.
92
mental states. Since minds are not given in any of their
"intrinsic properties," there is no possibility that Arm
strong has begged the question about mind's nature or that
self-awareness discloses any uniquely mental properties.
Hence, Armstrong believes he has overcome his most serious
obstacle, and that consequently, the road is open to "dis
cover" what minds really are. And he is confident we will
discover that minds are brains. Thus, Armstrong presents
the following picture wherein he develops a view of mind
neutral about its nature.
Psychologists very often present us with the following
picture. Man is an object continually acted upon by cer
tain physical stimuli. These stimuli elicit from him cer
tain behaviour, that is to say, a certain physical re
sponse. In the causal chain between the stimulus and the
response fall physical processes in the central nervous
system, and nothing else at all, not even something "epi-
phenomenal." At the same time the theory cannot mention
the central nervous system in its account of the concept
of mind. If we now think of the psychologist's picture
the outline of a solution is in our hands. As a first
approximation we can say that what we mean when we talk
about the mind, or about particular mental processes, is
nothing but the effect within a man of certain stimuli,
and the cause within a man of certain responses. The
intrinsic nature of these effects and causes is not some
thing that is involved in the concept of mind or the par
ticular mental concepts. The concept of a mental state is
the concept of that, whatever it may turn out to be, which
is brought about in a man by certain stimuli and which in
turn brings about certain responses. What it is in its
own nature is something for science to discover. Modern
science declares that this mediator between stimulus and
response is in fact the central nervous system, or more
93
crudely and inaccurately, but more simply, the brain.8
Here, then, is a description of the mind as a neutral "in
termediator" or "transformer" between whatever stimuli
affects us and the resulting response to that stimuli. And,
since we know nothing about the "intrinsic nature" of this
intermediary, it could very well turn out to be the brain,
which in fact, science has "declared" it to be.
2
Clearly Armstrong's Central-State theory is similar to
the identity claim of U. T. Place and J. J. C. Smart. The
latter also gives a "topic neutral" analysis of minds and
mental states. As Smart writes,
My suggestion is as follows. When a person says, "I see a
yellowish-orange after-image," he is saying something like
this: "There is something going on which is like what is
going on when I have my eyes open, am awake, and there is
an orange illuminated in good liqht in front of me, that
is, when I really see an orange.
Just as Armstrong argues that we know nothing about the
"intrinsic nature" of minds or mental states, Smart argues
that when we experience a colored sense-datum, we are aware
only of "something going on." In neither case are any spe
cific properties (as opposed to relations) revealed; for
clearly if any properties were revealed, and were those
properties uniquely mental, then the identity claimed
between mind and brain would be doomed. But even though
8Ibid., p. 79. 8Smart
94
Armstrong's view is similar in this respect to that of Place
and Smart, Armstrong is anxious to state the differences as
well.
Armstrong cites as the major difference between his
view and that of Place and Smart, the fact that he gives an
account of mental states in terms of causes and effects,
whereas they concentrate solely on the effects in us of spe
cific stimuli. Thus, Armstrong interprets Smart as analyz
ing mental phenomena entirely in terms of something acting
on us, without mentioning what in many instances is equally
as important, our reaction to whatever it is that is affect
ing us. This is exemplified by the yellowish-orange after
image Smart likens to the stimulus we normally receive when
we really do see an orange. In Smart's example, there is no
mention of our response to such a stimulus. (It must be
noted, however, that Smart does not clearly describe his
position in terms of "effects" as Armstrong interprets him
as doing. In the passage Armstrong cites, Smart speaks in
terms of "something going on" and in terms of resemblance.
To justify his interpretation, Armstrong must interpret
"something going on" or resemblance as meaning "effect";
otherwise there is no clear statement by Smart that this is
what he means.)
But omitting "responses" from the analysis of mental
states is crucial to Armstrong since it prevents us from
giving a satisfactory analysis of such mental events as
95
intentions and perceptions. Armstrong argues that such men
tal items require mention of behavior as well as stimulus,
output as well as input.
Now if we consider some other mental processes it
is at once clear that this sort of analysis solely in
terms of the effects of a stimulus can have no hope of
success. Suppose I form the intention to go out and get a
drink. There may well be no typical physical stiuations
which have the effect of creating this state in me. The
account of intentions must clearly proceed instead in
terms of the behaviour that such an intention initiates.
The intention is an inner cause of a certain sort of
response, not the inner effect of a certain sort of stimu
lus. Of course, the intention is an effect of certain
causes, but it cannot be defined in terms of these
causes.10
And this applies to perceptions as well. They, too, require
mention of behaviour or response in addition to mentions of
causal stimulus. Not only must we analyze perception in
terms of standard perceptual conditions and stimuli, we must
also, if we are to be convinced a person has perceived any
thing, determine if he can "discriminate in his behavior
between classes of objects."11
As Anthony Kenny remarks in his Action, Emotion and Will,
we pick a man's lack of perceptual powers by a certain
inefficiency in conduct. So, even in such a case as per
ception, reference to certain sorts of responses for which
the perception gives us a capacity is at least as impor
tant for elucidating the concept as reference to certain
sorts of stimuli.12
Such considerations lead Armstrong to reject Smart's
and Place's analysis as not going far enough, and, conse
quently, his definition of a mental state will have packed
into it the notion of being a cause, not just an effect.
1'Armstrong, p. 80. --Ibid., p. 81. 12Ibid,
The difficulties in Place's and Smart's position incline
me to look to the response rather than the stimulus in
seeking a general account of the mental concepts. The
concept of a mental state is primarily the concept of a
state of the person apt for bringing about a certain sort
of behaviour. Sacrificing all accuracy for brevity we can
say that, although mind is not behaviour, it is the cause
of behaviour. In the case of some mental states only they
are also states of the person apt for being brought about
by a certain sort of stimulus. But this latter formula is
a secondary one.1*
Here, then, is Armstrong's attempt to remedy what he takes
to be defective in the work of Place and Smart, with the
result that a mental state will be seen not only as the
result of stimuli of various sorts, but as a cause of behav
iour as well. And with this emphasis on the behaviour side
of our mental life, Armstrong is forced to carefully dis
tinguish his views from Behaviourism.
3
The behaviourism Armstrong disavows is the brand that
identifies mind with behavior. This identification is un
satisfactory to Armstrong simply because it does not allow
the mind to be the cause of behaviour, a fact which is at
the heart of his claim. A thing cannot be the cause of it
self, so that if mind is behaviour, it can no longer serve
as the cause of behavior.
The issue between Armstrong and the Behaviourist comes
into focus over the nature of dispositions.
Admittedly, there is one great divergence from Behaviour
ism: the mind is not to be identified with behaviour, but
1 *Ihid>., p. 82.
97
only with the inner principle of behaviour. But, in eluc
idating our formula, there has been talk about tendencies
to initiate, and capacities for, behaviour. And are not
these perilously close to the Behaviourist's dispositions?
There is some force in this. In talking about disposi
tions to behave Behaviourism did come quite close to the
version of the Central-state theory being defended here,
. . . But Behaviourism and the Central-state theory still
remain deeply at odds about the way dispositions are to be
conceived. * *
Armstrong sees the Behaviourist as committed to a "Phe-
nomenalist" account of dispositions whereas he is a "Real
ist." A "Realist" believes that if we truly ascribe a dis
position to an object, we are asserting there to be some
non-dispositional, "categorical" state of that object that
"stands behind" the dispositional manifestations and is
responsible for them. For example, we might cite some
molecular structure of sugar as the "categorical state"
causally responsible for the sugar's soluability.
To this we may oppose what may be called a Realist account
of dispositions. According to the Realist view, to speak
of an object's having a dispositional property entails
that the object is in some non-dispositional state or that
it has some property (there exists a "categorical basis")
which is responsible for the object manifesting certain
behaviour in certain circumstances, manifestations whose
nature makes the dispositional property the particular
dispositional property it is.15
A phenomenalist denies what the Realist asserts. He
denies that there exists a "categorical basis" for disposi
tions; the dispositions may themselves be ultimate facts of
the objects of which they are dispositions. Armstrong
quotes H. H. Price as representative of this view.
1 ‘ 'Ibid. . p. 85. 15Ibid., p. 86.
There is no a priori necessity for supposing that all dis
positional properties must have a "categorical basis." In
particular, there may be mental dispositions which are
ultimate.1
It is clear why Armstrong must be a Realist about dis
positions. If mind is to be a cause of behavior, and more
specifically, if dispositions are to be causes of behaviour,
then they must be identified as actual states "standing
behind" and responsible for behaviour. Under the phenomen-
alist account this could never be.
The tremendous difference between this and the "Phenomen-
talist" account of disposition emerges when we consider
that, on this "Realist" view of dispositions, we can think
of them as causes or causal factors. On the Phenomenalist
view, dispositions cannot be causes.17
Consequently, it is important that Armstrong clearly demon
strates why we must be Realist and not Phenomenalists about
dispositions, something I think he has failed to do. Two
points need be made about the distinction between a Realist
and Phenomenalist as Armstrong construes it. First, he is
never clear about whether the disposition is to be identi
fied with the categorical basis or whether it is to be iden
tified with the dispositional manifestations. For instance,
he writes that "having a dispositional property entails that
the object is in some non-dispositional state"18 or again,
. . . the Realist view asserts, in asserting that a cer
tain piece of glass is brittle, for instance, we are ipso
facto asserting that it is in a certain non-dispositional
state which disposes it to shatter and fly apart in a side
variety of circumstances.18
16 Ibid.
18Ibid., p. 86.
17Ibid., p. 88.
18Ibid.
99
Here the categorical state clearly is not the disposition
but is its cause; the disposition apparently is to be iden
tified with the effects of his cause. But later in the dis
cussion, Armstrong identifies the disposition with the "ca
tegorical basis." "But if brittleness can be identified
with an actual state of the glass, then we can think of it
as a cause, or, more vaguely, a causal factor, in the pro
cess that brings about breaking."20 Here, as opposed to the
earlier quotes, the disposition is the categorical basis and
the effects are its manifestations. Clearly, this is the
thrust of Armstrong's positions. If dispositions are to be
seen as causes, then they must be seen as the "categorical
basis" and not simply as the manifestations. And Armstrong
does assert this. "It is simply that the states are identi
fied in terms of their manifestations in suitable condi
tions, rather than in terms of their intrinsic nature."21
Nevertheless, if the disposition is to be identified with
some actual state of the object, it is confusing to call
such a state a "non-dispositional state."
Fortunately, Armstrong's argument does not rest on this
point, though it would be helpful were he consistent. What
Armstrong does hope to show in his argument is that a Phe
nomenalist has no justification for ascribing dispositions
to objects when those objects are not presently manifesting
the dispositions. If he is right, the Phenomenalist could
20Ibid., p. 88. 21Ibid.
not assert of a rubber band which had stretched on inch at
Tx, under force F, that it would behave similarly at T2
under the same force, F. On the other hand, a Realist can
clearly assert that the band will react the same way simply
because there is no reason to doubt that the band's "cate
gorical properties" had remained the same. And, since those
properties had remained, and since there is a causal rela
tionship between the "Categorical properties" and their
manifestations, we are fully warranted in asserting, as a
matter of "physical necessity," that the band will repeat
its previous behavior.
But a phenomenalist is prohibited from saying this,
according to Armstrong. The only possible justification a
Phenomenalist could have for such an assertion is "that
numerically the same band behaved in this way on other occa
sions."22 But "What," asks Armstrong, "is the magic in
numerical identity?" Might not a thing change it's proper
ties? And if so, what assurance is there that a thing's
dispositional properties might not change? The Phenomenal
ist cannot reply that since we have every reason to believe
the "categorical properties" have remained the same we can
then safely assume the dispositional properties will remain
unchanged as well. He is barred from such a reply because
he denies any "necessary relationship" between the "catego
rical basis" and the dispositional manifestations. Since
2 2Ibid., p. 87.
there is only a contingent relationship between the latter
two, one could never be justified in asserting a continuance
of that relationship. How could one show a continued con
tingent relationship between "categorical properties" and
"unfulfilled possibilities?" Since this is an impossibil
ity, Armstrong concludes that the Phenomenalist about dispo
sitions must remain a skeptic about dispositions except at
those times when the disposition is manifesting itself. The
main points in this argument are contained in the following
passage.
But what answer can the Phenomenalist about dispositions
give? For him, a disposition does not entail the exis
tence of a categorical state. The only reason he can give
for saying that the band would have stretched one inch
under force F at T1( is that numerically the same band
behaved in this way on other occasions. But now we may
ask the Phenomenalist "What is the magic in numerical
identity?" A thing can change its properties over a
period of time. Why should it not change its disposi
tional properties? How does the Phenomenalist know what
the band's dispositional properties are at Tx? He may
reply "We have every reason to think that the relevant
categorical properties of the object are unchanged at T1#
so we have every reason to think that the dispositional
properties are unchanged." But since he has asserted that
the connection between categorical properties and disposi
tional property is not a necessary, he can only be arguing
that there is a contingent connection between categorical
properties and the fact that the band has that disposi
tional property at Tj. But how could one ever establish
a contingent connection between categorical properties and
unfulfilled possibilities? It is not as if one could
observe the unfulfilled possibilities independently, in
order to see how they are correlated with the categorical
properties! It seems that the Phenomenalist about dispo
sitions will be reduced to utter scepticism about disposi
tions, except on occasions that they are actually mani
fested. 2 *
2 3Ibid.
But if this argument is successful in discrediting the
Phenomenalist, there seems no good reason why it may not
also be turned on Armstrong's Realist account as well. That
is, the argument proves too much: the principle of identity
and change which Armstrong uses against the Phenomenalist is
a two-edged sword; it affects his belief that dispositions
are both grounded in actual states of objects and have a
"physically necessary" relationship with their manifesta
tions .
Armstrong believes that as long as we have reason to
believe the categorical properties have remained unchanged,
we have every reason to believe the dispositional properties
have remained unchanged as well. Hence, we can conclude
that the object will manifest the same kind of behaviour in
relevantly similar circumstances. Crucial, then, is the
identity of the "categorical properties" through time. But
it is just such identity that Armstrong questions in criti
cizing the Phenomenalist. "What," he asks, "is the magic of
numerical identity?" The question could well be asked of
him. His point is that an object may change it's proper
ties, including its dispositional properties. But if he
admits this, he must be admitting that an object may change
its categorical properties as well, a thing's dispositional
properties with an actual state, "categorical basis" or
"categorical property." But if this is so, he too must be a
skeptic about dispositions, since he himself writes: "Why
should not an object change its dispositional properties?" 2 * *
Hence, I can see no reason why his criticism of the Phenom
enalist does not apply equally to his Realism. As important
as this issue is to Armstrong, it seems clear that he has
failed to establish the view he thinks necessary in showing
that dispositions may be causes in the mental life.
4
By way of summary, we see that Armstrong asserts a con
tingent identity between mind and brain. Since the identity
is contingent, Armstrong must give "logically independent"
descriptions of both mind and brain without begging the
questions about the nature of mind or introducing any
uniquely mental properties. Armstrong does this by giving a
"topic neutral" analysis of mind. Introspection reveals
nothing about the "intrinsic nature" of the mind via mental
states; and since we are aware of no properties of mental
states, we can hardly be said to be aware of any uniquely
mental properties. Further, by stripping all properties
from mental states (or the mind), Armstrong avoids begging
the question about the ultimate nature of mind. Mind's
nature remains an open question to be solved by scientific
investigation.
So far, Armstrong's position is similar to Place's and
Smart's. There is a fundamental divergence, however.
2 ‘ ‘Ibid.
104
Whereas Smart and Place emphasize the stimulus side of our
mental life, Armstrong stresses both the stimulus and the
response. This is expressed in his definition of a mental
state— the controlling definition in his work— as a state
apt for the bringing about (causing) of behaviour. The
inclusion of the response side of our mental life allows
Armstrong to give a more thorough account of mental events,
e.g., intentions and perceptions.
But Armstrong's stress on the response, and hence
behavioral, side of our mental life does not make him a
Behaviorist. Indeed, he is anxious to repudiate Behaviour
ism since the latter does not allow mind to stand as a cause
of behavior. This is clear when one looks at dispositions.
Armstrong construes dispositions as actual states of
objects, and hence, they can be causes, e.g., kindness could
be identified with an actual state of one's brain and as
such could stand as a cause of kindly behavior. A Behav
iorist must deny this since he cannot allow any "inner work
ings" as somehow standing behind and causing behavior. All
there is, according to the brand of Behaviourism Armstrong
considers, is the behavior. And, since dispositions as con
strued by Armstrong would be just such an "inner working,"
the Behaviorist must opt for a different analysis of dispo
sitions. Hence Armstrong's disavowal of Behaviorism.
In light of all this it may be clear why Armstrong sees
as the best model for understanding his claim, the statement
105
"The gene is the DNA molecule."
At this point we see that the statement "The gene is the
DNA molecule" provides a very good model for many features
of the statement "The mind is the brain." (I am greatly
indebted to Brian Medlin for this very important model.)
The concept of the gene, when it was introduced into biol
ogy as a result of Mendel's work, was the concept of a
factor in the person or animal apt for the production of
certain characteristics in that person or animal. The
question then arose what, in fact, the gene was. All
sorts of answers were possible. For instance, the gene
might have been an immaterial principle which somehow
brought it about that my eyes are the colour they are. In
fact, however, biologists have concluded that there is
sufficient evidence to identify that which is apt for the
production of hereditary characteristics as the substance
to be found at the centre of cells: deoxy-ribo-nucleic
acid. This identification is a theoretical one. Nobody
has directly observed, or could ever hope to observe in
practice, the details of the causal chain from DNA mole
cule to the colouring of the eye. But the identification
is sufficiently certain.25
Hopefully, we now have a general understanding of Arm
strong's program. It is now time to critically examine his
crucial claims in some detail.
2 5Ibid., p. 90.
CHAPTER IV
ARMSTRONG'S ANALYSIS OF SELF-AWARENESS
We saw in the last chapter how the admitted contingency
of Armstrong's identity claim forces him to produce logic
ally independent descriptions of the two terms of the iden
tity. The one term, brain, is ultimately defined osten-
sively and mind (or mental states) is defined in terms of
what is given to us in acts of self-awareness. But, since
the description of mind or mental states presents Armstrong
with some of his most difficult problems, his analysis of
self-awareness is obviously most important. Hence, we will
now examine how Armstrong actually does analyze our acts of
awareness directed toward our mental states, or as it is
traditiaonally discussed, our acts of "self-awareness."
With this in mind, it is important first to notice that
Armstrong clearly asserts that we know nothing about the
"intrinsic nature" of our mental states: they remain "topic
neutral." In our common-sense grasp mental states remain
vague and indeterminate, their "whatness" something to be
revealed only by the further workings and discoveries of
science:—
As a first approximation we can say that what we mean when
we talk about the mind, or about particular mental pro
cesses, is nothing but the effect within a man of certain
106
107
stimuli, and the cause within a man of certain responses.
The intrinsic nature of these effects and causes is not
something that is involved in the concept of mind or the
particular mental concepts. The concept of a mental state
is the concept of that, whatever it may turn out to be,
which is brought about in a man by certain stimuli and
which in turn brings about certain responses. What it is
in its own nature is something for science to discover.1
However, the fact that we know nothing about the "intrinsic
nature" (whatever that means) of our mental states does not,
according to Armstrong, preclude us from knowing something
about them. To the contrary, he holds that whether an act
of awareness is directed toward one of our mental states, we
are aware of at least one of three items. First, we may be
aware of a disposition to cause future behavior;2 second, we
may be aware of an occurrent causing of ongoing behavior;3
and, third, we may be aware of a resemblance between an
occurrent mental state and seme other existent or nonexis
tent mental state (s).1 * But the only reason which Armstrong
advances in support of these claims comes in the form of an
analogy with sense perception— specifically touch perception
— wherein we are supposedly aware of the same three types of
components.5
In this chapter I shall argue that Armstrong's analogy
with touch does not hold up, and that at best we might be
aware of resemblances via touch, but surely not of the
causal relation or of dispositions to cause. This will not
1Ibid., pp. 78-79. 2Ibid., p. 82.
3Ibid.. p. 96. "ibid.. p. 83. 5Ibid.. p. 97.
show that we might not be aware of such items in our mental
life; but it will collapse the only support which Armstrong
offers for his position, leaving only his (surely counter
intuitive) assertions concerning the content of our aware
ness of our mental states. In developing this line of
reasoning, it will be helpful to divide the discussion into
three sections: a presentation of what Armstrong thinks is
given in acts of awareness directed toward our mental
states; a presentation of the analogy Armstrong offers to
support such claims, and a criticism of the analogy.
2
As noted, even though Armstrong asserts that we know
nothing about the "intrinsic nature" of our mental states,
he does allow and insist that we do know something about
them. Mental states do have properties or relations; and
when one portion of the brain is "aware" of another portion
(both being identical with some mental state), the former
brain slice is "aware" precisely of certain of those partic
ular properties or relations. First, on Armstrong's ac
count, the brain state standing as the object of awareness
might be grasped as an instantiation of the dispositional
property of "aptness to cause."6 The aptness here is an
aptness to cause certain sorts of behavior. And, taking
6Ibid., p. 82.
109
"aptness" in its ordinary sense, a thing is apt to do some
thing when it has a tendency or a proclivity to behave in a
certain way under certain specified conditions. Thus, when
the brain slice identical with an act of self-awareness is
"directed" toward some other slice of the brain identical
with, for instance, the seeing of one's dog, the former
brain slice is "aware" of the tendency, proclivity or dispo
sition of the latter to cause certain sorts of behavior
under certain designated conditions.
Two points in this account need stressing. First, we
are "directly" aware of such dispositions; that is, the
brain is "directly" aware of them. This is in keeping with
Armstrong's statement that "we have direct awareness of
mental states."7 Second, when the brain is "directly aware"
of its dispositions, it is not "aware" of any process cur
rently going on. Dispositional states are not occurrent
states; they are states actualized only under the right set
of circumstances. The implication from Armstrong's defini
tion of "mental state” is that when the brain is aware of
one of its states, via the dispositional content of that
state, it is not aware of what is currently happening, but
of what would happen if certain conditions came to pass.
The brain, in these cases, then, is "aware" of what would
happen if other events, which are not occurring, were occur
ring. But if this counter-factual analysis strikes one as
7Ibid., p. 93.
110
being at odds with the nature of our mental states, that is,
if one wants to argue that when we are aware of our mental
states we are aware of something "current," Armstrong allows
for cases of this also.
In addition to direct awareness of dispositions, Arm
strong allows that the brain may also be aware of some
occurrent state of itself which is actually causing bodily
behavior.8 Though this does not, as in the case of dispo
sitions, follow from Armstrong's definition of "mental
state," he allows for it in numerous passages.
So when I acquire by introspection the information that,
for example, I am sad now or that I have a certain sort of
perception now, this information is information about cer
tain of my behavior producing or potentially behaviour
producing states.
Or again, "By inner sense we become aware of current happen
ings in our own mind."* Under this possibility, then, when
one brain slice is "aware" of another brain slice, the
former is "aware" of the actual causal relation which the
latter presently bears to some on-going activity. Thus,
since the brain can be "directly aware" of the causal rela
tion between itself and some behaviour, we have, so far, at
least two things which may be given in acts of awareness
"directed" toward mental states. But dispositions and
causal relations are not the only features that may be given
to the brain when it is directly aware of its own states.
The relation of resemblance may also be given.
8Ibid. . p. 96. *IJ2i£., P* 95.
Ill
In some cases, indeed, it will emerge that certain sorts
of mental states can only be described in terms of their
resemblance to other mental states that stand in causal
relations to behaviour is very indirect indeed.19
Thus, the brain may also be "aware" of the resemblance one
arrangement of atoms— which a brain state is— has to another
arrangement of atoms, and this resemblance can be described.
It is clear, then, that at least two relations and one dis
positional property may be disclosed in our awareness of our
mental states.
Clearly Armstrong is quite aware of what his position
amounts to:
For consider what we would be introspectively aware of.
(1) we would be directly aware of an extraordinarily
abstract, and purely relational, state of affairs. We
would be aware that something of whose non-relational
properties we had no direct awareness at all was operating
to produce certain behaviour. (2) The awareness would be
a direct awareness of causes: a direct awareness that
certain behaviour was being produced. Indeed, in some
cases, the direct awareness would be a direct awareness of
the mere possibility of the behaviour being produced if
circumstances were different. It would be a direct aware
ness of counter factual truths.11
3
At this point, one must look for the reasons Armstrong
advances for supposing he is right about our self-awareness,
and it is here that his analogy between sense perception and
"inner sense" comes into play. Armstrong compares ordinary
sense perception and "inner sense.” In the former we are
aware of what goes on "outside" us, whereas in the latter we
10Ifcid., p. 83. 11 Ibid.. p. 96.
are made aware of what goes on "inside" us.
By sense perception we become ».ware of current physical
happenings in our environment and our body. By inner
sense we become aware of current happenings in our own
mind.12
Again,
Bodily perception, indeed, serves as an excellent model
with which to grasp the nature of introspection, for it
has a further important resemblance to "inner sense." In
introspection we are aware only of states of our own
mind, not of other people's minds. In bodily perception
we are aware only of states of our own body, and not of
other people's bodies.11
Thus, to render plausible his claims about the intuitive
content of our mental states, Armstrong argues that in one
species of sense perception, touch, we may be, and often
are, aware of the same three sorts of components as present
themselves to "inner sense." That is, Armstrong will argue
that through the sense of touch we are made aware of dispo
sitions to cause, causal relations, and resemblances, just
as we (alledgedly) are whenever we utilize our faculty of
"inner sense."
When philosophers think of perception they are all too
liable to think of vision. But if we will turn from
vision to touch we shall find, to our surprise, that a
model is available which largely duplicates the features
of introspective awareness as conceived by a Causal analy
sis of mental states. The case in point is the perception
[by touch] of pressure.11* [insertion mine]
It should be stressed that this analogy with touch per
ception is the only reason which Armstrong offers to sub
12Ibid., p. 95.
1 * * Ibid., pp. 96-97.
1sIbid., p. 96.
113
stantiate the above claims about what is given to "inner
sense." If the analogy fails, then he is left simply with
his mere assertions that in "inner sense" relations and a
dispositional property are divulged. On the other hand,
were the analogy to hold, what is given to the sense of
touch as described by Armstrong would not prove that the
brain is actually aware of itself in the manner in question,
but only that some type of perception of causality, disposi
tions, and resemblances is possible.
4
The model Armstrong constructs to depict the alleged
similarity between "inner sense" and touch perception
focuses on the case of what is given to us when some object
presses on the small of our back.
Suppose I feel a pressure in the small of my back. What
am I aware of? It may be that I am aware of no more than
this: something I know not what is pressing upon my back.
I might say it was something material, but what is a mate
rial object in this context except "that which is capable
of exerting pressure?" I might not even know whether it
was something solid, something liquid (such as a jet of
water), or something gaseous (jet of air). My awareness
of "something which has the relation to me of pressing on
me." Here is a perceptual parallel to the bastract and
relational awareness that is being attributed to "inner
sense."15
Now since Armstrong holds that "pressing on me," or "pres
sure," is a "causal notion," in such instances as he
describes we must be directly aware of a causal relation.
Of course, if the analogy between "inner sense: and touch
15Ibid., p. 97.
114
perception is to be complete, we must also be directly
aware, through touch of dispositions and of resemblances
among various instances of "pressure."
Our first question, then, must be whether it really is
the case that when we touch or are touched by some object,
what we directly perceive is a causal relation. Also, we
must ask whether we are, at least sometimes, presented with
dispositions to cause, such that the immediate content of
our perception consists, primarily, in the relation between
a tendency and unfulfilled possibilities. Finally, is the
non-inferential data of such a perception, in any instance,
the resemblance between it and some other perception. Arm
strong must make a case for each claim if he is bo make his
analogy stick and thereby give us some reason to suppose
that he is even possibly right in his analysis of our appre
hension of the content of our mental states.
To see if Armstrong does make his case, we must closely
examine the analogy which he has chosen. The case he con
structs is designed specifically to isolate the causal rela
tion as an object of direct awareness. Armstrong argues, in
effect, that if the object pressing on the small of our back
is given in none of its properties, then, if we are aware of
something, we must be aware simply of the pressing itself.
But must this be the conclusion which we draw, even from a
case so deliberately devised to force such a conclusion? Do
we have to reify "pressure" in such a way that instead of
i
S '
115
feeling the object doing the pressing we feel only the
pressing? There certainly seems no necessity in so doing.
On the contrary, there is open to us an alternative descrip
tion of such cases, which is not only faithful to the facts,
but which does not force us into the uncomfortable position
of having to argue that we can touch a causal relation.
This alternative becomes clear when we consider how we de
scribe the various other ways objects come into contact with
our bodies. For instance, we resist concretizing "resting"
when some object rests on us. We do not conclude that we
feel the "resting" as something above and beyond feeling the
object that is resting on us. What would it be like to feel
the "resting" as something other and separate from feeling
the object that is resting on us?
The same applies to the various other ways an object
can come into contact with us. Do we feel the "caroming" as
something other than feeling the object that just "struck"
us? Further, objects "jab" us, or "stab" us, or "poke" us.
Are we tempted to make "jabs," "pokes," or "stabs" the
objects of direct awareness? Are they what we feel when
some object "strikes" us, or are such terms used to describe
the different ways objects can strike us? I would suppose
the latter. But if this is the case with "jabbing," et al.,
why should it be different with "pressing?" "Pressing" is
just another way an object may be in contact with us. Why
must we complicate the issue by saying that we feel both the
116
object and the "pressing," or both the object and the "pok
ing," the "stabbing," or the "jabbing?"
However, there is no a priori reason why we cannot feel
a causal relation; the case is surely to be decided by see
ing what exactly is felt when some object presses on the
small on one's back with just the right degree of force and
with no apparent qualities. And for my part, it has never
been my experience to feel the causal relation Armstrong
speaks of. When objects come into contact with my back,
there is the awareness of the object (and the object is
usually, if not always, given as something hard, soft,
pointed, etc.), and sometimes awareness of the depression of
the skin at the point of contact. Saying that the object is
"pressing on me" is not to refer to some further tactual
sensation; it is one of many ways of describing the total
event, and is only one in a family of terms used for such
description. The "pressing" is not something I feel in
addition to feeling the object. Of course, Armstrong might
feel the causal relation itself in conjunction with feeling
the object, or he might feel the causal relation by itself.
But if he is making this claim, he owes us some description
of what the causal relation feels like. We have numerous
ways of describing the way objects feel and relate to us,
but where would one begin describing how a causal relation
would feel to the touch? I suspect that this inability to
describe a causal relation rests on the fact that causal
117
relations are simply not felt. Armstrong apparently has
never felt one either since he gives absolutely no descrip
tion or information on what it is like to do so.
But even if these considerations are without weight,
and we are directly aware of actual causal relations through
touch, Armstrong must still show how it is that in touch we
are directly aware of causal dispositions. This is in keep
ing with his definition of mental states as apt for the pro
duction of certain sorts of behavior. Since we have direct
access to our mental states, we must also have direct access
to "aptnesses" or dispositions. Thus, before the analogy is
complete between "inner sense" and sense perception, Arm
strong must show how we are directly aware, through touch,
of dispositions or "aptnesses to cause." But that we are so
aware seems inconsistent both with the nature of disposi
tions and with how we come to know or become aware of dispo
sitions .
A disposition is a feature of an object such that if
certain conditions prevail, the object having the disposi
tional feature will undergo a change. Thus, sugar has the
dispositional feature of being soluable. Sugar will dis
solve when placed in liquid. It differs in this respect
from granite. Under normal conditions, sugar does not dis
solve— though it could— and hence, we have the distinction
between the occurrent states and the dispositional states of
objects. Even though a dispositional state may be grounded
118
in "categorical properties" of objects,16 this does not mean
that the dispositional state is made manifest under all con
ditions. In fact, it is not. Only under specific condi
tions do dispositional features make themselves known to
suitably placed observers. Thus, only under certain
conditions— for instance the throwing of a large rock— is
the dispositional feature of brittleness in window glass
made known to onlookers.
But not only are dispositional states or features not
continuously apparent, they have the additional feature of
not being apparent upon direct inspection. This surely is
one of Hume's points. We only find out the dispositional
features of things after having had experience of how the
objects in question behave under various circumstances.
That sugar is soluable is not apparent upon looking at sam
ples . That window glass is brittle is known only after see
ing it shatter under certain conditions. And clearly we see
only the shattering, not the disposition to shatter. Simi
larly with whiskey. Whiskey is intoxicating to be sure, but
the intoxicating properties are not given upon first percep
tual contact, and are never immediately or directly per
ceived. Dispositional features or states are not, then, in
the same category with "sensible" properties. It does not
seem that we see, hear, taste, smell, or feel them, ever.
But if we don't perceive dispositional properties, we obvi
ously don't immediately perceive them. We learn about them
16Ibid., pp. 85-88.
f
i
119
only by familiarizing ourselves with the reactions of
objects to various conditions. But is this "learned" char
acter of dispositions compatible with the case Armstrong
presents us?
Two points need to be emphasized with regard to the
"learned" character of dispositions. First, Armstrong must
argue that we are immediately or directly aware of disposi
tions. But as has been pointed out, this is incompatible
with the nature of dispositions, which we learn about by
perceiving certain actual events which are not themselves
identical with the dispositions. Second, Armstrong argues
that we have no knowledge of the "intrinsic nature" of the
object pressing on our backs. It remains a "something I
know not what" and stands decidedly in the background, leav
ing the causal relation as the dominant perceptual feature.
But if the object remains a "something I know not what,"
what sense is there in saying we are immediately aware of a
dispositional property or feature of the object? As Arm
strong constructs the case, the object pressing on my back
is itself not given as either a physical object or a jet of
water or a blast of air. We do not know what it is. How,
then, can we determine what its dispositional properties
are? If dispositional properties require familiarity with
certain actual events before we come to know them, how can
we immediately and directly know them? At best, what we can
do is recognize some feature of the object, for instance its
hardness, and then conclude what the object might do to us
were it to strike us under different circumstances. But
this would require some awareness of the object itself—
knowledge of its "intrinsic" features, properties or
"nature"— something Armstrong will not allow. In addition,
knowledge of dispositions can never be part of what is
immediately given; it is always inferred from the given. We
simply do not immediately perceive dispositional properties.
Thus, in light of the nature both of dispositional proper
ties and of how we come to know them, it doesn't appear that
we could know or be aware of them in the manner Armstrong
requires: immediately, directly or non-inferentially.
But Armstrong need not be arguing that we are immedi
ately aware of a disposition of the object; he might be
arguing that we are aware of a dispositional property of the
causal relation itself.
In this case we can say that we are aware that the object
is exerting a force apt for the bringing about of motion
in our body, although it is not actually bringing about
such motion.17
One might interpret this passage as asserting that it is the
relation itself that is apt for the bringing about of alter
ations in one's body. If this is what Armstrong has in
mind, then it will be the causal relation and not the object
that bears the dispositional property. Further, given Arm
strong's "realist" account of dispositions, it will be the
17lMd. / p. 98.
— 121
causal relation that will have the "categorical basis" for
the dispositional property, that is, the "non-dispositional
state . . . which is responsible for the object manifesting
certain behaviour in certain circumstances . . ,"18 Under
this interpretation, then, we have a case where we are not
only immediately aware of (feel?) a causal relation, but are
also immediately aware of (feel?) a dispositional property
of that relation. And if this is what Armstrong has in
mind, we are again required to be directly aware of a dis
positional state, thus disposing this position to the same
criticisms as stated above.
Possibly to avoid such criticisms, Armstrong presents a
case wherein some force is exerted against us, and which we
resist.19 Given in such a situation, Armstrong argues, is
the awareness of what the force would "do" were we to give
in to it. But if this knowledge is given in such circum
stances, it surely is not immediately given. It would seem
that our knowledge of what the force would "do" is depen
dent, at least, upon the strength of the force. If this is
true, then we infer from the strength of the force what
would happen were we to give in to it. Force by itself does
not yield uniform effects in the future; the degree of force
helps determine what the affects will be. So it would seem
that even if we are directly aware of the causal relation,
which I have argued is doubtful, what the relation might or
19Ibid., p. 86. 19Ibid., p. 98.
122
will "do" in the future seems to be inferred. And if this
is the case, then Armstrong's analogy surely fails in this
respect.
Armstrong does not attempt to make a case for the third
area of comparison between "inner sense" and sense percep
tion; that is, for our ability to "pick out" the resemblance
between the causal relation "pressing on me" and some other
causal relation. Further, he does not attempt to describe
how, or in what respects, one causal relation might resemble
another, even though he does assert this possibility regard
ing mental states. He argues that when we are directly
aware of some of our mental states, the content of our acts
of awareness might be the resemblance between (among) the
mental state being "picked out" and some other mental
state(s), which resemblance can be described.20 Conse
quently, if the analogy with touch perception is to be com
plete (and if, in fact, Armstrong wants to include resem
blance in the analogy), we must be able, at least in some
instances, to "pick out" resemblances among different dispo
sitional states of objects.
My response to this possibility is twofold. First,
with respect to perceiving resemblances, the problem of
"immediacy" again presents a difficulty for Armstrong. If
we do not perceive dispositions, much less immediately per
ceive them, how can it be held that we perceive resemblances
20Ibid.. p. 83.
t
123
among them? Second, if I am right about not perceiving the
actual causal relation via touch, then the same question
needs answering regarding resemblances among different
causal instances. It hardly makes sense to say that we per
ceive resemblances where we do not perceive the objects sup
posedly resembling each other. Since we do not perceive
dispositions, and since it is doubtful that we perceive the
causal relation, it is certain that we don't perceive resem
blances among the former and unlikely that we perceive them
among the latter. This is not to argue that we might not
be able to perceive, immediately, resemblances with respect
to certain properties, e.g., sounds, melodies# shapes,
colors, etc. But in such cases there is no question about
perceiving the objects between which the resemblances hold.
At least one of the objects is perceived. But at issue here
is whether any of the objects Armstrong presents us with is
ever perceived, much less immediately perceived, i.e.,
causes and dispositions. And since it seems clear that we
do not perceive such objects, it is difficult to believe it
possible to immediately perceive resemblances among them.
5
To summarize our case against Armstrong here: he
argues that we have direct awareness of our mental states
and that what is given in these acts of awareness are "caus
ing," "aptnesses to cause," and resemblances. The mind (or
mental states) remains in the background as something vague
124
and unspecified. To substantiate his claims, Armstrong
develops an analogy with tactual perception wherein sup
posedly, we are aware of the same three components. Thus,
when we touch an object, or an object touches us, even
though the object itself remains vague and indeterminate,
the causal relation between us and the object stands out.
We are directly aware of it. I have argued, to the con
trary, that what we feel in touch perception is the object
itself, and sometimes the dislocation of the flesh at the
point of contact. We do not feel the "pressing" as some
thing above or beyond feeling the object, any more than we
feel the "resting," "supporting," "caroming," "tapping," or
"poking." These are ways of describing the nature of the
contact between the object and our body. Second, to argue
that we are directly aware of the disposition of either the
object pressing on us or of a causal relation, which Arm
strong calls "pressing," is to run counter to how we learn
about dispositions. If we know nothing about the "intrinsic
nature" of the object doing the pressing, then we cannot be
said to know what its dispositional properties might be.
Further, if Armstrong is arguing that we are directly aware
of the dispositional properties of the causal relation it
self, this knowledge is, to the contrary, inferred, and is
not an immediate data of experience. And if it is inferred,
then Armstrong's analogy in this respect would fail.
Finally, Armstrong must show that in some instance, at
125
least, we are directly aware of the resemblances among dif
ferent causal relations and between different dispositional
states. In both instances I have argued that this is doubt
ful since there is no good reason to believe we immediately
perceive either causal relations or dispositions.
In light of these considerations, it does not seem that
Armstrong has made a case for his analogy between "inner
s e n s e " and sense perceptions. This analogy is important for
Armstrong's position since it is the only reason given to
support his contentions about what might be given to us in
the acts of awareness directed toward our own mental states.
If this analogy fails, Armstrong's analysis stands as a set
of counter-intuitive claims about our mental life, i.e.,
that one slice of brain tissue is "aware" of (1) the causal
activity of another brain slice, (2) the dispositional
properties of another brain slice, and (3) the resemblance
of the latter brain slice to some other brain slices. Of
course, being counter-intuitive does not show such asser
tions to be false; but it would appear that Armstrong needs
more than his alleged analogy with touch perception to show
them to be true. Until then, they simply stand as implaus
ible assertions. And his recourse to topic-neutrality of
mental state descriptions is clearly an ad hoc device whose
only recommendation is the way in which it permits the
identity to be contingent.
CHAPTER V
THREE IMPLICATIONS OF ARMSTRONG'S
IDENTITY CLAIM
At the core of Armstrong's Central State Materialism is
the continegent identity claim that mind is identical with
brain. As innocuous as such a claim might first appear, it
is the claim itself that forces on Armstrong most of his
serious problems. For instance, in the last chapter we saw
Armstrong forced to argue that we are directly aware of
causes, dispositions to cause, and resemblances among our
mental states, all to reconcile his purported "topic neu
tral" analysis of our mental states with his "contingent-
identity" claim. But, as we will see in this chapter, such
a reconciliation is not the only problem generated by Arm
strong's claim.
There are at least three other problems that require
examination. The first concerns the non-existence of one
member of the identity claim. We seem to find Armstrong
arguing that sensations are identical with brain states,
even though sensations or "raw feels," do not exist. But
how could something non-existent be identical with something
which exists? Second, we will see that Armstrong argues, in
effect, that minds are identical with brains even though we
126
127
do not know what minds are. Thus, we will have an identity
claim wherein we are wholly ignorant of one of the two enti
ties claimed to be identical. This will bring in its wake
the possible reply by Armstrong that we are justified in the
latter claim by scientific considerations. But this, in
turn, elicits the question whether the Identity Thesis, as
construed by Armstrong, is a scientific hypothesis in the
sense that it can be settled by observation. Third, there
is a set of considerations revolving around introspection an
and the objects of introspection which, if true, would indi
cate serious defect in Armstrong's position. And, Arm
strong's inability to solve these problems strongly suggests
that his version of materialism is indefensible.
To begin with, we concentrate on Armstrong's claim that
sensations are identical with some brain state. Much to-do
is made about sensations whenever the Identity Theory is
discussed, because sensing something is nowadays thought to
be a prime example of a mental event. That is, sensing or
seeing red, feeling a pain, etc., appears to be radically
different from a firing of one's C-fibers. Were one to look
into the brain of someone in pain, or read off some graph
sheet the brain impulses of the person, it is thought that
one could never see or in any way witness the pain itself.
At best, one might see some minor movement in one section of
the brain, though this is not probable. What one would see
would be the sharp "spike" pattern on the electroencephalo
128
graph. But this is as senseless to some as would be the
question whether we can see gravity. Such things simply are
not seen, though their effects might be. So a sharp cleav
age appears between immediately felt sensations, "raw
feels," and neural events. As Feigl writes, "the crux of
the mind-body problem consists in the interpretation of the
relation between raw feels and the neural processes.1
The distinction between the two is brought out in the
following passage by John Hospers. After describing the
minute physiological alterations and movements of the hear
ing process, Hospers writes,
The entire process just described takes only a small frac
tion of a second, but now, when the auditory nerve has
carried the stimulus to the appropriate portion of the
brain, something new and different occurs: you hear a
sound, you have an auditory sensation. This is "something
new under the sun." It is something quite different from
anything that went on earlier in this brief but complex
process . . . It is an awareness, a state of conscious
ness. 2
This distinction is underlined by Wilder Penfield, a leading
neurophysiologist.
Let us consider the brain-mind relationship briefly. This
is a problem to which a psychologist must turn his atten
tion if he is concerned with psychology . . . It is a
boundary which, as some philosophers explain it, does not
exist at all. But for the neurophysiologist there is a
working boundary that does exist. Physiological methods
bring him nearer and nearer to it. But he comes to an
Herbert Feigl, "Mind-Body, Not a Pseudoproblem," in
Dimensions of Mind, Sidney Hook, ed. (New York: Collier
Books, i960) , p. 35.
2John Hospers, An Introduction to Philosophical Analy
sis, 2d ed. (Englewood Cliffs, New Jersey: Prentice Hall,
Inc.), p. 379.
s
129
impasse, and beyond that impasse no present method can
take him . . . *
But if one wanted to defend the Identity theory (some
thing, according to Penfield, that only philosophers try to
doi) one would try to show that such "raw feels" are identi
cal with some neural events, processes, states, etc. One
would try to show that there is nothing "new under the sun"
or that the "impasse" Penfield speaks of really does not
exist. This, of course, is what Armstrong seeks to do.
Consequently, he proposes a contingent identity claim of the
form A = B, where A is the "raw feel," sensation or phenom
enal data, and B is the brain state, event, or process.
What I will argue here is that under Armstrong's analysis of
sensations or raw feels, the latter turn out to be nonexis
tent entities. That is, I will argue that Armstrong reduces
sensations to perceptions and in turn defines perceptions as
"acquirings of belief" in such a way that sensations them
selves disappear from the picture altogether. Further, I
will argue that Armstrong's obvious alternative in the face
of this criticism results in a blatant circularity which
surely confuses the issue.
It would indeed be a strange kind of contingent iden
tity claim if one term of the identity ceased to exist after
the identity was established. For instance, it would not be
warmly received were one to say that "the morning star is
sWilder Penfield, p. 8.
130
identical with the evening star," and then add, "but there
really is no morning star." To say Dr. Jekyll is identical
with Mr. Hyde, but then deny that there is a Dr. Jekyll,
seems inconsistent with the claim being made. Or again, to
identify lightning with an electrical discharge does not
then imply that there is no lightning. In the same vein, it
would not do to say that occurrent sensations of pain are
identical with some brain states, and then deny that there
are any occurrent sensations of pain. Nevertheless, this is
what I suspect Armstrong's analysis comes to: and this is
clear when we consider his treatment of sensations in Sec
tion 14 of A Materialist Theory of Mind. What he appears to
do is to analyze sensations as perceptions, which in turn he
analyzes solely in terms of "acquirings of beliefs." The
sensation as we normally speak of it, the occurrent "raw
feel," the experience of pain, for instance, is simply left
out. There are, in effect, no such things.
That Armstrong does take sensations to be perceptions
is seen in his description of bodily sensations.
It will be argued in this chapter that such sensations are
nothing but bodily and tactual perceptions, using the word
"perception" in the neutral sense that is compatible with
failure to correspond to physical reality.1 * (Italics mine)
Sensations then are perceptions. But what does Armstrong
include under sensations?
If asked what should be included among bodily sensations
we might instance such things as sensations of pressure,
‘ ‘Armstrong, p. 306.
131
warmth and movement, together with pains, itches, tickles,
and tingles.®
An occurrent feeling of pain, then, is a sensation, which in
turn is a perception. The obvious question concerns what
Armstrong takes a perception to be. Given that answer, we
will be able to see what an occurrent feeling of pain, a
"raw feel," turns out to be in the final analysis for Arm
strong.
Armstrong discusses perceptions in Section 10 with this
as a brief introduction.
We come now to an account of the concept of perception.
Here I have been unable to see any way of presenting the
position except by an analysis that has two stages. In
the first stage, it is argued that an account of percep
tion can be given in terms of the acquiring of beliefs
about the physical world.6
Again,
It leads us to the view that perception is nothing but the
acquiring of true or false beliefs concerning the current
state of the organism's body and environment.7 (Italics
mine)
Thus, sensations of pain are perceptions and percep
tions are "nothing but" the acquirings of beliefs about
one's body or environment. But if perceptions are only
acquirings of belief, where is the occurrent "raw feel?"
Where is the event supposedly identical with the specified
neural event? That Armstrong does not take into account
such "raw feels" is evident from this characterization of
perception.
5Ibid. . p. 307. 6Ibid.. p. 208. 7Ibid., p. 209.
132
When it is said that somebody perceives something/ then
the mental event that takes place is simply the acquiring
of information and misinformation about the environment/
information or misinformation that may be "discounted/"
for one reason or another, by the perceiving subject.8
(Italics mine)
Acquiring information is the same for Armstrong as acquiring
beliefs ("Information and beliefs are identical."*) so we
have a picture of a sensation as "simply" the acquiring of a
belief about one's body or about one's surrounding
environment.
Feeling pain will then be simply gaining some informa
tion about one's body; there will not be that "something new
and different" that Hospers speaks of, that is, the experi
ence of pain. To be sure, Armstrong speaks of feeling
pains, seeing red, and the felt intensity of tickles. But
insofar as these events are described solely in terms of
perceptions and perceptions are described solely in terms of
acquiring beliefs, what can Armstrong mean by "feeling" a
pain? "Feeling" a pain is clearly left out. If X is "sim
ply" Y, or if X is "nothing but" Y, what can there be of X
that is not Y? Under this analysis, there simply is no room
for the "raw feels." In opposition to this analysis I would
suppose the "ordinary man"— here meaning one who is not
holding a theory about perception, etc.— thinks he is feel
ing something when his hand hurts, that what he is feeling
is a pain, that this hurts and that this hurting is not
•ibid. •ibid., p. 210.
133
simply acquiring a belief about an indeterminate disturbance
in his hand. In ordinary discourse, when one says that he
is having a pain in his tooth, he is not talking simply
about acquiring a belief, actual or potential, about the
state of his tooth. He thinks he is reporting afeeling, an
awareness of something that hurts, a condition of his tooth.
The feeling is the ground of his acquisition of a belief,
not the acquisition itself.
Armstrong's apparent reduction of sensations to acquir
ings of belief is buried under his use of "feeling" language
throughout. But what I am suggesting here is that he cannot
mean by "feeling" or sensation what is ordinarily meant by
these terms. He can't be referring to Feigl's "raw feels,"
simply because his analysis of sensation eliminates such
occurrent feelings. Armstrong analyzes sensation ("raw
feels") as perceptions and perceptions as "acquirings of
beliefs." But "acquirings of beliefs," as Armstrong dis
cusses them, do not entail and appear to exclude, having any
sensations.
Armstrong's alternatives-in light of these criticisms
seem to be two: (1) he might reply that by "acquirings of
beliefs" he simply means sensations or "raw feels," or
(2) he might reply that what he means by "acquirings of
beliefs" is that one comes to believe X on account of
some "feeling" or sensation one had. But if he means
(1) then his analysis of sensation and perception becomes
134
tautologous. That is, if acquirings of beliefs are identi
cal with sensations, and if sensations are "nothing but"
perceptions, and if perceptions are "simply" acquirings of
beliefs, then Armstrong is arguing that acquirings of
beliefs are the same as acquirings of beliefs. But if he
means (2), the notion of "acquiring a belief" would be
dependent upon, and would obviously not rule out, having
sensations. But if this is what Armstrong means, one might
wonder why all the bother. On this interpretation, sensa
tions would be perceptions and perceptions would be beliefs
gained as a result of having sensations. Sensations would
then be beliefs gained by or caused by sensations; they
would be beliefs caused by themselves. And this point is
not affected if Armstrong rejects a causal relation between
sensations and acquirings of beliefs in his definitions. No
matter what, the relation is circular, and such circularity
gets one nowhere.
In review of this first point then, it would appear
from what Armstrong writes, that he offers a reductive
analysis of sensations in that sensations are reduced or
eliminated altogether from his account of mind. But if this
is the case, Armstrong is arguing that sensations are iden
tical with some brain states, even though there really are
no sensations. But part of Armstrong's program is the
identification of sensations with neural processes of some
sort. Needless to say, if sensations turn out to be non
p -
135
existent entities, we then have a case where some non
existent entity is identical with some other existent en
tity is identical with some other existent entity. But this
seems conflicting at best. Armstrong may reject this
assessment of his position and reply either that by "acquir
ing of belief" he means sensation, in which case his analy
sis is tautologous; or he may reply that one comes to hold
the beliefs in question as the result of some sensation, in
which case his analysis is circular. In neither case has he
given an account of the relation between "raw feels" and
neural events.
2
There is yet another problem for Armstrong's contingent
identity claim, one resting upon our supposed ignorance of
our mental states. As has been pointed out, Armstrong is
careful to depict the content of our acts of awareness,
especially those directed toward our mental states, as
indefinite and obscure. For instance, Armstrong speaks of
the good biological reasons for the lack of information
afforded by introspection about our mental states.
Knowledge of the presence within us of potential causes of
behaviour is obviously valuable in the conduct of life
. . . But to know the intrinsic nature of the process, or
even where in the body the process was, would be of little
value in the state of nature.10
As with the object pressing on the small of one's back
1°Ibid., p. 99.
(discussed above), mental states (or the mind) are charac
terized as a "somethin? I know not what." But when plugged
into the identity claim, this has the effect of saying that
"something I know not what" is identical with some brain
state (and this difficulty is augmented by the fact that we
are not presently clear even about what constitutes a brain
state, or how one brain state can be clearly demarcated from
another) . This is unlike the point above about the non
existence of sensations in that mental states do exist; its
just that we do not know what they are. But then this
becomes a claim wherein A is claimed to be identical with B
even though we do not know what A is. But how can we pick
out that A is identical with B if we are ignorant about A?
This is important in light of the contingency of Armstrong's
identity claim. The truth of his claim is in the discovery;
we find out by investigation whether he is right or wrong.
But how could we know that "A = B" is a contingent truth
under the restrictions of ignorance placed on us by Arm
strong?
Armstrong would probably justify his claim by pointing
out the cumulative evidence localizing and establishing an
undeniable link between mental and cerebral activity. This
evidence ranges from the relatively common experience of
being struck on the head and the resulting loss of con
sciousness, to the systematic probing of one's brain with an
electrode and the ensuing mental states reportable by the
137
subject. Granted, our knowledge about the brain and its
workings is in relative infancy, still Armstrong would argue
that we know enough to justify our conviction of a close tie
between mind and brain. Indeed, since there is nothing in
mental experience that suggests that such experience cannot
at least be caused or in some other way connected with the
brain, when we begin investigation as to what the intrinsic
nature of mental experiences might be, it is only natural
that we look to the brain. Hence, Armstrong might reply
that even though we do not know what the A is in A = B, it
is not strange at all that we can assert that the A is
identical with something we do have some knowledge about.
Since the connection is obviously there, and since mental
states are not given in their intrinsic nature but only as
causes, potential and actual, we are fully justified in
looking to the brain for the disclosure of the nature of
mental states. Thus, Armstrong is willing to leave to the
neurophysiologist the demonstration of the truth or falsity
of his claim.
This willingness lands us squarely within the bounda
ries of the purported scientific character of Armstrong's
claim, about which Armstrong is quite clear.
It (the book) does not attempt to prove the truth of this
physicalist thesis about the mind. The proof must come,
if it does come, from science: from neurophysiology in
particular.11
llIbid., p. 2.
138
Science will also disclose the intrinsic nature of mental
states.
As a first approximation we can say that what we mean when
we talk about the mind, or about particular mental pro
cesses, is nothing but the effect within a man of certain
stimuli, and the cause within a man of certain responses.
The intrinsic nature of these effects and causes is not
something that is involved in the concept of mind or the
particular mental concepts. The concept of a mental state
is the concept of that, whatever it may turn out to be,
which is brought about in a man by certain stimuli and
which in turn brings about certain responses. What it is
in its own nature is something for science to discover.12
Again,
The question then arises "What in fact is the nature of
these inner states?" What are these inner causes like?
And here no logical analysis can help us. It is a matter
of high-level scientific speculation.13
It is clear that Armstrong sees himself as clearing the way
for science to handle the factual content of his claim.
Once he has settled whatever conceptual or logical issues
that surround his identity claim, science will be able to
proceed unimpeded by philosophical criticisms at its flanks.
Granted that this is how Armstrong views the situation,
I will argue that Armstrong expects too much from science in
this regard; there is little, beyond a correlation of
events, that science could produce as "proof" of his thesis.
There is no crucial test that might be performed that would
not only prove his thesis but disprove all other competing
theses. The situation still stands surprisingly close to
Leibniz' description of it in The Monadology.
. . . it must be confessed that perception and that which
12Ibid., p. 79. 13Ibid., p. 89.
139
depends upon it are inexplicable on mechanical grounds,
that is to say, by means of figures and motions. And sup
posing there were a machine, so constructed as to think,
feel, and have perceptions, so that one might go into it
as into a mill. That being so, we should, on examining
its interior, find only parts which work one upon another,
and never anything by which to explain a perception . .
We can, of course, inspect a brain more closely than in
Leibniz' time, though much of his point still stands. If a
neurophysiologist, such as Penfield, probes a brain and
records the electrical activity of the brain, he will be
recording "figures and motions" to quote Leibniz. What he
will not record is the "raw feel," the sensation as experi
enced by the subject. For these, the neurophysiologist will
have to await the response of the person. And one reason he
would have to await the response of the patient is that
there is nothing about brain tissue that, prima facie,
resembles a feeling of pain, or thinking, or believing, etc.
Simply by looking at the brain, without responses from the
person whose brain it is, one could never tell anything
about the mental life of the person, especially prior to the
establishment of massive correlations between states of the
brain and mental states.
What is it about the brain that would indicate that the
subject was presently thinking if the subject made no
response? Though Armstrong is not daunted by this position,
^Gottfried Wilhelm von Leibniz, The Monadology, as
found in Leibniz, Selections, Philip P. Wiener, ed. (New
York: Charles Scribner's Sons, 1951), p. 536.
140
he expresses the point well.
For even in this case a supreme physicist would need, be
side his knowledge of physics, a physiological-psycholog
ical "dictionary." By means of this dictionary, having
first worked out what the physical state of a particular
subject's brain must be at a certain time, he could then
say what the corresponding mental event in the subject's
mind must be. But the "dictionary" could never be com
piled in the first place simply on the basis of knowledge
of physics.15
Thus, in relation to Armstrong's claim about the factual
nature of his claim, there are two preliminary points to be
made. (1) First, there is nothing about the brain itself,
in the absence of previously established correlations be
tween mental activity and brain activity, which would indi
cate that it was in any way connected with the mental life.
Grey matter, by itself— even analyzed to its atomic struc
ture— simply does not "point to" mental activity (Nor for
that matter, does mental activity, on the face of it, resem
ble chemical or atomic activity.). Thus, no neurophysiolo
gists could, simply by looking at the brain confirm or dis-
confirm Armstrong's claim. The obvious corollary here is
that prior to our ability to observe a subject's brain and
infer that he is thinking about his lost dog, we must al
ready have established massive one-to-one or one-to-many
correlations between brain activity and mental activity.
But this drives an epistemological wedge between the mind
and the brain. If we have to infer, from observing a sub
ject's brain, that he is feeling a pain, then the brain
lsArmstrong, p. 50.
141
state and the feeling of pain cannot be the same thing. We
do not infer A from itself; we infer it, if we infer it at
all, from something other than itself. It would surely be
strange to say that we do infer B from A though A and B are
the same thing, or what is equally strange, were we to say
that we infer B from A though there really aren't any A's.
(2) Second, as is pointed out by Jaegwon Kim, there is no
set of observations which would confirm the identity thesis
without at the same time confirming a correlation thesis.
That is, there is no set of observations that would estab
lish the identity thesis while discontinuing all other pos
sible theses.
It is often emphasized that a particular psycho-physical
identity (e.g., pain and brain state B) is a factual
identity. From this some philosophers seem to infer that
the Identity Theory is an empirical theory refutable or
confirmable by experience. This is misleading, however.
To begin with, a particular psycho-physical identity
statement is not confirmable or refutable qua identity
statement; it is confirmable or refutable insofar as, and
only insofar as, the corresponding correlation statement
entailed by it is confirmable or refutable by observation
and experiment. There is no conceivable observation that
would confirm or refute the identity but not the associ
ated correlation.16
But if this is true, and no good reason has been
advanced by Armstrong to doubt it, then there will be no
observation data to which Armstrong (or his future scien
tist) could point to as proof of the truth of his "physical-
ist thesis about the mind.” This is, no doubt, why many
16Jaegwon Kim, "On the Psycho-Physical Identity
Theory," in Modern Materialism: Readings on Mind-Body
Identity, John O'Connor, Ed. (New ^forks Harcourt, Brace &
WorTcf," Inc., 1969), pp. 227-235.
142
identity theorists argue for their position, not on the
grounds of observational results, but on methodological con
siderations, such as simplicity and parsimony, or on grounds
such as the unity of science. But since Armstrong advances
no arguments of this kind, we will not examine them here.
The point here is that if Armstrong expects science to vin
dicate his thesis while at the same time disconfirming all
other theses, he is apparently mistaken.
3
There is a third set of considerations which might cre
ate problems for Armstrong and it revolves around the object
of introspective awareness. What, if anything for Arm
strong, are we directly aware of when we are directly aware
of our mental states? Some philosophers think that it is at
this point that the identity theory breaks down.17 Since we
are aware of something when we are aware of a mental state
and since it is clear— at least for these philosophers— that
we are not aware of anything physical, we must, it is
argued, be aware of something non-physical.
Of course, were this argument sound, Armstrong's posi
tion would be false. So, it would be helpful to see what in
fact Armstrong does say about this line of argument, and if
this is not sufficient to refute the argument, we can dis
17For instance, see Jerome A. Shaffer, Philosophy of
Mind (London: Prentice Hall International, Inc., l3rt),
p. 46.
143
cuss what he could say to meet the objection.
First, the argument rests on the principle of the sub-
stitutivity of identicals which allows us to infer from the
premise A = B and a second premise containing an occurrence
of A, a conclusion formed from the second premise by replac
ing each occurrence of A b7 B. For instance, we might argue
Argument #1
1. The Sultan of Swat is Babe Ruth.
2. The Sultan of Swat hit 60 home runs in 1927.
3. Therefore, Babe Ruth hit 60 home runs in 1927. 1, 2
Identity
Now if such inferences based on identity were always
valid, then Armstrong might face some difficulty with this
inference.
Argument #2
1. m = b, where "m" denotes some mental state and "b"
denotes some brain state.
2. I am directly aware of m.18
3. Therefore, I am directly aware of b. 1, 2 Identity
This might prove difficult for Armstrong since it does
not appear, even to him, that we ever notice any physical
feature of our brains when we are introspectively aware of
our mental states. Further, Armstrong does argue for prem
ises (1) and (2). Each mental state is identical with some
brain state, and we are directly aware of at least some of
our mental states. How, then, might Armstrong reply to this
1 8Awareness is here used to indicate a conscious state
— distinguished from a state we are in when we are con
scious.
144
argument?
It seems that there are two possible replies to this
argument. First, Armstrong might allow the inference and
argue that, contrary to appearances, we are directly aware
of at least some brain states. Second, Armstrong might deny
the inference by pointing out that substitutivity of identi
cals is a valid inference rule only in extensional lan
guages. A language rich enough to include "is aware of"
would be viewed as intentional.
As to the first alternative, Armstrong might freely
admit the inference. After all, he defines introspection as
a "self-scanning process in the brain."1* In introspection,
one portion of the brain scans another. Thus, there is evi
dence that Armstrong might not think the inference in Argu
ment #2 odd at all.
And if this further mental state, which qua mental state
is simply a state of the person apt for the production of
certain behaviour, can be contingently identified with a
state of the brain, it will be a process in which one part
of the brain scans another part of the brain.20
There is no question here about the non-materiality of
the object of awareness or of the awareness itself. Both
are physical states of the brain. We simply may be directly
aware of the brain and not know that it is the brain that we
are aware of, we might not be aware of the brain "as" the
brain.
Here the question arises, and it is the same question
1’Armstrong, p. 324. 2 0Ibid.. p. 94.
145
that will arise under the second alternative: how is it
that we are not aware of the brain as the brain? For
instance, there is nothing about thought that would indicate
that it is physical in nature. Armstrong recognizes this.
"But certainly we are not aware of the mental states as
states of the brain."21 Again, "It must be admitted, howev
er, that when we are aware of our own mental states we are
not aware of them as having any spatial properties."22
But if we are not aware of the brain as the brain, what
are we aware of it as? If we are not aware of our mental
states as having any spatial properties, then are we not
aware of them as having non-special properties? Of course,
if Armstrong admits this, he has admitted the falsity of his
view.
It may still be objected that when we are aware of our
mental states we are directly aware that they are things
that are non-spatial in nature.
This is an observation-claim, and if it is correct, of
course it refutes my view.2*
But does Armstrong have to admit this? It is not obvi
ous that he does. Armstrong might reply in one of two ways.
First, he might simply deny that we are directly aware of
any uniquely mental properties. That is, he might reply in
terms of his topic neutral analysis of our mental states.
Under this analysis, even though we are directly aware of
our mental states, we are not directly aware of them as they
really are. We are not aware of them in their intrinsic
21Ibid., p. 78. 22Ibid., p. 117. 2sIbid.
146
nature. "Mental processes have a nature of their own,
although this nature is not directly given to us."21*
Thus we are simultaneously aware of our mental states,
that is our brain, yet not aware of it as the brain. What,
then, are we aware of it as? Nothing definite. As it is
presented to us, the object of introspective awareness is
ontically neutral. This does not mean we are not directly
aware of the brain. We are, though this awareness is vague
and oblique. We do not perceive the brain as it really is
but only as a "something" which is causally active in us.
Nevertheless, it is the brain, and not something else which
we grasp in introspective awareness. This is just the point
of the topic neutral analysis. There is nothing definitive
about our mental states that would lead us to think that
they are either mental or physical. Hence, there is no need
to bring in a set of non-physical properties as the objects
of introspective awareness.
Second, Armstrong could reply in terms of alternative
languages, meaning and reference.25 That is, Armstrong
could admit that we do have two ways of talking about our
selves, one mental and the other physical. Further, he
could freely admit that these two ways of speaking do not
2* * Ibid., p. 92.
25It is clear that Armstrong does not reply in these
terms and there is even question whether he consistently
could do so; nevertheless, he might.
147
mean the same. That is, he could admit that when I say "I
have a throbbing pain," I do not mean that some brain pro
cess is occurring.
But Armstrong could easily point out that from the mere
fact of difference of meaning, it does not follow that "I
have a throbbing pain" refers to something other than "a
brain process is currently going on." Definite descriptions
often refer to one and the same thing even though they are
not synonymous. This was clearly pointed out in Frege's
example of the two expressions "the morning star" and "the
evening star." These two expressions, though they differ in
meaning, still refer to the same object, the planet Venus.
If they did mean the same, one could know that the morning
star is the same as the evening star simply by understanding
the meanings of the two expressions. But this is clearly
not the case.
Consequently, even though we have different ways of
speaking about our mental lives and even though these dif
ferent ways of speaking do not mean the same, it does not
follow from this fact alone that we are necessarily refer
ring to two separate spheres, the mental and the physical,
or that we are not, in both instances, referring to the
brain. Our mental language may just be one way of describ
ing, albeit indirectly and vaguely, how the brain appears to
us. Thus, it would seem that Armstrong could simply accept
Argument #2 maintaining that we are aware of brain states
148
though we do not thereby know or believe them to have neuro-
physiological properties; that is, we are not aware of them
as brain states.26
26In addition, it does not seem that Armstrong would
have to admit that our mental language refers at all. That
is, insofar as our mental language is thought to denote
uniquely mental properties, substances, processes, events,
etc., Armstrong might simply deny that it does so. He
might deny that he is committed, ontologically, to such
entities simply because they are not needed to make true any
of his assertions. For instance, Armstrong could follow
Quine's lead.
We can very easily involve ourselves in ontological com
mitments by saying, for example, that there is something
(bound variable) which red houses and sunsets have in com-
mon; or that there is something which is a prime number
larger than a million. But this is, essentially, the only
way we can involve ourselves in ontological commitments:
by our use of bound variables. The use of a l l e g e d names
is no criterion, for we can repudiate their namehood at
the drop of a hat unless the assumption of a corresponding
entity can be spotted in the things we affirm in terms of
bound variables . . . To be assumed as an entity is,
purely and simply, to be reckoned as the value of a vari
able . . . The variables of quantification, "something,”
"nothing," "everything," range over our whole ontology,
whatever it may be; and we are convicted of a particular
ontological presupposition if, and only if, the alleged
presupposition has to be reckoned among the entities over
which our variables range in order to render one of our
affirmations true. (Willard Van Orman Quine, From A Logi-
cal Point of View, 2d ed. rev. (New York: Harper, Row,
1963), p. 13.
Again,
To show that some given object is required in a theory,
what we have to show is no more nor less than that that
object is required, for the truth of the theory, to be
among the values over which the bound variables range.
(Willard Van Orman Quine, Ontological Relativity and Other
Essays (New York: Columbia University Press, 1969), p.
94.
Thus, Armstrong might allow for the meaning of mental-
istic discourse without countenancing mental entities. He
possibly could allow for the full range of expressions in
mental language, even if that language appears to pick out
non-spatial properties of mental states, but without admit
ting the existence of such properties. He could do this
only if the existence of such non-spatial entities, proper-
149
Of course, Armstrong does not have to admit the infer
ence in Argument #2. He does not have to admit that when
ever we are directly aware of our mental states that we are
directly aware of our brains. That is, he can opt for the
second alternative and deny that Argument #2 establishes its
conclusion. He can do this by showing that in certain "con
texts," the inference in question is not valid. For
instance, returning to Argument #1, we can produce a variant
of that argument to show that substitutivity of identicals
does not hold in all languages.
Argument #3
1. I am aware that Babe Ruth hit 60 home runs in 1927.
2. Babe Ruth is the Sultan of Swat.
3. Therefore, I am aware that the Sultan of Swat hit 60
home runs in 1927. 1, 2 Identity.
Armstrong might point out that the conclusion here does
not necessarily follow, since I may know nothing about the
Sultan of Swat and hence, may not be aware that he hit 60
ties and the like were not presupposed by some of his asser
tions or were needed for the truth of some of his asser
tions. If this were the case, then he would be implicitly
"committed" to the existence of such things; otherwise, he
does not have to reckon them to be values of any variables,
that is, he does not have to admit their existence in his
theory. Of course, if the identity theory Armstrong argues
for is true, Armstrong will not deny the existence of mental
entities. He will simply point out that these mental
entities are physical in nature. Thus, it would seem that
if Armstrong wants to admit the inference in Argument #2, he
might do so while yet deny that this forces him to admit
mental properties.
150
home runs in 1927. Thus, Armstrong may reply to Argument #2
that simply because one is aware of X, and X = Y, one cannot
conclude that one is necessarily aware of Y as well.
Contexts in which truth value is preserved through the
substitution of identicals are called referentially trans
parent. But instances where we may not always substitute
identicals without change of truth value are termed refer
entially opaque. Referential opacity refers to the inabil
ity of replacing a word or sentence with another word or
sentence which refers to the same thing and still maintain
the truth value of the original word or sentence.27
Thus, if Armstrong wanted to blunt the force of Argu
ment #2, he might argue that such epistemic words as
"aware," "know," "believe" etc., generate opaque contexts.
He could point out that it does not necessarily follow from
one's awareness of M, and M's identity with B, that one is
aware of B. Thus, if he wants to avoid the charge that
under his analysis we must be directly aware of our brains,
he could build a reply around the notion of referential
opacity.
27It should be noted that "aware of" is not clearly
opaque and not clearly transparent. For instance, there are
cases where it might be considered clear that the context
was transparent. When one is aware of the present President
of the United States, one is aware of the foremost graduate
of Whittier High School. Then again, there are cases where
it might be thought clear that "aware of" generates referen
tial opacity. One might be aware of one's destructive
inclinations yet be ignorant of the fact that such
inclinations were "really" to be identified with father
hatred.
151
In retrospect, it is not obvious that Armstrong need be
threatened by arguments centered around what is given to us
in acts of introspective awareness. He can reply to such
arguments as #2 in either me of two ways. First, he can
admit that we are directly aware of our brains even though
we are not aware of them as our brains. Then again, if he
would rather not admit direct awareness of our brains, Arm
strong could defuse "awareness arguments" in terms of refer
ential opacity. And, given such defenses, it is not obvious
that Armstrong need be seriously concerned by such argument.
4
We conclude from this chapter, then, that Armstrong's
position may be criticized in two areas and questioned in a
third: (1) Prior to an identification of sensations with
brain states, Armstrong defines sensations, or "raw feels,"
out of existence. (2) Armstrong's identity claim has the
consequence that we can know mental states are identical
with brain states though we are wholly ignorant about what
mental states are. Further, we argue that science is of
little help in clearly establishing the identity in ques
tion. (3) Armstrong's identity claim seems to imply that,
contrary to appearances, whenever we are directly aware of a
mental state we must be directly aware of some brain state.
This could, if true, prove embarrassing since even Armstrong
affirms that we are not aware of our mental states as physi
152
cal in any respect. But if we are of something when we are
aware of our mental states yet not aware of anything physi
cal, are we not then aware of something uniquely non-physi
cal? Were the answer to this affirmative, Armstrong's
materialism would have to be abandoned. But it is not obvi
ous that Armstrong must give anaffirmative answer.
First, Armstrong might freely wish to admit that we are
directly aware of our brains whenever we are aware of a
mental state. We just might not be aware of the brain as
the brain. But what then are we aware of it as? Nothing
specific. This would be in accord with Armstrong's topic
neutral analysis of our mental states. The brain would be
directly given to us in acts of introspection though it
would be given as a vague somewhat whose intrinsic proper
ties remained obscured.
Second, if Armstrong wished to deny that we are
directly aware of our brains, however oblique that awareness
might be, he might construct a reply around referential
opacity. He might wish to argue that it does not necessar
ily follow from one's direct awareness of M, and M's iden
tity with B, that one must be directly aware of B. Thus, he
might argue that even though we are directly aware of our
mental states, and even though our mental states are identi
cal with some brain state, it does not follow that we must
be directly aware of our brains.
Thus, it would seem that Armstrong is covered both
153
ways. He can either admit that we are directly aware of our
brains and employ topic neutrality to defend this admission
against obvious criticism, or he can utilize referential
opacity to refute the charge that his identity claim implies
direct awareness of the brain in the first place. Whether
either of these replies is satisfactory is a question which
must await further discussion.
CHAPTER VI
CONCLUSION
In broadest terms I have tried to demonstrate two
points: (1) that Armstrong's arguments against what he
takes to be Cartesian Dualism are not compelling, and
(2) that his positive account of the mind-body relationship
is beset with serious flaws.
Regarding the first endeavor, I have tried to demon
strate these points. First, Armstrong employs a rigid
empiricist principle— that we can have no "positive knowl
edge" of a thing without having sense experience of it,
which, though it might rule out knowing mental substance,
would also rule out knowing physical substance (as Armstrong
describes physical substance) and self-knowledge. This lat
ter point follows from Armstrong's causal and dispositional
definition of our mental states. I argue that since we
have never had sense experience of either causes or disposi
tions, it is difficult to see how we could have "positive
knowledge" of them. But this runs contrary to what Arm
strong is most anxious to assert, i.e., that we do have
knowledge of physical substance and we do have self-knowl
edge. Consequently, if I am right, Armstrong must either
forego his empiricist principle or he must abandon his
154
155
knowledge claims concerning physical substance and self-
knowledge .
Second, I argue that Armstrong's claim that Dualism
fails to account for the "unity" of man is either clearly
mistaken— a thing may have parts (mind and body) yet be
"unified"— or question begging— Armstrong cannot assume
man's "simple" nature at the outset.
Third, Armstrong's demands that the Dualist present a
principle of individuation for "spiritual" objects amenable
to sensory inspection. But I argue that this is an unfair
request given what such a principle would have to be— non-
spatial.
Finally, I have tried to show that Armstrong's ques
tions about the mechanism of mind-body interaction rely both
on a suppressed conviction that mind and bodies, being so
different, just cannot causally affect one another, and on a
refusal to clarify even the mechanism of standard physical
causation. But as is pointed out, there is no a priori rea
son why minds and bodies cannot causally interact, and the
very questions Armstrong asks about mind-body interaction
may be asked about standard physical causation.
I conclude from the first section, then, that Armstrong
offers no decisive reason or set of reasons to reject Dual
ism. Of course, Dualism may yet be false; but if it is, it
is apparently not so for the reasons Armstrong advances.
Regarding the second endeavor, I try to show that at
156
the heart of Armstrong's positive account of the mind-body
relationship rests an implicit denial of the existence of
mental events (at least sensations) at the outset, an
unfounded conviction that science will somehow vindicate his
claims, and a set of inconsistent propositions which vitiate
his position.
Thus, I argue that Armstrong implicitly rules out the
existence of sensations ("raw feels") simply by definition.
But this has the effect of denying the existence of sensa
tions prior to their identification with some brain state,
the net result being that Armstrong asserts that sensations
are identical with brain states though there are no sensa
tions .
This is accomplished by defining sensations as percep
tions and perceptions as acquirings of beliefs. Sensations
turn out for Armstrong to be "simply," "only," and "nothing
but" acquirings of beliefs either about our own bodies or
our surrounding environment. But if a sensation is "nothing
but" an acquiring of a belief, where is the "raw feel?" It
appears to be omitted, to be completely left out of the
discussion. But if this is the case, then the identity
asserted between sensations and brain states is that between
a non-existent entity and an existent one. And, given what
Armstrong says on the matter, there is no clear way to avoid
this conclusion.
Second, I point out that Armstrong argues that minds
157
are identical to brains even though we are ignorant of the
first term of the relation. That is, as Armstrong construes
the identity claim, we know nothing about the "intrinsic
nature" of minds. Nevertheless, we are required to know
enough about them to know what they are identical tol This
has the effect of arguing that A = B, though we know nothing
about A in its intrinsic nature.
Third, Armstrong might reply to this point that science
will vindicate his claim. But I argue that this confidence
is unjustified since there appears to be no crucial test
which would decide the issue in favor of an identity of
events rather than a correlation of events.
Finally, there is an implication at the heart of Arm
strong's position which some philosophers have thought fal
sifies Armstrong's version of the identity claim. That is,
since mental states are identical with brain states, and
since we are directly aware of some of our mental states, it
would seem to follow that we must be directly aware of at
least some of our brain states. But it seems obvious that
we are not directly aware of any neurophysiological process
or event or state at any time when we are introspectively
aware of one of our mental states. Hence, Armstrong's
account cannot be the true account.
Armstrong could possibly reply to these charges in at
least two ways. Whether these replies are in any way satis
factory is not here determined. But he could accept the
158
inference as it stands and argue that we are directly aware
of our brains whenever we are introspectively aware of one
of our mental states. To support this, Armstrong could
bring his topic neutral analysis to bear and argue that we
just are not aware of the brain as the brain. We are aware
of it as something indeterminate. And, since the object of
introspective awareness is nothing definite, there is
nothing to prevent this object from being the brain.
Second, Armstrong could argue that since "is aware of"
generates opaque contexts, it does not follow from one's
direct awareness of M, and M's identity with B, that one
must be directly aware of B. Hence, if Armstrong wanted to
deny the implication that under his analysis we must be
directly aware of the brain, he could possibly do so by
relying on the notion of referential opacity.
Nevertheless, in light of criticisms (1) and (2) above,
and the fact that the defenses in (3) are not sufficiently
worked out, we fail to see how Armstrong has accomplished
either of his two tasks. He has neither shown Dualism to be
false ("unsatisfactory") nor has he shown his Central State
Theory to be the "theory on the right track." Again, he
might be right in both instances, though it is not clear
that he is from what he writes.
BI B LIO GR APH Y
i
159
BIBLIOGRAPHY
Anderson, Alan Ross, Ed. Minds and Machines. Englewood
Cliffs, N. J.s Prentice-Hall, Inc., I?64.
Armstrong, D. M. A Materialist Theory of the Mind. New
York: Humanities Press, 1968.
Beck, Lewis White, Ed. Eighteenth-Century Philosophy. New
York: The Free Press, 1966.
Campbell, Keith. Body and Mind. Garden City, New York:
Doubleday & Company, Inc., 1970.
Chappell, V. C., Ed. The Philosophy of Mind. Englewood
Cliffs, N. J.: Prentice-Hall, Inc., Hf62.
Edwards, Paul, and Pap, Arthur, Eds. A Modern Introduction
to Philosophy. New York: The Free Press, 1973.
Hook, Sidney, Ed. Dimensions of Mind. New York: Collier
Books, 1966.
Hospers, John. An Introduction to Philosophical Analysis;
2nd ed. Englewood Cliffs, New Jersey: Prentice-Hall,
Inc., 1960.
Hume, David. An Inquiry Concerning Human Understanding.
Indianapolis; New York: Bobbs-Merrill Company, Inc.,
1963.
Kaufman, Walter. Philosophic Classics, Bacon to Kant.
Englewood Cliffs, N. J. : Prentice-Hall, Inc., T968.
Loux, Michael J., Ed. Universals and Particulars: Readings
in Ontology. Garden City, New York: Doubleday & Com-
pany, Inc., 1970.
Morick, Harold, Ed. Introduction to the Philosophy of Mind.
Glenview, 111.: Scott, Foresman and Company, 1970.
O'Connor, John, Ed. Modern Materialism: Readings on Mind-
Body Identity. New York: Harcourt, Brace & World,
Inc., ld69.
160
161
O'Connor, Tillman Ber " ' " ‘ ’ ' Philosophy; 2nd
Penfield, Wilder, and Rasmussen, Theodore. The Cerebral
Cortex of Man. New Yorks The MacMillan Company, 1950.
Penfield, Wilder, and Roberts, Lemar. Speech and Brain
Mechanism. Princeton, N. J.s Princeton University
Press, IU58.
Popkin, Richard H., Ed. The Philosophy of the 16th and 17th
Centuries. New York's The Free Press, 1966.
Quine, W. V. Ontological Relativity and Other Essays. New
York and Londons Columbia University Press, 1969.
Quine, Willard V. From A Logical Point of View. New Yorks
Harper & Row Publishers, 1963.
Russell, Bertrand. The Problems of Philosophy. New Yorks
Oxford University Press, l9*?0.
Ryle, Gilbert. The Concept of Mind. New Yorks Barnes &
Noble, Inc., 1967.
Shaffer, Jerome A. Philosophy of Mind. Englewood Cliffs,
N. J.s Prentice-Hall, Inc., 1968.
Wiener, Philip P., Ed. Leibniz Selections. New Yorks
Charles Scribner's Sons, 1951.
Wolfson, Harry Austryn. The Philosophy of Spinoza. Vols.
I & II. New Yorks Schocken Books, 1969.
Ed. New York

Asset Metadata

Creator Walker Kendrick Wayne (author)

Core Title A Critical Examination Of D. M. Armstrong'S Materialistic Theory Of Mind

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Philosophy

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Philosophy

Language English

Advisor Willard, Dallas (committee chair), Dreher, John H. (committee member), Robb, J. Wesley (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c20-527987

Unique identifier UC11225851

Identifier 7506452.pdf (filename),usctheses-c20-527987 (legacy record id)

Legacy Identifier 7506452.pdf

Dmrecord 527987

Document Type Dissertation

Rights Walker Kendrick Wayne

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA

Linked assets

University of Southern California Dissertations and Theses