Citation: 18 Cardozo L. Rev. 153 1996 - 1997
Content downloaded/printed from
HeinOnline (http://heinonline.org)
Wed Mar 23 10:58:49 2016
-- Your use of this HeinOnline PDF indicates your acceptance
of HeinOnline's Terms and Conditions of the license
agreement available at http://heinonline.org/HOL/License
-- The search text of this PDF is generated from
uncorrected OCR text.
-- To obtain permission to use this article beyond the scope
of your HeinOnline license, please use:
https://www.copyright.com/ccc/basicSearch.do?
&operation=go&searchType=0
&lastSearch=simple&all=on&titleOrStdNo=0270-5192
PRAGMATISM AND REALISM
HilaryPutnam*
Let me begin by asking what will undoubtedly seem to many,
if not all of you, a most peculiar question: How did it happen that
the first philosopher to present a completely worked out version of
direct realism in the entire history of modem philosophy was none
other than the American pragmatist, William James?
The peculiarity of this question is expected because James's
pragmatism is often thought of (especially by those who have accepted
Richard Rorty as their guide to pragmatist ways of thinking)
as a species of antirealism. The very fact that James ardently
defended a direct realist account of perception will come as a shock
to you, assuming that is how you are accustomed to think of pragmatism.
And even those of you who are ultimately convinced by
my reading of William James, may still want to challenge my claim
that James was the first to successfully promulgate a direct realist
picture. Had Thomas Reid not already just done that? And wasn't
James's good friend and friendly (if occasionally scathing) critic,
the great founder of pragmatism itself, Charles Sanders Peircealso
a direct realist years before James was? What a bizarre view
of the history of modem philosophy the lecturer seems to have!
Additionally, some may find the question peculiar because it is
not evident why one should care. Presumably, the title Pragmatism
andRealism did not suggest that you were to listen to a dry discussion
of the history of philosophy.
I will address this last concern. Although the Putnams, Ruth
Anna and myself, are presently writing a book about William
James that will provide full textual support for the interpretation I
sketch in this lecture, my interests today are not primarily historical
ones. What I shall understand by the term "direct realism" today is
not a particular metaphysical theory; rather it is our implicit and
everyday conviction that in experience we are immediately aware
of such common objects as trees and buildings, not to mention
other people. I am interested in James's defense of direct realism-of
what he called "natural realism," because I see overcoming
the traditional picture of perception-a picture according to
which our sensations are as much an impassible barrier between
ourselves and the objects we perceive as a mode of access to
* Cogan University Professor, Harvard University
CARDOZO LAW REVIEW
them-as absolutely necessary if philosophy is to ever stop "spinning
its wheels" in a futile attempt to locate a resting place in the
dispute about metaphysical realism and antirealism.1 Although, in
the end, the version of direct realism that I would defend is not
James's, it was his defense of direct realism that led me to appreciate
the issue's fundamental importance.
Let me return to the initial concern: How can a pragmatist
also be a realist?
James:Pragmatistand Realist
Let me begin with the admission that there are antirealist elements
in James's philosophy (even if he would not have regarded
them as such). Although James's theory of truth is both subtle and
complex (so much so that neither his critics nor his extravagant
admirers have done it justice), I argue elsewhere2 that, at the end
of the day, it does commit James to a degree of antirealism about
the past (though not exclusively about the past) which is quite unacceptable.
As Bertrand Russell aptly recognized, however, it is
possible to admire James's theory of perception without admiring
his theory of truth. This observation, however, does not really
speak to the concern. Given that James had a realist side, how did
he reconcile it with a theory of truth that has startlingly antirealist
consequences? For two reasons, that too should not really be so
great a puzzle. First, no one seems to have a problem recognizing
that Peirce was both a realist and a pragmatist (though I would
argue that Peirce's definition of truth as the opinion to which inquiry
would converge if indefinitely pursued has the same antirealist
consequences about the past, etc., as that of James's theory).'
The second reason is that neither James nor Peirce admitted that
their theories of truth possessed these antirealist consequences.
They were wrong. If one rejects the facile view that Peirce was a
1 For a discussion of the importance of overcoming this picture, see Hilary Putnam,
Sense, Nonsense and the Senses:An Inquiryinto the Powersof the Human Mind, 41 J. PHIL.
445 (1994) [hereinafter Putnam, Sense, Nonsense and the Senses].
2 For a discussion of James's theory of truth, see A COMPANION TO WILLIAM JAMES,
(R.A. Putnam ed., forthcoming 1997).
3 For Russell's view of James, see HILARY PUTNAM, PRAGMATISM: AN OPEN QUESTION
(1995).
4 See Hilary Putnam, Pragmatism,in 95 THE ARISTOTELIAN Soc'y 291 (1995) [hereinafter
Putnam, Pragmatism].
5 In Peirce's case however, the story is more complex because Peirce was willing to
make empirical assumptions (ones which are, however, incompatible with today's physics)
to guarantee that his theory of truth would not be incompatible with his realism. Given his
fallibilism, Peirce might even be happy that his theory has been refuted by a scientific
[Vol. 18:153
PRAGMATISM AND REALISM
complete realist while James thought anything could be true provided
it was satisfying to believe, then pragmatism can be seen as a
rich source of insights that was not free from errors.
My concern today, however, is not with the antirealist side of
James's thought-manifested to speakmuch too quickly and inaccurately
in his identification of what- is true with what will ultimately
be "verified" in his own special sense of the word-but with
his realism.
James's efforts to work out a satisfactory form of realism began
with what can be seen as a turn from psychology to philosophy,
were it not misleading to speak of a "turn" in the case of a thinker
who had published essays on philosophical topics beginning in the
1870s. Nevertheless, James did not begin to devise his own systematic
metaphysical view until the 1890s. Indeed, in his monumental
work, The Principlesof Psychology (although he voiced philosophical
opinions on a wide range of subjects), his official stance was
that he was bound as a psychologist to assume the working philosophy
of the experimental psychologist, which he took at that time to
be a form of dualism coupled with a belief in psychophysical parallelism.
'The position of The PrinciplesofPsychology, as far as metaphysics
is concerned, is that while this working philosophy is
doubtless inadequate, its inadequacy need not trouble the psychologist
qua psychologist.6 James reversed himself, however, in the
Epilogue to Psychology:Briefer Courseand again in a lecture given
to the American Psychological Association, when he declared that
good psychology requires a more coherent metaphysical stance
than we presently possess, and in his 1895-96 seminar on "the Feelings,"
when he began to devise the views that he would later present
to the world as Essays in Radical Empiricism.7 (By the way,
the discovery that James had worked out the essence of his "radical
discovery. The discovery in question-by Stephen Hawking-is the discovery that there is
such a thing as the irretrievable destruction of information. This refutes Peirce's claim that
we are entitled to believe that scientific investigation could discover the answer to any
factual question if sufficiently continued, and that claim is necessary to Peirce's defense of
the realistic character of his notion of truth.
6 James writes, in the preface to Principlesof Psychology, that:
This book, assuming that thoughts and feelings exist and are vehicles of knowledge,
thereupon contends that psychology when she has ascertained the empirical
correlation of the various sorts of thought or feeling with definite conditions
of the brain, can go no farther-can go no farther, that is, as a natural science.
If she goes father she becomes metaphysical.
WILLIAMS JAMES, Preface to Volume One of THE PRINCIPLES OF PSYCHOLOGY 6 (1981)
(emphasis added).
7 This address, The Knowing of Things Together, was published in The Psychological
Review in March 1895, reprintedin WILLIAM JAMES, ESSAYS IN PHILOSOPHY 71-89 (1962).
1996]
CARDOZO LAW REVIEW
empiricism" as early as 1895-96 is due to a young scholar at
Harvard named David Lamberth, on whose research I am relying
in this paragraph.8)
The fundamental idea of James's new metaphysics is the radical
rejection of the subject/object split. This rejection seems to
have been provoked first of all by James's attention to the phenomenology
of perception. "Phenomenology" has been used by different
philosophers to stand for very different projects. What unifies
those projects is that they reject the picture of experience in classical
associationist psychology as a false description dictated by a
received philosophical, view (originally, the view of British empiricism),
that was imposed willy-nilly on a very complex set of phenomena,
and that they all aim to replace this imposed picture with
an accurate and philosophically useful description of the actual
character of experience. In this sense, one may say that in The
Principlesof Psychology, James was already doing "phenomenology"-indeed
we know that Husserl was inspired by that book.9
Now, the received empiricist psychology, coupled with mind/matter
dualism (as in the working philosophy of The PrinciplesofPsychology),
leads to the following picture of perception: (1)in perception
we receive "impressions" which are immaterial, totally differentand
separated by a metaphysical gulf in fact-from all the material
objects we normally claim to perceive; and (2) from the character
of our internal mental impressions we infer how things are in the
external physical world.
Nothing, however, could be farther from the way perception
seems to be, or farther from the "phenomenology" of perception.
Phenomenologically, we seem ourselves to be perceiving tables and
chairs (or cabbages and kings), not immaterial intermediaries.
What we see in James's first attempts to work out his new metaphysics10
isthe conviction that the phenomenology of perception is
the best guide to a correct ontology. In effect, James has entertained
the heretical thought: What if all the philosophers are wrong
and the way it seems to be is the way it is?
8 David D. Lamberth, James's VarietiesReconsidered:Radical Empiricism,the ExtraMarginaland
Conversion, 15 AM. J. THEO. & PHIL. 257 (1994); DAVID D. LAMBERTH,
METAPHYSICS, EXPERIENCE AND RELIGION IN WILLIAM JAMES'S THOUGHT (forthcoming
1997) (provides a detailed periodization of the development of James's metaphysics).
9 Cf. BRUCE WILSHIRE, WILLIAM JAMES AND PHENOMENOLOGY: A STUDY OF "THE
PRINCIPLES OF PSYCHOLOGY" (1968).
10 I am referring again to the Epilogue in WILLIAM JAMES, PSYCHOLOGY: BRIEFER
COURSE 395 (1984), and to William James, Notesfor Philosophy20b: PsychologicalSeminary-The
Feelings,in WILLIAM JAMES, MANUSCRIPT LECTURES 212-29 (1988) [hereinafter
MANUSCRIPT LECTURES].
[Vol. 18:153
PRAGMATISM AND REALISM
In addition to denying that the immediate objects of our perceptions
are immaterial "sensations" or "impressions," James further
denies that we are aware of a substantial self (or as he puts it,
a substantial "consciousness")." In James's own terminology,
Notes for Philosophy 20b: PsychologicalSeminary-The Feelings,
what is given is a "field" of objects which only upon reflection appear
either as perceivable "external" objects, or (when we engage
in an alternate mode of reflection) as "sensations" or affectations
of our own subjectivity. The datum, the phenomenon, the pure experience,12
in itself "has no such inner duplicity," James writes. 3
Here too, James regards the phenomenology as the best guide to
an ontology. Reality in itself does not consist of two radically different
sorts of things-subjects and objects-with a problematic relation.
Rather, it consists of the data-the phenomena.-and it is
just that these can be thought about in different ways.
As mentioned previously, James is describing the phenomenology
of perception; but he is also doing more than that. By suggesting
that we can take that phenomenology seriously in a way
that philosophers have long thought we could not, he is proposing
that we return to a standpoint close to what he calls the "natural
realism" of the common man.
11 James's denial of a substantial "consciousness" is, of course, not a denial that we are
conscious! Although this charge is occasionally brought against James, it seems to depend
on the idea that one cannot believe that we are conscious unless there is a thing which is
our "consciousness," or at least a substantial self (or at least a transcendental ego, such as
Kant's famous "I think"). What James was denying was the need for any of these metaphysical
items, not the fact that Jones is conscious of the purring of the cat, or whatever.
James's denial that we are aware of any such item is, of course, a denial that Kant-who
did not think we experience the transcendental self-would have agreed with, as, for quite
different reasons, would Hume.
12 A term James took from Avenarius, by the way, although he later seems to have
quite forgotten this debt, since he spoke in quite scornful tones of that thinker's unreadability!
This was pointed out to me by David Lamberth. To my knowledge, Ignas K.
Skrupskelis was the first person to suggest this in his introduction to MANUSCRIPT LECTURES,
supra note 10.
. 13 Although the sentence: "Experience... has nosuch innerduplicity .... comes from
Does 'Consciousness'Exist?, reprintedin WILLIAM JAMES, ESSAYS IN RADICAL EMPIRICISM
3, 6 (1976) [hereinafter JAMES, ESSAYS IN RADICAL EMPIRICISM], the thought is already
present in James's Notes for Philosophy20b: PsychologicalSeminary-The Feelings,
supra note 10. See id. at 228 ("But nothing postulated whose whatness is not of some
naturegiven in fields, that is not of field stuff, datum-stuff, experience stuff, 'content.' No
pure ego, for example, and no material substance. This is the hypothesis that we are trying
to work out.") (footnote omitted).
1996]
CARDOZO LAW REVIEW
A HistoricalDigression
In trying to defend "natural realism" (even if with some philosophical
additions and revisions), James, though probably unbeknownst
to him, possessed an Aristotelian way of thinking about
perception-a way of thinking that was dominant in the later middle
ages. Aristotle held that in perception, the very form of the
object perceived is in the mind. To be sure, there are difficulties
with this view.14 Some of those difficulties are with the essentialism
involved in the Aristotelian notion of "form," and those difficulties
become serious when we hold that the form of the object is
also in the mind when we merely think about the object, as well as
when we perceive it. It does seem that in conception, the form
Aristotle speaks of is what he regards as the "essence," and I find
serious difficulties with the idea that we can only think about things
whose "essence" is known. It has been suggested to me by Gisela
Striker, however, that in the case of perception, it is unlikely that
Aristotle has so demanding a notion of form in mind. It is plausible
that the form that we receive in perception is simply the sensible
form-the color, shape, texture, sound or whatever.15 Reading
Aristotle in this way, it would seem that what he is saying is simply
that in perception we are aware of the sensible properties of external
things-their shape and/or color, etc.
It may be, of course, that Aristotle also thought that his talk of
the same thing (the form) being in two places at once (in the object
perceived and in the mind) explained how such direct awareness of
sensible properties of external things is possible; if so, he was mistaken
as the explanation is empty. (It is also possible, as indeed the
rather strange prose of De Anima at this point suggests might be
the case, that he is simply using a figure of speech to say that we
are directly aware of properties of external things, as today the figure
of speech that something is "in my mind" is used.) Whatever
Aristotle may have intended, and whether part of what he intended
must be rejected as unhelpful, he at least believed that we do have
an awareness of the sensible properties of "external" things, and
that this is not to be cashed out as meaning that we merely have
"images" or "representations" of those things before our minds (as
holds the view that has been dominant ever since Descartes). Furthermore,
Aristotle's ideas were faithfully supported by Aquinas.
14 Cf. HILARY PUTNAM, Aristotle after Wittgenstein, reprintedin WORDS AND LIFE 62
(James Conant ed., 1994) [hereinafter PUTNAM, WORDS AND LIFE].
15 I did not consider the possibility of this reading in Aristotle after Wittgenstein, and if I
had I would have been more charitable to the Aristotelian view. See id.
158 [Vol. 18:153
PRAGMATISM AND REALISM
Aristotle was aware that perception requires many things-in
the case of sight, the form must somehow be transmitted through
the air to the eyes, and there must be physical modifications in the
eyes16 and (perhaps) in the cognitive organ (which Aristotle understood
to be the heart and blood stream) before the form can be in
our psyche. Perception is supervenient on physical processes. That
is not, however, an obstacle to thinking that what results from all
these transactions between the organism and the environment is
the perception of the way something is, and not only in the Pickwickian
sense of the presence of a "representation" which in some
mysterious way refers to the way something is.
By Descartes's time, however, Scholasticism and Aristotelianism
were in ill repute, and the old notion of a "substantial form"
smacked of a metaphysics that was no longer acceptable. As is
usually the case in philosophy, the baby was tossed out with the
bathwater, and the representational theory of perception that continues
to dominate philosophy and psychology (and, today one can
add "cognitive science" as well) came to be the only possible view.
To be a direct realist was to be a "naive realist," and being naive is,
of course, bad. It is against the background of the consensus that
has endured for centuries, that "naive realism" had been refuted
once and for all, that James's amazing philosophical radicalism has
to be appreciated.
The Importance of the Issue
Before I describe James's version of direct realism, I wish to
comment further about the importance of the issue. One reason I
feel I must do this, is that after being at the center of philosophical
attention in the first quarter of the twentieth century, the philosophy
of perception receded from view (and almost disappeared
from serious attention after the death of J.L. Austin in 1960). Indeed,
of the great figures in Anglo-American philosophy, only Peter
Strawson maintained a steady, if intermittent, interest in the
16 This has been challenged by Myles Burnyeat. For Burnyeat's paper and a reply defending
the interpretation given here, see ESSAYS ON ARISTOTLE'S DE ANIMA (MARTHA C.
NUSSBAUM & AMP-LIE 0. RORTY EDS., 1992). VICTOR CASTON HAS POINTED TO PASSAGES
IN ARISTOTLE'S BIOLOGICAL WRITINGS IN WHICH ARISTOTLE SPECULATES THAT COGNITION
MAY INVOLVE BUILDING OF A KIND OF REPRESENTATION IN THE BLOOD STREAM!
SUCH REPRESENTATIONS ARE NOT, OF COURSE, IN OUR PSYCHE, AND ARE NOT IDENTICAL
WITH THE PERCEPTUAL EXPERIENCE; STILL THE FACT THAT ARISTOTLE HIMSELF WAS WILLING
TO SPECULATE IN THIS WAY SHOWS THAT AN "ARISTOTELIAN" VIEW NEED NOT BE
HOSTILE TO THE IDEA THAT SUBPERSONAL PROCESSES OF A MODEL BUILDING KIND HAVE A
ROLE TO PLAY IN THE ETIOLOGY OF PERCEPTION. (ARISTOTLE AS THE FATHER OF COGNITIVE
SCIENCE?)
1996]
CARDOZO LAW REVIEW
topic. A second and less sociological reason is that one aspect of
the traditional picture of perception has now been given up, and
this may give the illusion that the picture is no longer as problematic
as it once was.
When I described the traditional picture, I mentioned two elements:
in perception we receive "impressions" that are immaterial,
totally different-separated by a metaphysical gulf-in fact from
all the material objects we normally claim to perceive; and from
the character of our internal mental impressions we infer how
things are in the external physical world. Today, however, a majority
of philosophers appear to be hard-core materialists (even if they
prefer the more innocuous label of a "naturalist"), and these philosophers
would say that our "impressions" are little more than
brain events or processes. These philosophers would say that the
representational theory of perception no longer requires us to posit
a "metaphysical gulf" between impressions and external objects.
In fact however, the new view is at least as problematical as
the old one. Philosophers who advocate the new view concede that
we are not aware of our "sensations" as brain processes. We are
aware of the blue of the sky as a color extended over an area, not
as a cortical process. The claim advanced by Donald Davidson and
others that the experience is "identical" with a cortical event trades
on a notion of "identity" that seems to me entirely meaningless.
(Davidson has admitted, under the pressure of criticism from W.V.
Quine, that the criterion he originally offered for what he calls "token
identity of events of different types" was fatally flawed. 17)
Moreover, even if we were to concede that our sense impressions
are brain processes-that is, even if we suppose that the notion of
"identity" can be made sense of in this context (though we know
not how!)-the picture remains one in which our cognitive
processes extend no further than an "interface" between us and the
external objects (an interface consisting now of cortical processes
rather than processes in a mental substance). The objects of perception,
those cabbages and cabinet ministers, impinge on our
mind/brains only causally, not cognitively.
This feature of the current materialist version of the representational
theory of perception, the feature that the picture of cognition
provided by that picture is in the end a purely causal and not a
normative one, has been used by Richard Rorty to argue against
the idea that we can think of reference as a relation to objects in
17 For a fuller discussion of this issue, see Putnam, Sense, Nonsense and the Senses,
supra note 1, at 477-83.
[Vol. 18:153
PRAGMATISM AND REALISM
the world. "We are connected to the world causally, not semantically,"
I have heard Rorty say more than once.'8
With these remarks, Rorty builds on an argument against the
"given" used by Wilfrid Sellars. Sellars assumed what I have called
the picture of our impressions (our "raw feels," in his terminology)
as an "interface" between us and the world (although he does not
seem to have been willing to identify them with brain events). Sellars
further observed that any given impression-report, (e.g., "I am
experiencing E now"), must have a multitude of causes, and that
there is nothing in any one of those causes, qua causes, to single it
out as "the appropriate" cause of my verbal response or my verbalized
thought. Even if my verbal report (or my thought) is in part
caused by the very quality of the raw feel I am trying to describe, it
may still be a misreport-my report cannot causally depend only
on the quality of my "raw feel." If it is to be a report in a language
with stable meanings, it must also depend on my prior linguistic
conditioning, on attention, and on "set," etc. Like Rorty, Sellars
thought that to postulate a semantic relation between linguistic
items and nonlinguistic objects which determines when I am successfully
"referring," is to postulate a mystery relation.
Rorty generalized Sellars's argument (once accepted it invites
immediate generalization), and concluded that if we are connected
to the world "causally but not semantically" (a fair summary of
Sellars's view), then our words have no determinate real counterparts.
Sellars attempted to avoid going as far as Rorty did when he
postulated a holistic relation between our conceptual schemes and
reality that he called "picturing"; postulating further in Peircean
fashion, that with the progress of science our schemes come to
"picture" the world more and more accurately. Rorty however, rejects
Sellarsian "picturing" on the grounds that it is as occult a relation
as reference allegedly is.
Rorty and Sellars are right in saying that any given event can
be traced to a multitude of different causes. As William James
wrote in a different context:
Not a sparrow falls to the ground but some of the remote conditions
of his fall are to be found in the milky way, or in our federal
constitution, or in the early history of Europe. That is to
say, alter the milky way, alter the [facts of our] federal constitution,
alter the facts of our barbarian ancestry, and the universe
would so far be a different universe from what it now is. One
18 This paragraph and the four that follow are adapted from HILARY PU' NAM, Realism
without Absolutes, reprintedin PUTNAM, WORDS AND LIFE, supra note 14, at 279, 285.
1996]
CARDOZO LAW REVIEW
fact involved in the difference might be that the particular little
street-boy who threw the stone which brought down the sparrow
might not find himself opposite the sparrow at that particular
moment; or, finding himself there, he might not be in that particular
serene and disengaged mood of mind which expressed
itself in throwing the stone. But, true as all this is, it would be
very foolish for anyone who was inquiring the cause of the sparrow's
fall to overlook the boy as too personal, proximate, and so
to speak anthropomorphic an agent, and to say that the true
cause [of the sparrow's fall] is the federal constitution, the westward
migration of the Celtic race, or the structure of the milky
way. If we proceeded on that method, we might say with perfect
legitimacy that a friend of ours, who slipped on the ice upon his
door-step and cracked his skull, some months after dining with
thirteen at the table, died because of that ominous feast.19
Yet to conclude (from the fact that there is a great difference
between asking for the referent of a term and asking for the causes
of a particular "tokening" of that term) that we must abandon the
idea that there is a relation of reference that holds between some
of our terms and objects in the world, is to engage in a gesture of
repudiation with respect to our conception of ourselves as thinkers
in a world that is so sweeping as, in the end, to invite the suspicion
that it is simply an empty pose. In this suspicion, I am naturally
joined by metaphysical realists. Metaphysical realists, of course, deplore
the rejection by both Rorty and Sellars of the very idea of
semantical words-world relations. If, however, (as most of the contemporary
ones do) they also wish to endorse a "bald" version of
naturalism,20
they cannot simply posit a semantical words-world relation;
they must also show that it can be reduced to non-semantical
relations and facts-to posit irreducible semantic relations is no
better from their point of view than to posit immaterial sense data.
Attempts to reduce semantic relations to non-semantic ones have
been utter failures-to the point that we presently have no idea
what such a reduction could conceivably look like.21
It was in this philosophical climate that I advanced my own
attempt at a "middle way" between antirealism and metaphysical
19 WILLIAM JAMES, GreatMen and Their Environment, in THE WILL TO BELIEVE AND
OTHER ESSAYS IN POPULAR PHILOSOPHY AND HUMAN IMMORTALITY 216, 216-17 (photo.
reprint 1960) (1898).
20 1 owe the term "bald naturalism" to JOHN McDOWELL, MIND AND WORLD (1994).
21 For a critical examination of such attempts, see HILARY PUTNAM, RENEWING PHILOSOPHY
(1992) [hereinafter PUTNAM, RENEWING PHILOSOPHY].
[Vol. 18:153
PRAGMATISM AND REALISM
realism (my so-called "internal realism ' 22) in the 1970s and 1980s.
While I still defend some of the ideas that were involved in those
attempts (in particular the denial that reality dictates one unique
description is as central to my thinking as it ever was), it is now
clear that that attempt too was fatally flawed by its allegiance to
the traditional conception of our sensations as an "interface" between
us and the world. Thus, I can understand from my own experience
how, even if we did not very much discuss the philosophy
of perception in the 1970s and 1980s, we were-and too many of us
remain-the inheritors of a Cartesian (or Cartesian cum materialist)
picture of perception, a picture that had become all the more
coercive for not being critically discussed in the way it was at the
time Russell wrote The Analysis of Mind-a book that, for all its
retention of parts of the traditional picture, is extremely aware of
the difficulties to which that picture leads, as most of today's philosophers
are not.
Another (shorter)HistoricalDigression
Even if you are now convinced (I trust) that it is important to
overcome the representational theory of perception, the picture of
our sensations as "between" us and those "external objects," there
is still the question as to our historical accuracy in giving James and
not Thomas Reid the credit for being the first modern philosopher
to revive direct realism. It is true that Thomas Reid thoroughly
understood the disastrous consequences of the representational
theory, and that (both in the Inquiry and in the Intellectual Powers)
he called-indeed, he vigorously polemicized-for a return to
a direct realism.23 My reason for not ultimately counting Reid as a
successful advocate of the direct realist cause, is that he retains the
idea that sensations are nonconceptual and internal "signs" (as opposed
to sensingsof what is there) as an essential part of his epistemology
and ontology. For example, Reid writes:
[In perception] there is something which may be called the sign,
and something which is signified to us ...by that sign ....Thus
when I grasp an ivory ball in my hand, I have a certain sensation
of touch. Although this has no similitude to anything material,
yet, by the laws of my constitution, it is immediately followed by
22 A name which I now find to have been an unhappy one, for the reasons given in
Putnam, Sense, Nonsense and the Senses, supra note 1, at 461-63.
23 See John J. Haldane, Reid, Scholasticism and CurrentPhilosophyof Mind, in THE
PHILOSOPHY OF THOMAS REID 285 (Melvin Delgarno & Eric Matthews eds., 1989).
19961
CARDOZO LAW REVIEW
a conception and belief that there is in my hand a hard smooth
body of a spherical character.24
This differs from the standard Cartesian account only in adding
that, owing to "the laws of my constitution," I am able to form
a "conception" of the object itself from this sensation even though
it "has no similitude to anything material." This power, however,
necessarily appears to be something mysterious in Reid's account,
since the mind has as "input" something different in kind from
what it forms conceptions and beliefs about. Similar remarks apply
to Peirce's defense of direct realism-indeed, Peirce cites Reid as
the one who got it right.25
James's Radical Empiricism
James's name for this area of his philosophy is "radical empiricism";
James does not, however, use the term "direct realism," but
the much better'term "natural realism" appears a number of times
in his Essays in RadicalEmpiricism,26 and he repeatedly insists that
radical empiricism is close to natural realism, or revives it, or shows
that it can be maintained. We have already provided an answer to
the puzzle as to how a pragmatist came to revive natural realism; it
was James's typically pragmatist insistence that,we take seriously
the way in which we think about-and have to think aboutperceptual
experience in the course of living our lives that virtually
forced natural realism upon him. And James is certainly correct
that in the course of living our lives, we have to think of ourselves
as living in what he calls "a common world," that each of us must
think that she is aware of the other's body and not simply a representation
of it, and we all have to think that we are aware of many
of the same objects. It is true that in elaborating a philosophy
which took this idea as correct, James was led to some elaborate
metaphysical construction which I do not presently have time to
discuss. For that reason, towards the end of the this lecture I shall
speak to the further question: To what extent can we preserve natural
realism without accepting all of James's metaphysics of radical
24 THOMAS REID, ESSAYS ON THE INTELLECTUAL POWERS OF MAN 332 (James Walker
ed., 1850).
25 Cf. 5 COLLECTED PAPERS OF CHARLES SANDERS PIERCE 5.56, 5.444, 5.539, 6.95
(Charles Hartshorne & Paul Weiss eds., 1934).
26 "Natural realism" is a much better term because, as pointed out in J.L. AUSTIN,
SENSE AND SENSIBILIA (1963), there is a great deal that isproblematic with the traditional
epistemologist's use of "direct" and "indirect."
27 It is for the same reason that Dewey followed James in this, and that Peirce earlier
sought-even if, in my view, unsuccessfully-a return to direct realism.
164 [Vol. 18:153
PRAGMATISM AND REALISM
empiricism? I shall further suggest that we can find resources for
doing this in the writings of J.L. Austin.28 Even if, however,
James's way was not the only way, or even the best way to defend
natural realism, the fact remains that it was the first way to be proposed
after the fateful Cartesian turn in modern philosophy. 9
James's way involved what Russell (not James himself, for
whom the term "monist" was anathema) described as a "neutral
monist" ontology. In such an ontology, the properties and relations
we experience are the stuff of the universe; there is no nonexperiential
"substratum" (this is an idea of James's with which
Russell was sympathetic), and these experienced or experienceable
properties and relations (James is unfortunately a little vague at
this crucial point) make up both minds and material objects.
Moreover, minds and material objects, in a sense "overlap"; the
very thing I experience as a sensation of red is, in another context,
also what I refer to as "a patch of color on the wall." Illusions do
not prove that we never "directly experience" external objects; in
James's radical empiricist metaphysics, what the phenomena of illusions
go to show is that not every bit of pure experience has the
status of being a part of a "real" object, not that none do.
Of course, there is the obvious objection that the skeptical
epistemological problem has not been "solved." In James's account,
there is no absolutely certain way to know when we are subject
to an illusion and when we are not. But James would reply
that the problem is even worse for the traditional theory. As James
puts it for the Berkleyan school:
Our lives are a congeries of solipsisms, out of which in strict
logic only a God could compose a universe even of discourse.
... If the body you actuate be not the very body that I see
there, but some duplicate body of your own with which that has
nothing to do, we belong to different universes, you and I, and
for me to speak of you is folly.30
In short, James argues that several minds, each acquainted only
with its own private objects, could not arrive by any process of inference
at knowledge or even thought of one another.
28 For a fuller account, compare Putnam, Sense, Nonsense and the Senses, supra note 1.
29 Charles Taylor has objected that I am leaving Hegel out of account here. Even if
Hegel can be seen as a kind of direct realist (which is problematic), he seems to have had
no influence on James in this respect.
30 JAMES, ESSAYS IN RADICAL EMPIRICISM, supra note 13, at 37-38.
19961
CARDOZO LAW REVIEW
The advantage of pragmatism over traditional "foundationalist"
epistemology, in James's view, is that the way in which pragmatist
philosophers answer skeptical doubts is the way in which
doubts are answered in practice, by appealing to tests that in fact
work in our lives. If I think that what I see may be an illusion, I
can try to touch it, or look at it from a different position, or ask
other people to take a look. There are not, in James's view, two
sets of criteria for being "real"-commonsense criteria and philosophical
criteria.
With so much, Russell was able to agree in The Analysis of
Mind. In a chapter of that work dealing with sensations and
images (chapter eight), Russell heartily endorses James's view that
"the dualism of mind and matter cannot be allowed as metaphysically
valid," writing, "[o]n this subject we may again quote William
James. He points out that when we say we merely 'imagine' things,
there are no such effects as would ensue if the things were what we
call 'real."' ' 31 It is at this point Russell quotes James at length and
then restates the point in his own terminology: the difference between
so-called 'mental' phenomena and 'physical' phenomena is a
difference in the causal laws obeyed, not a fundamental dualism.
But there is an important aspect of what James calls "natural
realism" with which Russell was unable to agree (although it is not
clear that he realized that this was a point of disagreement). For
immediately after this, Russell proceeds to give precisely the description
of experience that James had previously rebelled against
in The Principles of Psychology: experience consists of color
patches, etc., and we think we see tables and chairs because we
make "inferences" of various kinds!32 Evidently Russell is willing
to follow James about color patches ("the sensation that we have
when we see a patch of colour simply is that patch of colour, an
actual constituent of the physical world, ' 33) but not about tables
and chairs.
Here is an example (my own rather than his), of what bothers
Russell: while it is true that we normally see tables and chairs as
just that, very often we do not see the side away from us; yet seeing
something as a chair is seeing it as something which has an unseen
side, and one of a certain kind. Because that knowledge, the
knowledge of the unseen side, cannot be perceptual (this seems
31 BERTRAND RUSSELL, THE ANALYSIS OF MIND 137 (1921).
32 Id. at 140.
33 Id. at 142.
166 [Vol. 18:153
PRAGMATISM AND REALISM
evident to Russell) it must be inferred. 4 We can only find what is
really given in sensation by stripping away all "mnemic" (i.e., conceptual)
contributions. But James's view is that if there is an original
nonconceptual element in perception, we are unable to get
back to it (or can get to it only in free reverie-precisely when we
are not cognizing!). The question, how much of what we perceive is
"given" and how much is "added," is, he says in one place, like the
question: Does a man walk more essentially on his left leg or his
right? 35 Although James was not willing to go as far as Kant and
treat perception as a passive exercise of the same conceptual powers
that are exercised in judgment,36 the practical effect of this part
of his doctrine is the same. What we perceive, insofar as the perception
is available to us as a source of knowledge, is a sort of
fusion of sensation and conception. Given his constant emphasis
on the richness and variety of what is given in experience, he could
never accept the view that all we really see are color patches! The
Russell of The Analysis of Mind was strongly influenced by James,
but-only so far: his direct realism stopped with the color patches.
James's natural realism is full bodied.
James's Excesses
While I admire this full bodied natural realism, I indicated earlier
that James's "radical empiricism" contains elements that I find
excessively metaphysical. One such element is James's restriction
of what there is to "pure experience"; however we understand that
puzzling notion, this appears to be too restrictive. Although James
does at times try to allow not only what is perceptually experienced,
but even what is conceived as "pure experience," this strain
in his thought is never worked out, and is, indeed, in conflict with
his own account of conceptual thinking.37 For James, the world is
the experienceable world; since James has an admirable reluctance
to rule out any kind of talk that does real work in our lives, he is
forced to reinterpret talk of unobservables in physics, of
counterfactual connections, and of mathematical talk, etc., in ways
that are unconvincing and ultimately unsuccessful.
34 In the Notes for the seminar on "the Feelings," James insists that this "pointing" to
something more is part of what he calls the "content" of the "datum"; this is the exact
opposite of Russell's view.
35 WILLIAM JAMES, THE MEANING OF TRUTH 120 (1975).
36 For a recent discussion of the importance of this insight of Kant's, see McDOWELL,
supra note 20.
37 For a criticism of that account, see Putnam, Pragmatism,supra note 4.
19961
CARDOZO LAW REVIEW
In another respect, however, James's ontology is not restrictive
enough. To explain what ismeant by this, I shall close by bringing
in the great British philosopher J.L. Austin, a second figure in the
history of natural realism's revival. I have ventured a hypothesis as
to the origin of James's defense of natural realism. I do not know,
however, the aetiology of Austin's rejection of the whole idea of
sense data as private representations of an external reality, except
that that rejection may have come very early in Austin's life.38 In
any case, he is very likely to have been acquainted at least with
Russell's account of James's views in The Analysis of Mind.
It was, by the way, through reflecting on Austin's Sense and
Sensibilia, as I did repeatedly starting in the 1970s, that I first began
to take direct realism seriously. Although my views are now
closer to Austin's than to James's, I must admit that I was not at
first convinced by Austin.39 Only after I began to teach courses on
the philosophy of William James, and to focus on his RadicalEmpiricism,
did I begin to see that the endless pattern of recoil in
modern philosophy (from extravagant versions of realism to
equally extravagant versions of antirealism and back again) can
never be brought to rest unless we challenge the picture of the
mind, and particularly the picture of perception, that makes it seem
impossible to take our ordinary talk of perceiving and thinking
about objects seriously unless one reinterprets it in terms of a representational
theory of the mind. Understanding how that theory
fails to provide the desired "foundation" for our ordinary talk
(since it is just as much a mystery, in the end, how the supposed
"mental representations" can refer to objects as it is how,our ordinary
talk can do so!) makes it seem Rortian nihilism must be the
only option that remains (although that too, I would argue, is only
38 In 1936 (Austin was born in 1911), Austin and Isaiah Berlin held a class on Lewis's
book, CLARENCE I. LEWIS, MIND AND THE WORLD-ORDER: OUTLINE OF A THEORY OF
KNOWLEDGE (1929), in which Austin characterized Lewis's doctrine of qualie-specific,
sensible characteristics-as "complete nonsense." Cf. ISAIAH BERLIN, PERSONAL IMPRESSIONS
107 (Henry Hardy ed., 1981). Unfortunately, Berlin does not give Austin's grounds,
and so one cannot tell whether this means that Austin already held the views that he was to
defend in Sense and Sensibilia,supra note 26.
39 In HILARY PUTNAM, Models andReality, reprintedin 3 REALISM AND REASON: PHILOSOPHICAL
PAPERS 1 (1983), I explicitly rejected Austin's views. It now seems to me that
this rejection led me directly into a cul-de-sac with respect to the realism/antirealism issue.
For an explanation of this remark, see Putnam, Sense, Nonsense, and the Senses, supranote
[Vol. 18:153
PRAGMATISM AND REALISM
the illusion of an option, a fata morgana which disappears the moment
one tries to embrace it).4"
The most striking feature of the view put forward in Sense and
Sensibilia,at least on a first reading, is the rejection of something
that the tradition takes as self-evident: namely, that even non-veridical
experience must be analyzed on a perceiver-percipient
model. If I dream, or am subject to the illusion, or even hallucinate
that I see a building, I really do perceive something on the traditional
view-it is just that what I perceive is not a physical something
but a "mental" something. James (and Russell when he was
following James), retained just this feature of the traditional view.
To be sure, they denied that such mental somethings are made of a
different "stuff" from the physical things (this denial was a consequence
of their different versions of "neutral monist" metaphysics);
but they accepted the perceiver-percipient model. But on Austin's
still more radical approach, when, for example, I dream that I see a
building, I do not perceive anything-I only seem to perceive
something. With one stroke, Austin banishes the last vestige of the
tradition's "sense data." (Well, not entirely, you may object-we
still feel pains, for example, and a pain is not a physical object.
That is true, but Austin's point is that the tradition regards feeling a
pain and seeing a table as essentially similar-in both cases I have
"sensations"; and it is the use of the notion of a "sensation" in connection
with perceptual experiences, be they "veridical" or "nonveridical,"
that Austin regards as complete nonsense.)
I am almost at the end of my time, and to even sketch Austin's
view would require a lecture as long as this one has been. I will
only mention one obvious objection to Austin's account (the one
that bothered me for a long time): if we give up the idea that there
is a mental object in the case of non-veridical experience, how are
we to explain the similarity between the non-veridical experience
and a corresponding perception? To be sure, Austin's account does
not prevent us from saying that how an experience seems, depends
on our neural state in such-and-such ways; these causal dependencies
are matters of scientific fact. But do we not want to say more
than that? Do we not want to say that there is something identical,
a "common factor," present in the two experiences? Austin's view,
which I have come to share, is that one has here only the illusion of
40 For a discussion of Rorty's position, see PUTNAM, RENEWING PHILOSOPHY, supra
note 21; see also HILARY PUTNAM, Realism Without Absolutes and the Questionof Realism,
in PUTNAM, WORDS AND LI'E, supra note 14, at 279-94, 295-314.
1996]
CARDOZO LAW REVIEW
an explanation. Further inquiry into that question will have to be
in the discussion period.41
I will say, however, that Austin's strategy was not available to
James; James's whole metaphysical outlook is that everything that
seems to be present in experience is, in some sense, there-is, as
James sometimes puts it, a "bit" of pure experience. To go from
James's bold presentation of his metaphysics of radical empiricism
to Austin's Sense and Sensibilia, is to go from a bold metaphysical
construction with deep roots in traditional empiricist metaphysics
(roots which James repeatedly acknowledged, even as he tried to
correct what he saw as the errors of that empiricism), to a bitingly
cold attempt to achieve a certain kind of clarity. To some, this will
seem a loss. But at the end of the day, it is, I believe, a gain; and it
even preserves, I believe, the genuine moment of insight in James's
ambitious metaphysical project, which was James's realization that
our ordinary ways of talking and thinking about our perceptual experiences
should be taken seriously in philosophy. In taking so seriously
our common sense picture of ourselves as having access to a
common world-taking that picture so seriously because it, and the
actions that are interwoven with it and give it content, are essential
to our lives together, not just as knowers but as moral agentsWilliam
James was, in the best sense, both a "pragmatist" and a
"realist."
41 This question is addressed in Putnam, Sense, Nonsense and the Senses, supranote 1,
at 473-83.
[Vol. 18:153