Sie sind auf Seite 1von 16

The Arts: Being through Meaning

Author(s): George P. Stein

Source: The Journal of Aesthetic Education, Vol. 5, No. 4 (Oct., 1971), pp. 99-113
Published by: University of Illinois Press
Stable URL:
Accessed: 28-11-2015 06:50 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact

University of Illinois Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Aesthetic

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions

The Arts:Being through Meaning


A pianist was once asked about the meaning of a composition he had

just played. He replied by playing it. His assumptionwas that the meaning of a work of art can be rendered only through itself and not by a
verbal translation.
This assumption is of course implicitly questioned by anyone who
says anything about what a work of art means. And the fact is that
philosophers, critics, educators and those being educated do often ask
and attempt to answer the question: "What does this poem (or painting,
or symphony) mean?" If these questions and the answers are themselves
to be considered meaningful, some theory of meaning in the arts would
have to be implicit in the questions and answers.Such a theory will have
of course some effect upon the sorts of questions asked and upon the
nature of the answers given. The problem of a theory of meaning in
the arts is simply an interestingone to some. But to those involved in the
design of educational curriculums from kindergarten through college,
the problem should be, interesting or not, a major one. The purpose of
this paper is to reduce the ambiguity of the question, "What does this
work of art mean?" to the point where it may assist curriculumdecisions
in the arts.
To begin with, there are a couple of "nonassumptions"to be made:
1. There is no assumption as to the essential nature of "art" or "the
arts." Whatever is claimed with respect to the nature of meaning in the
arts requires no adherence to any of the classical or contemporarydefinitions of art.
GEORGEP. STEIN is professor of philosophy and chairman of the department of
philosophy at Bloomfield College, New Jersey. His most recent publication is

The Ways of Meaning in the Arts (1970).

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



2. There is no assumptionas to whether or not art is definable at all.

It is assumed that people recognize or learn to recognize that certain
things or events are works of art, in some way similar to the way we
recognize or learn to recognize that certain things are bridges, games,
horses, disasters, etc. It had been, before Wittgenstein, traditional to
assume that one could not reliably make this sort of recognition unless
one was equipped with a "definition"of art. But it is now common for
theorists to believe that works of art are recognized as such by virtue of
the strands of similaritiesamong the things called works of art.
The main claims of this paper will be that:
1. Works of art come to have meaning in one or more of several
definable ways.
2. The ways in which works of art have meaning, once defined, enable
questions and answers about "the meaning of a work of art" to be put
more precisely.
3. The ways in which works of art have meaning are among the
strands of similarities enabling us to recognize works of art as such; and
thus open up the various avenues of understanding and "appreciating"
we believe should be accessible to the educated person.

What makes a thing a work of art? And can we answer this question
without the help of a definition of art?
Certain things, through the usual processes of education, come to be
recognized by viewers as works of art simply by their noting that this is
what the works are called by others. And other things get recognized as
works of art in virtue of similarities to already recognized works of art.
One kind of similarityis the similarityin the ways works of art come to
have meaning or are meaningful. There are of course other kinds of
similaritieswhich "stamp"things as works of art, but here we are interested only in the "meaning"similarities.
This manner of recognition has the ring of circularity about it. But
consider how we recognize and value games. After a snowstorm, one
child says to another, "Let's see how many times we can each hit that
tree with a snowball, three throws each." The game is recognizedquickly
and is usually valued and attended to for one or more of a number of
reasons: it is competitive, some skill is involved, some chance is involved,
score can be kept, etc. In a very different setting, such as a stockbroker's
office, a man who buys and sells his securities, keeps his score, etc., can
also be said to be playing a game, although there is no competition in

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



the usual sense, very little skill according to some, and (unlike in the
snowball game) a great deal of money can be "made." The gamemaking reason in this case is simply the score-keeping.
Analogously, we often recognize or learn to recognize works of art
quickly. Works of art have many art-making characteristics.And some
of these characteristicshave been selected singly or in combination by
theorists as the art-making characteristics,i.e., the definition or essence
of art. But having bracketed out the problem of defining art, we are
here selecting one of the art-making characteristics, namely meaning, for
analysis. The value of the analysis, in addition to its intrinsic value as
analysis, is the clarification it can introduce into discussions about the
meaning of works of art. And incidentally, but importantly for education, a work of art, once stamped as a work of art by virtue of its way
of meaning, is open to valuable ways of attending to it other than
through its meaning.
What are the ways in which works of art have or convey meaning or
become meaningful?
A. Relational Meaning

I. A. Richards said, in an anticipation of what philosophers have

identified as emotive meaning, "Many arrangements of words evoke
attitudes without any references being required en route. They operate
like musical phrases."l Clement Greenberg has written:
This sort of view has led to theories of art (and moral behaviour) as emotive
behaviour and as the conveyor of "emotive statements," theories which have
been at times widened to include effects other than emotions, e.g., images and
thoughts. This is not to say that earlier theorists in the history of aesthetics
have not recognized the connection between art and its causes or consequences,
but that from Richards on there begins a line of development in which these
causes or consequences became the meaning of art. Plato had recognized that
the poet's peculiar power is "enthusiasm" rather than "wisdom." But after
Richards it became easy to say that the wisdom of the artist consists in his
Picasso, Braque, Mondrian, Miro, Kandinsky, Brancusi, even Klee, Matisse
and Cezanne derive their chief inspiration from the medium they work in.
The excitement of their art seems to lie most of all in its pure preoccupation
with the invention and arrangement of spaces, surfaces, shapes, colors, etc., to
the exclusion of whatever is not necessarily implicated in these factors.2
Here emotions, feelings, excitements become the meaning of the work

I. A. Richards, Principles of Literary Criticism (New York: Harcourt, Brace,

1934), p. 267.

2"Avant-Garde and Kitsch." Partisan Reader 1934-44 (New York: Deal

Press, 1946), p. 381.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



of art. No meanings are "necessarilyimplicated" in colors and surfaces.

Yet a definite and understood sense of the term meaning in the arts is
being used. We will call this sense of the term: relational meaning. A
work of art may obviously have more than one recognized relational
meaning. How do we know when a relational meaning is being asked
One form in which this kind of meaning is referred to is in a question
such as: "What is the meaning of art?" and in its answer. This particular question is unambiguous,so that when Ananda K. Coomaraswamy
in Introduction to Indian Art said that
"Art in India" and "art" in the modem world mean [my emphasis]two
very differentthings.In India it is the statementof a racial experience...
The names and peculiaritiesof individualartistseven if we could recover
them would not enlightenus: nothing dependsupon genius or requiresthe
knowledgeof an individualpsychologyfor its interpretation,3
it is clear that the writer was comparing Indian art with modern art
with respect to their differences within the relationships constituted by
the mutual influence of art and a (social and individual) psyche.
Similarly when Dewey defined art as organization of energies, the
definition consisted in part of the relating of works of art to "those
features that make any experience worth having as an experience."This
of course was not Dewey's account of meaning in art in the sense in
which he claimed art expressesmeanings, but it was his account of what
is sometimes called the meaning of art and of what we are here terming
a relational meaning of art. In the latter sense the work of art does not
tell us about organizationof energies; it makes no meaningful statement.
It organizesenergies; and the statement of this fact is a statement about
a relational meaning. Thus in Wordsworth's"Prelude,"the lines
... the wind and sleetyrain,
And all the businessof the elements,
The singlesheepandthe one blastedtree,
And the bleakmusicfromthat old stonewall
build a sense of desolation in the reader but do not discuss the "building" of the mood. Dewey however did discuss the building and his
statement about it was a statement of a relational meaning of the
Another instance of "relational meaning" is implicit in such "purist"
theorists as Hanslick, Bell, and de Gourmont. When the particular arts
in which they were mainly interested have no cognitive content it is
Introductionto Indian Art (Madras:Theosophical,
'A. K. Coomaraswamy,
1923), p. v.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



because the aesthetic relevance of what we are terming relational meaning is denied by them. Works of art may have causes or consequences
but these theorists held that such causes or consequences are irrelevant
to the work of art as work of art.
Hanslick, for example, was concerned with destroying the notion of
emotive meaning (before the latter became the philosophical concept
used by Ogden, Richards, Ayer, Stevenson and others). To define the
relationshipsobtaining between a given piece of music and its causal or
consequential emotions, Hanslick thought, was to illuminate no part of
that work of music.4 In painting, Bell, although he used the phrase
"esthetic emotion," considered that emotion to be "unrelated to the
significance of life"5 or to the emotions of life resulting from such painting. And in literature where it is more difficult to hold such an extreme
view (since the basic materials of literature are words), Remy de
Gourmont, theorizing Mallarme's desire "to put some smoke between
the world and himself," said "Mallarme's work is the most marvelous
pretext for reveriesyet offered men weary of so many heavy and useless
What they had in mind (and this is a usage which is sufficiently
subject to definition to sanction its qualified use) when they denied
meaning to a work of art was, at least in part, that the work of art had
in fact no relevant emotional, image-al, or intellectual consequences or
causes. This denial suggests the theoretical possibility that if a work of
art should have relevant consequences, then those consequences would
be its (relational) meaning. For them the work of art had either a
musically intelligible structure or a sensuously attractive sound (which
were considered by them to be relevant to the work as art), but it had
no meaning since it was not relatable to an element outside of the work
of art.
Now the concept of relational meaning assumes no definition of the
work of art, or of the essentially aesthetic part of the work of art. It
takes the work of art as it occurs. For example, in the Brahms Double
Concerto in A Minor it notes certain consequences,such as are indicated
in comments by Specht, who describes it as "one of Brahms' most
inapproachable and joyless compositions";7and in Tovey's characterization of the same work as having "vast and sweeping humor" which
Edward Hanslick, The Beautiful in Music (New York: Novello, Ewer and
Co., 1891).
Clive Bell, Art (London: Chatto and Windus, 1914), p. 26.
6 Remy de Gourmont, Decadence
(New York: Harcourt, Brace, 1921), p. 155.
7Richard Specht, Johannes Brahms (London: J. M. Dent and Sons, 1930),
p. 300.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



some think "the most deadly crime possible to a great work"8 of music.
Both of these may be verifiable consequences empirically correlated with
and so related to the work of music. Taken with all the other reactions
to the Double Concerto, in their aggregate, these constitute the relational meanings of the Double Concerto. Which relational meanings are
justified can only be determined by the continuous historical criticism
that is being applied to the Double Concerto. That different and even
contradictory relational meanings coexist seems to me a matter primarily
for musical history and sociological determination, rather than aesthetic
determination, since the relational meanings do not depend upon aesthetic analysis of notes which do not change from generation to generation (except in performance interpretations), but on how people react
to those notes: what emotions are aroused in them, what images, if any,
they form to accompany the music, what arguments the music suggests.
This of course does not preclude the possibility that a changed or
deepened understanding of the Double Concerto's structure will alter
such reactions. The assertion of relational meanings emphasizes the fact
(once disputed) that art exists in no vacuum free of a social atmosphere. Whatever pure or intrinsic qualities a work of art may have, it
has causes and effects, factors influencing it and influenced by it, ideas
which bring it about and ideas brought to life by it. A complete statement of all the relational meanings of a work of art, if this were a
possibility, would be an account of its significance in the life of which
it is a part. If the artist's breakfast can actually be indicated as being
responsible for some elements in the work of art, then the statement of
that responsibility would be a relational meaning. When some other
factor (e.g., the artist's attitude towards some aspect of life) is responsible for elements of the work of art, then a statement of that responsibility could be called a relational meaning, perhaps a more important
B. Interpretive Meaning

A second usage of the word meaning in the arts can be called "interpretive meaning."
For the purpose of framing a preliminary definition, let us consider an
illustration of interpretive meaning. Take Freud's elaboration on how
art gets to mean something. In this elaboration is contained an implicit
assertion of a particular sort of meaning-claim for works of art:
... to those who are not artists the gratification that can be drawn from the
8D. F. Tovey, Essays in Musical Analysis III
Press, 1936), p. 145.

(London: Oxford University

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



springs of phantasy is very limited; their inexorable repressions prevent the

enjoyment of all but the meagre day-dreams which can become conscious. A
true artist has more at his disposal. First of all he understands how to
elaborate his day-dreams so that they lose that personal note which grates
upon strange ears and become enjoyable to others; he knows too how to modify them sufficiently so that their origin in prohibited sources is not easily
detected. Further, he possesses the mysterious ability to mould his particular
material until it expresses the ideas of his phantasy faithfully; and then he
knows how to attach to this reflection of his phantasy-life so strong a stream
of pleasure that, for a time at least, the repressions are out-balanced and
dispelled by it.9
Now it seems to me that whatever verdict we pass upon the truth of
this statement we must recognize that it indicates one sense in which
works of art make a meaning-claim. It asserts that a work of art gives
us in some way some particular information, in this case "the ideas of
[the artist's] phantasy." To the question: "What does this work of art
mean?" Freud has written:
Stefan Zweig ... has a story which he calls Vierundzwanzig Stunden aus dem
Leben einer Frau (Four and Twenty Hours in a Woman's Life). This little
masterpiece ostensibly purports only to show what an irresponsible creature
woman is, and to what excesses, surprising even to herself, an unexpected experience may drive her. But the story tells far more than this: when it is subjected to an analytical interpretation it represents without such apologetic
tendencies something quite different, something universally human or rather
masculine. And such an interpretation is so obvious that it cannot be denied.
It is characteristic of the nature of artistic creation that the author, who is
a personal friend, was able to assure me that the interpretation given by me
was completely alien both to his mind and his intention, although many
details were woven into the narrative which seemed expressly designed to indicate the secret clue ... 10
And Freud proceeds with a thorough psychoanalytic reading of the
story. We are not interested in the truth of any part of Freud's statement about the meaning content of Zweig's story. We are considering
the claim implicit in it that the work of art tells us something, even if
the mediation of a theory is necessary, and also the form in which Freud
maintains this claim: the form of significant discourse. He apparently
is saying something we understand when he says: "Zweig's story really
means . . . ," etc.

Here of course some proponent of a "story-teller" theory of fiction

could make the following objections:
9Sigmund Freud, A General Introduction to Psychoanalysis (Garden City:
Garden City Publishing Co., 1938), p. 327.
Sigmund Freud, "Dostoevski and Parricide," Partisan Review, Vol. 12, No.
4 (Fall 1945), 530.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



1. That the interpretive meaning here is gratuitous, i.e., that the

meaning that Freud delineates is the meaning of perhaps an interesting
story but not Zweig's story which is a mere recounting of what has
happened or might happen;
2. That the interpretive meaning here is a formal consideration involving an elaborate theory and its possible connection with reality, i.e.,
that one could easily predict the meaning of "Freud's"story from the
general claims of his psychoanalytictheory, and that this is an interesting
but playful deductive procedure from a theory that perhaps has some
3. That the interpretive meaning is not what the artist intended, i.e.,
that Zweig did not intend to convey the meaning of which Freud
speaks; and
4. That one does not apprehend this interpretive meaning if one
comes to the conclusion that Freudian theory is scientifically unsound.
A consideration of these objections serves to bring out the formal
characteristicsof the kind of meaning involved here. Answering them in
1. Freud insists that "many details were woven into the narrative
which seemed expressly designed to indicate the secret clue," i.e., his
procedure is to be empirical about this Zweig story and to relate certain
details of it to certain theoretical constructions which make the details
more meaningful, i.e., more communicative of certain facts--a procedure with which one can in principle hardly argue. This procedure
allows Freud to make a statement of the form "A means B" where
Zweig's story as it appears literally means Zweig's story as it appears
interpretivelyor in its interpretivemeaning. Thus in principle Freud is
constructing another story only in the sense that we end up, if we are
convinced, with another more accurate apprehension of what Zweig's
story has to say than we would have had without the Freudian interpretation. We are led in this case with the help of Freudian theory to
determine what the story "really has to say," that is, what, on the
assumption of a more or less plausible theory, it can be said to mean
2. It is true that we would probably know what sort of answer a
"Freudian aesthetician" might give to the question "What does 'Four
and Twenty Hours in a Woman's Life' interpretively mean?" if we
knew beforehand that he was a "Freudian aesthetician," that is to say,
that he adhered generally to the theory stated by Freud above. But

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



there is no doubt that the word "meaning" as applied to the arts is

being used here significantlyboth in question and answer: on the basis
of one interpretation Zweig's story, when read in the light of theory
having some degree of probable truth, tells us many facts about the
(conscious or unconscious) wishes of the artist. And presumably if the
artist were to agree to the interpretationFreud makes he would have a
greater degree of self-knowledge. His story would interpretively mean
something to him, i.e., give him some information about himself. We
could leave it as optional whether or not we generalize this knowledge
into "something universally human or rather masculine," an extension
of the scope of the meaning not necessary for our present argument.
If we have more information about Zweig after reading the story (with
the Freudian interpretation in mind or imposed later upon the story)
we would say that his story interpretivelymeans at least this additional
3. Whether Zweig did or did not intend to convey certain information is not important except biographically, or psychologically; just as
when I say "It is raining" it is not important for a determination of the
meaning of those three words to ascertain whether or not I intend to
convey information, true or false. As a matter of fact Freud is asserting,
however paradoxical it may sound, that Zweig is saying something interpretively meaningful which for reasons of inner moral censorship he
cannot make literally meaningful. And this must be considered for what
it is worth as a scientific claim.
4. The meaning that is conveyed to Freud is obviously dependent
upon the existence of an elaborate psychological theory. But this is
sometimes true of the simplest linguistic statements. In discussing problems of translatability,Malinowski remarkedupon the necessity in some
cases of understanding a whole culture and its history before understanding the meaning of one sentence of that culture's language.1l
C. Contextual Meaning

We have often heard it said that ordinary language states meanings

while art expresses meanings. And some theorists have claimed that
whatever the differences between "stating" and "expressing,"they are
both forms of communication and that art is in some way a superior,
more effective form of communication than ordinary language. On the
basis of the latter claim we can be led to the unusual position of asserting
" C. K.
Ogden and I. A. Richards, Meaning of Meaning, 8th ed. (New York:
Harcourt, Brace, 1944), pp. 301-02 (Supplement 1, B. Malinowski).

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



that a statement in ordinary language such as "There sometimes seems

to be a spiritual entity pervasively present in our experience of natural
objects" is less communicative than lines 93-102 of "Tintern Abbey":
And I have felt
A presencethat disturbsme with the joy
Of elevatedthoughts;a sensesublime
Of somethingfar moredeeplyinterfused,
Whosedwellingis the light of settingsuns,
And the roundoceanand the livingair,
And the blue sky,and in the mind of man:
A motionand a spirit,that impels
All thinkingthings,all objectsof all thoughts,
And rollsthroughall things.
In one sense of the word expressive,Wordsworth'spoetry is obviously
more expressive than the more or less literal translation above: certain
feelings are much more vividly expressed in the poetry than in the
"translation."But this is an evincing and an evoking more adequately
described in terms of relational meaning.
Much more is being created and/or recreated in the poetry than
specific feelings. Let us examine the poetry in the light of a sort of
meaning we may call contextual. Some experiences we have give us an
awarenessextending beyond the moment and revealing to us the sources,
movement, and culmination of the sorts of interactions we have with
our (physical or cultural) environment. When an object or event is
contrived to present and convey such an awareness, we call the contrivance a work of art. And what it is contriving is what we are calling
contextual meaning. Now let us take another look at "Tintern Abbey":
Five yearspast: fivesummers
with the length
Of fivelong winters!And againI hear
These waters....
With these lines Wordsworthbegins his poem, and later in the poem we
find that those five years this landscape has been of special value to the
poet: it has given him serenity and pleasure in an otherwise chaotic
world. He finally comes to feel the presence of a source for that serenity
and pleasure. When we have lived through Wordsworth's poem and
have finally felt the expressedquality near Tintern Abbey we can reconstruct two elements which stamp the work as art:
1. We become at least aware of, even if we do not completely share
Wordsworth'sprior values (prior to the expressiveact which eventuated
in the expressiveobject, the poem), visual and emotional states.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



2. We experience the existence of a new objective entity: the poem

which in its words, sounds, and rhythms embodies a new experience.
It is certainly possible to imagine Wordsworth living through those
five years hearing "these waters," revisiting and actually lying down in
the sun to


These waters, rolling from their

With a soft inland murmur...,
and to:


These hedge-rows, hardly hedge-rows,

little lines
Of sportive wood run wild ..


without ever having written "Tintern Abbey." There would then certainly have been an experience, but none of the contrivances of poetry
would have been used to create an ordering of events designed to reveal
either to Wordsworth or his possible readers the total context within
which specific events would "add-up" to mean something. The poem
does in fact do the latter and tends to convince the reader of the existence (in some important sense) of the sort of presence he senses.
To be aware of a contextual meaning we do not need an interpretive
theory of the sort which generates an interpretive meaning. But the
distinction between contextual meaning and relational meaning is not so
obvious. The events of the past which may be apparent in the hushed
reverberations (in Santayana's phrase) of the present experience of the
work of art, are events which upon analysis we could theoretically relate
to the work of art. In a contextual meaning there are "these waters"
outside of and related to the poem, and we have apparently simply
another relational meaning. The difference lies in the importance of
context in a contextual meaning. Cezanne's Sainte-Victoire may have
the contextual meaning ascribed to it by critics who speak of its revelation of spiritual forces or of the manner in which it echoes the rhythm
of the universe. But none of its relational meanings could be stated as:
this is the way the world outside this frame can be seen - a world of
spatial depth through nuanced surfaces, where the painting simply indicates this way of looking at the world with no construction of a universal
context. Yet this way of looking at the world is not unrelated to such a
possible context. Any element which is related to the work of art in a
relational meaning may be brought into a context of other elements

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



such that where previouslythat element was directly related to the work
of art, it would, in a contextual meaning, take its place in a meaningful
context having a quality and significance of its own. The painting properly has both meanings- one of the values in recognizing both meanings is that there then should be no dispute about which is the meaning.
D. Referential Meaning

It is quite possible for the products of human activity to convey

meaning without there being an understanding of how such an activity
conveys the meaning. That Rubens in The Battle Between Emperor
Constantine and Maxentius conveys a meaning is obvious to everyone
who looks at such an allegorical scene. But this is not as obvious in a
Mondrian painting. And yet while it may be true that this is so because
the Mondrian painting conveys no meaning in fact, it may possibly be
the case that since we have not been sure how the Rubens painting
conveyed meaning that therefore we have no method by which to
determine the meaning of the Mondrian painting. We can attempt to
define such a method in a way continuous with ordinary usage of the
word "reference."
In the sciences, we find a set of symbols, signs, or marks used in
definable ways to refer to certain aspects of their spheres of interest, i.e.,
we find a method for "telling about" something. The obvious search, if
art is to have any referential meaning, is for a defined system of marks
and sounds by which it may refer to aspects of its sphere of interest.
By way of contrast to this search for method we can note that Plato,
the source philosopher of many imitational, representational,and referential theories of art, was not interested in defining the method of reference employed by the artistic endeavor. He was concerned with pointing
out that "the soul which has seen most of truth shall come to the birth
as a philosopher, or artist, or some musical and loving nature" and at
some later time, in a way beyond complete control, recall that truth.
In the Apology Plato knew "that not by wisdom do poets write poetry,
but by a sort of genius and inspiration; they are like diviners or soothsayers who also say many fine things, but do not understand the meaning of them." That is, they wrote meaningful poems but did not know
by what method the poems came to have meaning.
Let us consider Mondrian's Composition in Grey, Red, Yellow, and
Blue. What we see offhand is a group of rectangularblocks of specifiable
primary colors- the boundaries of the blocks being vertical and horizontal lines. And let us call that a description of the surface of the
painting. At this point let us temporarilyhalt the analysis and see if we

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



can derive anything from this description that could be called meaning.
We now notice that we have not stepped outside the frame of the picture
except in the very weak sense that we have used words for the description of this painting which are used in other (artistic and nonartistic)
contexts. And we have in principle fully described the surface of this
work of art. We could go on to specify the size, relative position, and
relative hues, intensity, and values of the colors, i.e., all the objective
qualities of the painting.
There is, one may notice, the fact that using words like "yellow" or
"vertical" immediately introduces something exterior to the painting or
outside its frame. A whole group of comparisons to events and objects
outside the frame is made possible by the use of words which are in their
nature generalized so that they may apply to outside events and objects.
These words tend to pull us away from the surface but only in an unimportant because undirected way. If someone were to say, for instance,
that this yellow is like the yellow of many bananas he has seen and that
this painting therefore is about some aspect of bananas, we could only
ask for some rule or direction by which one could go from this yellow to
those bananas. In the absence of such a rule or direction (method) one
could say only that between the description of this work of art and the
description of the experience of a banana there exists only the word
"yellow" describing some element common to the experiences. Thus we
are left on the surface, or, more precisely, in our experience of the surface, despite the use of the word "yellow," because we have no rule by
which to leave the surface. Yet leaving the surface in an undirected way
means that we could derive some meaning (and this may be enough or
all there is in some works of art). These would be relational meanings
such as "this work of art provides perceptual delight of a certain sort,"
or "this work of art reminds us of many previous pleasant experiences
of bananas." But here no direction is given to the way in which we
leave the surface. Reactions differ from observer to observer, from age
to age, and there is no authority as to which reactions must, notwithstanding the artist's intention, be had. The sense configuration he produces may have effects other than he anticipates, intends, or desires, and
there is nothing improperin this happening. A variety and large number
of relational meanings may occur (and this has sometimes been used as
a measure of value of the work of art).
Let us further notice that there are no diagonals or curved lines in the
painting. All lines are straight, vertical, or horizontal; and this determines the fact that all the shapes are rectangles.This is obviouslynot like

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



the visual field, as the purity of color was like the calmness of the spectator in the visual field. Then what is its relation to the visual field?
Impressedby the vastnessof nature, I was trying to expressits expansion,
rest,and unity.At the sametime, I was fully awarethat the visibleexpansion
of natureis at the same time its limitation;verticaland horizontallines are
the expressionof two opposingforces;these exist everywhereand dominate
everything;their reciprocalaction constitutes"life." I recognizedthat the
equilbriumof any particularaspect of naturerests on the equivalenceof its
opposites.I felt that the tragicis createdby unequivalence.I saw the tragic
in a wide horizontalor a high cathedral.l1
From Mondrian's composition to the struggle against the tragic might
seem like a long jump, but in this painting we can see the preparations
for, execution of, and ensuing excitement of a very methodical leap.
Looking at this composition with Mondrian's view of "nature expanding" in all vision, we can see it as a visual field of no particular, determined origin or location. In this field we must see a necessarylimitation
if we are to see anything, any form. We see this limitation by means of
verticals and horizontals (curved lines could have accomplished the
same effect to some degree, but the limitations they suggest are not as
abrupt or dramatic as a straightline block).
What we have seen in this painting, as describedin the last paragraph,
is not a matter of simple vision or survey of the surface of the canvas.
It required seeing the verticals and horizontalsof this painting as limitation of the "visible expansion of nature," i.e., in terms of a concept
logically previous to the painting. Mondrian is of course not talking
about the universe (which may actually be expanding but certainly
not visibly). He must be understood as talking about the visual field
and its characteristics.
A concept "the visible expansion of nature" was thus one of the
causes of this painting and is also one of the relational meanings of the
painting. But it is also the pre-work-of-artconcept motivating the painting. This concept has a content made evident by the painting: the visual
field naturally expands when we open our eyes; it expands until it is
limited by two basic sorts of lines, verticals and horizontals; when verticals and horizontals are seen as forces they are sometimes seen to be
in balance, thus giving their visual fields qualities of rest and calm
which may be felt by a spectator. It is possible that this concept be a
constituent of a relational meaning of the painting and still be an inadequate pre-work-of-art concept. For instance there undoubtedly are

Piet Mondrian, Plastic Art and Pure Plastic Art (New York: Wittenborn,

Schultz,1945), p. 13.

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions



some spectators who, given the pre-work-of-art concept of this painting

might not have feelings such as would make the concept evident in this
painting. By virtue of the visible resemblance between the verticals and
horizontals of this painting and those seen in nature, one could say that
Mondrian has been painting nature but distorting it to extreme. This of
course is true but says very little: it says merely that "remote resonances"
are discernible. The questions "Why distortion?" "Why to extreme?"
bring to focus the method Mondrian is using. He is selecting elements of
a plastic language: straight vertical and horizontal lines and pure color
to point in a definite (its definiteness establishes to its extent the existence of a language) and unique (what make this work of art new) way
to certain "specific sections of reality": expansion, forces, equilibrium.
He points to those facts by giving us a particular instance of exemplification, in terms and materials of an art form, of a concept logically
previous to this painting.
Mondrian's painting now can be said to mean referentially (whatever
else it means) that a visible expansion occurs in visual fields and that
this expansion encounters certain characteristic obstacles (verticals and
horizontals) and has certain resolutions in a "dynamic equilibrium."
This meaning is presented in the painting by means of a pointed exemplification of such logically prior concepts as "visible expansion," "obstacles," "dynamic equilibrium" and "pure reality."

This content downloaded from on Sat, 28 Nov 2015 06:50:54 UTC
All use subject to JSTOR Terms and Conditions