100% found this document useful (4 votes)
487 views132 pages

Aesthetic Theories of French Artists - From Realism To Surrealism

Uploaded by

Myra Elisse Anit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (4 votes)
487 views132 pages

Aesthetic Theories of French Artists - From Realism To Surrealism

Uploaded by

Myra Elisse Anit
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 132

The

(iAesthetic Theories

ofYrench <iy[rtists

From Realism to Surrealism

by

Charles Edward Qaf^


Johns Hopkins Paperbacks JH-23 i^
/

THE AESTHETIC THEORIES OF FRENCH ARTISTS


THE AESTHETIC THEORIES
OF
FRENCH ARTISTS
FROM REALISM TO SURREALISM

BY
CHARLES EDWARD GAUSS

THE JOHNS HOPKINS PRESS


BALTIMORE and LONDON
Copyright © 1949, 1966 by The Johns Hopkins Press
All rights reserved. No part of this book may be
reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying,
recording, xerography, or any information storage and
retrieval system, without permission in writing
from the publisher.
Manufactured in the United States of America

The Johns Hopkins Press, Baltimore, Maryland 21218


The Johns Hopkins Press Ltd., London

ISBN 0-8018-0215-6 (clothbound edition)


ISBN 0-8018-0216-4 (paperback edition)

Originally published, 1949


Second printing, 1950
Third printing, 1957
Fourth printing, 1963
Johns Hopkins Paperbacks edition, 1966
Second printing, 1970
Third printing, 1973
To
MY PARENTS AND MY WIFE
who encouraged and aided

my education

and were ^patient through the long process


PREFACE
This study is a modest attempt to trace some of the
connections between art and other branches of human
endeavor and culture in the history of ideas. Its thesis is
that one should examine the history of ideas as it is
manifested in art.
I have included in my discussion only those artists and
those writings which I deemed sufficient to trace the course
I wished to outline. If the study is of any value others
better equipped than I may fill in the outlines.
In rendering the French quotations in English I have
made my own translations, though often better ones are
available, in order that any responsibility for misunder-
standing may be laid only to me.
I wish particularly to thank Dr. George Boas and Dr.
Lionello Venturi for their kind aid and interest in the
present undertaking. I am also grateful to them and to
Dr. Pedro Salinas and Miss Etta Cone for having made
available to me materials from their libraries which it

would have been otherwise difficult or impossible to obtain.


Acknowledgment is made to The Journal of Aesthetics
for granting me permission to incorporate here materials
published in that journal on surrealism.
Acknowledgment is also made to the following copyright
holders from whose works passages have been quoted in
this book: The University Chicago Press for selection
of
from D. C. Rich, Seurat and the Evolution of " La Grande
Jatte"; The Dial Press, Inc. for selections from C. J.
Ducasse, The Philosophy of Art; Dodd, Mead and Co.,
Inc. for selection from C. G. Jung, Psychology of the Un-
conscious; Faber and Faber Ltd. for the selections from A.
Breton, What is Surrealism?; Mr. Julien Levy for the
passage from S. Dali, Conquest of the Irrational, and the
vii
viii PREFACE

passage from J. Levy, Surrealism; The Open Court Pub-


lishing Co. for the sentences from B. Russell, Our Knowl-
edge of the External World; and to Mr. Walter Pach for
the selections from his article " Pierre-Auguste Renoir " in
Scribners Magazine for May 1929.
Translations have been made from materials held by the
following copyright owners: Art Catholique (L. Rouart
and J. publishers of C. Cennini, Le livre de Vart,
Watelin) ,

translated by Victor Mottez, and M. Denis, De Gauguin


et de Van Gogh au classicisme, theories 1890-1900; Beaux
Arts, Dictionnaire abrege du surrealisme^ and R. Rey, La
renaissance du sentiment classique; Jean Budry et Cie.,
T. Tzara, Sept manifestes dada; G. Cres et Cie., Jean de
Rotonchamp, Paul Gauguin 1848-1903; Durand-Ruel, L.
Venturi, Les archives de Vimpressionnisme; Figuiere, A.
Gleizes and J. Metzinger, Du cubisme; Librairie Floury, P.
Signac, D'Eugene Delacroix au neo-impressionnisme; Lib-
rairie Gallimard, A. Breton, Le Surrealisme et la peinture;
f^ditions Bernard Grasset, P. Cezanne, Correspondance,
edited by J. Rewald; Journal Havre-tlclair, issue of Sep-
tember 25, 1904 containing statement of Pissarro; M.
Henri Matisse, whose article, " Notes d'un peintre " ap-
peared in the Grande Revue for December 25, 1908; H.
Laurens (Librairie Renouard) E.Moreau-Nelaton, 3/an^t
,

raconte par lui-meme and H. Focillon, La peinture du


XIX^ et du XX^ siecle; Editions du Sagittaire, A. Breton,
Manifeste du surrealisme and Second manifeste du sur-
realisme. I thank all of these who gave me their kind
permission to print my translations.
And to all those who aided me with their comments or
in the preparation of the manuscript I extend my grateful
appreciation.
Charles Edward Gauss
The George Washington University
PREFACE TO THE PAPERBACK EDITION
The appearance of this book in a paperback edition is
particularly gratifying to me. When I originally prepared
the book for publication, I had in mind the general reader.
Consequently, I translated the quotations into English
and omitted scholarly cant as much as possible. I can
now, with the appearance of a popularly priced edition,
look forward to the realization of my original aim to reach
a wide reading audience.
It has been seventeen years since the book originally
appeared. Even then, it covered no material beyond the
mid-thirties of our century. Much has happened in the
world of art since then. Styles have come and passed.
Surrealism, the last movement I touch upon, is now practi-
cally dead. And yet my book has met with a steady
response that induced the publishers to reissue it, most
probably because it deals with the aesthetic ideas of artists
in that interesting century in France that embraces the
rise of realism with Courbet and Manet, the impressionist

and post-impressionist developments, and the dada and


surrealist movements. This was the century in which
painting, losing to the camera its older purpose of making
visual records, was searching for a justification of itself
through a new aesthetic. The elements of this new aesthetic
were being worked out in the various stylistic movements
just mentioned. By giving a reader some understanding
of these, I hope that my book will suggest to him how he
might look at the art of today, which is the flowering of
that new aesthetic. One cannot just read aesthetics and
appreciate art; art is meant to be looked at. But the new
painting can only be looked at correctly if one is given
some guidelines from the past.
Also, I hope that my book will suggest a sobering idea
X PREFACE TO THE PAPERBACK EDITION

to those who have not yet disembarrassed themselves of


a philistine attitude. I hope it will suggest that creativity
demands an ongoing development and that often the con-
tributors to a constituent tradition in the arts are not the
academics but the avant garde men, sometimes even when
they are throwing out nonsense —at least such was the
case among the French artists from Courbet to the
surrealists.
And last, I hope that my book will suggest that paint-
ing is not an enterprise isolated from all other cultural
endeavors, but is simply one of the ways, like all else he
does, in which man philosophically reflects upon the world
and his knowledge of it.
C.E.G.
Nova University,
Fort Lauderdale, Florida
TABLE OF CONTENTS

I Introduction 3

II The Realism of Courbet 9

III From Realism to Neo-Impressionism . 19

IV The Great Independents: Renoir and


Cezanne 35

V Symbolism and Fauvism: Gauguin, Denis,


Matisse 53

VI Early Cubism 69

VII Surrealism 79

VIII Coda 96

Bibliography 100

Index 109
I
THE AESTHETIC THEORIES OF FRENCH ARTISTS
CHAPTER ONE

INTRODUCTION
ARTISTS have frequently written about their art.
-tx. Numerous come to mind. The Treatise of
instances
Alberti on painting and the Notebooks of Leonardo da
Vinci are famous among the writings of the ItaHan painters.
Hogarth's Analysis of Beauty and Reynolds' Discourses on
Art came at the time of awakening interest in aesthetics in
England. There are posthumous publications of the
Journals of Delacroix, Redon, Whistler, of the correspon-
dence of Van Gogh, Gauguin, Cezanne. In the twentieth
century books and essays written by artists have multi-
plied extensively. Indeed, it has become quite the accepted
thing for artists today to publish some statement of their
aims or an explanation of their aesthetic, especially when
theirwork is of a new or strange character.
The literary preface is common for the professional writer,
and surely we should find nothing amiss in the extension
of the use of the preface to the plastic arts. If anything, it

is logically more necessary to have a declaration of artistic


intentions for works in this field than for literary efforts.
Literature uses the universally comprehensible language
of speech, but the medium of painting is not so readily
understandable to the average person. The fact that an
artist states his aesthetic credo indicates that he does not
trust the unaided sensibilities of the spectator before his
work. The arts have become sophisticated like the sciences.
The naive attitude of the average man is no longer suffi-

cient to comprehend them.


The naive attitude is that which contents itself with
whatever minimum mental effort is necessary to grasp the
4 THE AESTHETIC THEORIES OF FRENCH ARTISTS

practical life. axioms


It accepts the space of Aristotle, the
of Euclid, and snapshot art/ These seem to have the
advantage of being substantiated fundamentally in daily
experience; and their acceptance becomes a habit of mind,
tenaciously held and rarely questioned. Toward modern
science the average man is occasionally willing to admit
the insufficiency of his naive attitude. Awed by the con-
crete technological advances of science during the last two
centuries, he is willing to assume a docile and uncritical
attitude towards the new physics. Quantum theory and
relativity have merited his respect. Before them his atti-
tude becomes submissive. On the other hand, before
modern art, which his naive attitude renders him incapable
of understanding, the average man is anything but sub-
missive. The art of the new movements of the twentieth
century call forth his immediate reactions. Art which
cannot be understood is not to be left alone until one does
grasp it. It must be judged immediately, judicially.

Though we must recognize the existence of such an


attitude, we must not let this naivete be the court of
judgment for our art. It must be educated and refined, at
least to where it expresses that same humility that it
shows towards new scientific theories.
To know the questions of modern science we go to the
chemist, the physicist. To learn the questions of modern
art we must go to those who deal with them. We must see
the problems presented by the painter's materials, tech-
niques, and objectives, and we should understand the
various solutions of these problems. The artists them-
selves can best present these problems to us.
The study and pronouncements
of artists' manifestoes
has not, of course, been neglected. In art histories and
monographs on artists they have been examined for the
purpose of setting in clearer light the producing personali-
^

INTRODUCTION 6

ties and of better understanding their artistic productions.


Critics of art have referred to them for similar reasons, or
to support their own artistic judgments by comparison
or contrast. But it is possible to use these writings as
materialsupon which to base a further study. We may
probe back to the more fundamental philosophical ideas,
the metaphysics as it were, upon which the individual
taste of an artist or the do^rine of a movement in art
is founded. There are often certain leading ideas, presup-
posed notions, or tenets of faith which are like guiding
lines for much of the human activity of a time, being
observable in thought, science, and art. The theories of
the artists thus become materials for the manifestations of
ideas in history.
The legitimacy of basing a philosophical inquiry upon
the words of artists may be challenged. These theories
were not written as philosophy and whatever philosophical
content they have will be meager and poorly stated. Philo-
sophical questions are for the philosophers, not the
painters, to solve. Curt Ducasse ably expresses this
criticism:

When questions of this sort present themselves people generally


assume as a matter of course that, of all persons, the artist must be
the one best able to answer them. But the artist's business is to
practice art, not to talk about it. Indeed, an artist with a theory
should be regarded a priori with suspicion, for such an one is likely
to paint or sing with his intellectual conscience instead of his feelings,
and to give us, therefore, not works of art but moral documents.
It would not be much more foolish to expect a lecture concerning
the physiology of digestion from a man who can eat things indi-
gestible to others, than it is to expect answers to questions of the
sort mentioned from a painter on the strength of his ability to paint
pictures. But people, failing to see this, insist on making the artist
talk. They regard him as an expert, and naturally he admits being

one, he as well as his questioners innocently but unfortunately
overlooking the fact that it is at something very different from dealing
with theoretical questions concerning art that he is expert.
6 THE AESTHETIC THEORIES OF FRENCH ARTISTS

To this criticism two answers may be given. First, it


assumes that the proper business of the artist is the
making of a work of art and that all theoretical questions
about art belong to the philosopher. Such an idea that
there are some men who are doers and others who are
thinkers is the result of a false abstraction. Of course, the
artist while practicing his art is a doer, but his work has
some mental direction, and if he at all thinks about what
he does there is no reason why he should not talk about it.
Artists do theorize and we have a right to take the results
of this theorizing as materials for study.
Second, if the object in going to the artists' theories
were to find in them acceptable answers to purely philo-
sophical questions, the criticism of Professor Ducasse could
be legitimately urged against such a procedure. The phi-
losophy of an artist who theorizes may be as bad as that
of the who grows metaphysical. Speculative
scientist
philosophy is frequently upon shaky enough ground when
indulged in by philosophers, to say nothing of when others,
not so aware of weaknesses in their assumptions, add their
elaborations. However, we do not go to the theories of the
artists to find answers to aesthetic problems but turn to
them as materials for philosophic study. The purpose is

to analyze an aesthetic as viewed in the light of its pre-


suppositions.
The is a modest attempt to do this. The
present work
material used a selected group of representative writings
is

by French painters from 1855 until the beginning of the


Second World War. The study starts with the realism of
Courbet because here one finds the best link between the
art of the past and the new painting that was to come to
fruition in the twentieth century. By starting with it and
considering later theories in the light of it, the study that
follows is given the general thread of the history of an
idea.
INTRODUCTION 7

The story of French painting in the latter half of the


nineteenth century may
appear to be a series of revolu-
tions and counter revolutions. Yet it can be presented as

an intelligible pattern. From the time of the rebellion


against idealism French painting has been predominantly
realistic. Courbet, the impressionists, the neo-impression-
ists, Cezanne, were concerned with the appearance of the
external world. Renoir, the symbolists, the fauves, the
surrealists, though they may have
been primarily inter-
ested in expression or the revelation of the inner self, never
idealized their subject and retained a kind of primitive
naturalism. The cubists, though they seemed to discard
appearance for reality, were concerned with the nature of
the object as three dimensional and intellectually known.
Consequently every style of painting from the time of
Courbet may be looked upon as a variation of realism or a
reaction against its original form.
The dialectic of the historical development of philoso-
phy in France from the time of positivism is too compli-
cated to sum up in relation to the artistic development
which was much simpler. It is sufficient to notice that the
early positivism which limited knowledge to the immedi-
ately observable was soon shown insufficient and subjected
to criticismand emendation. The place of hypothesis, the
contributions of the human observer were noted and taken
into account. The reaction to the severe scientism of this
tradition began with Ravaisson and Bergson as these men
offered new interpretations of consciousness and intuition.
French philosophy divided into two schools, the scientific
philosophers and the intuitionists. Roughly this division
corresponds to the " scientific " painters, neo-impression-
ists, Cezanne, cubists, as opposed to the expressionist or
intuitionist painters, Renoir, the symbolists, the fauves,
the surrealists. Without an exhaustive study of every
8 THE AESTHETIC THEORIES OF FRENCH ARTISTS

minor development and change in French painting it is

impossible to connect the artistic and philosophical de-


velopments completely. I have preferred to keep the
thread of connection in the artistic field and then point out
such parallels with philosophy as may occur. This is
sufficient to support the contention that the aesthetic
theories of the artists do have philosophic roots or are
analogous to some ideas in philosophy.
Sometimes the relations pointed out are definitely ones
in which the artistic ideas depend on certain philosophical
presuppositions. This is the case in relation to realism.
At other times the artist in his aesthetic has consciously
borrowed and used ideas from another field. Such has
happened in surrealism. At still other times there may be
noted notions which are striking for their similarity be-
tween art and philosophy. An example of this is the
resemblance between the cubist painter's object and the
constructed object of the new realist philosophy. Since the
concern here is to demonstrate philosophic backgrounds all
three kinds of parallels are pointed out.
This history of ideas is a history of painterly problems
not of philosophical theories.

NOTES
^ The word snapshot is used instead of photographic to convey more
precisely the impression of uncritical, amateurish immediacy.
^Reprinted from The Philosophy of Act by C. J. Ducasse, p. 2,
by permission of Dial Press, Inc., copyright 1929 by Dial Press, Inc.
Also vide L. A. Reid, A Study of Aesthetics, p. 15, for another ex-
pression of this same notion.
CHAPTER TWO

THE REALISM OF COURBET


nr^HE ART world of the Second Empire was like an
*- exclusive club. It had its own standards and rules
of admission that forced the individual artist to its pattern.
There was no room for the independent or the revolu-
tionary. And so it was that in 1855 the jury of the Exposi-
tion rejected the work of Gustave Courbet. Incensed at
this rebuff, the artist rented a pavilion near the exposition
grounds and proceeded to hold his own one-man exhibition
of forty paintings, labeling his art realism. As the leader
during the fifties of the rising radical painters, Courbet
gained by this act historical if not popular notoriety and
set painting alongside of the rising new order in literature.
For the next fifteen years realism was destined to be the
center of opposition to the official and academic art of the
Second Empire.
Courbet's convictions on painting are contained in two
manifestoes. The first was published in 1855 in the cata-
logue of his independent exhibition. Though over his
signature it was most probably written for him by Champ-
fleury, the pioneerexponent of realism in literature.^ This
manifesto is very merely stating that realism is a
brief,
label applied as romanticism had been in 1830, that the
artist has studied the painters of history not to copy
them but " to know in order to be able to do," that he is
opposed to the theory of art for art's sake, and that he
wishes only to translate the manners and ideas of his
epoch, in other words " to make a Hving art." ^

In the number for December 25, 1861 of the Courrier


du Dimanche Courbet published a longer statement of his
10 THE AESTHETIC THEORIES OF FRENCH ARTISTS

theories, which had by now become crystalhzed in his own


mind, to explain his refusal to direct a school for artists
which the young realists had requested. His words sum
up the position of the realist:

I cannot teach my art, nor the art of any school, since I deny that
art can be taught, or as I maintain, in other words, that art is strictly
individual and is for each artist precisely the talent resulting from
his own inspiration and from his own studies of tradition. I add that
art or talent according to me should be for each artist only the means
of applying his own faculties to the ideas and things of the epoch in
which he lives.

Especially the art of painting should consist solely of the repre-


sentation of objects visible and tangible to the artist. Any epoch
should be reproduced only by its own artists, I mean to say, by the
artists who have lived in it. I hold the artists of one century
radically incompetent to reproduce the things of a preceding or future
century, or otherwise to paint the past or the future.
It is in this sense that I deny historical art applied to the past.
Historical art is in its essence contemporary. . . .

I also hold that painting is essentially a concrete art and does not
consist of anything but the representation of real and existing things.
It is a completely physical language using for words all visible objects.
An abstract object, one which is invisible, non-existent, is not of the
domain of painting.
Imagination in art consists in knowing how to find the most com-
plete expression of an existent thing, but never to suppose or create
that thing.
Beauty is in nature and is found in reality under the most diverse
forms. After it is found there it belongs to art, or above all to the

artist who knows how to see it. Rather, beauty is real and visible, it

has in itself its own artistic expression. But the artist has not the
right to amplify that expression. He cannot touch it without risk of
changing its nature and consequently of weakening it. The beauty
given by nature is superior to all the conventions of the artist.

Beauty, like truth, is a thing relative to the time in which it is seen


and to the individual fit to conceive it. The expression of beauty is
in direct ratio to the power of perception acquired by the artist.

There is the groundwork of my ideas on art. With such ideas, to


conceive the project of opening a school to teach conventional princi-
THE REALISM OF COURBET 11

pies would be to resume again the incomplete and banal ideas which
have up to now guided modern art everywhere.
There cannot be schools; there are only painters. Schools are only
of use to investigate the analytic procedures of art. No school can
bring one by resolution to the artistic synthesis. Without falling into
abstraction, painting cannot let a partial aspect of art dominate, be
it drawing, color, composition, or any other one of the many ways of

which only the ensemble constitutes art.^

How terrible! the must have thought upon


official artist

reading this. One cannot paint nymphs, angels, or muses.


This madman denies the artist the right to paint scenes
from classical antiquity, from the Bible, or from national
history. The artist cannot beautify his subject, for beauty
is supposed to be the bald, literal statement of fact. This
man is denying all the glorious traditions and teachings of
art. Great art has always been a thing of ideals, the result
of a noble imagination; this man, like a scientist, makes it

a work of fact. As well call a daguerrotype art!


Between this official artist and Courbet there was the
bridgeless chasm that separates the " realist " and the
•*
idealist." Idealism makes the world a construction of
our own minds and its aim is mathematical explanation
and abstraction. Realism assumes the world as given and
its aim is description and observation.

The meaning of realism does not change when we pass


from the realm of thought to art. The realist artist seeks
to omit the subjective and to reproduce the world as seen.
He is a descriptive artist who states without comment.
Realism in thought and art, therefore, are closely allied;
and it is possible to find in the world of the Second Empire
intellectual trends that parallel the rise of realist art.
The new thought of the mid-nineteenth century is posi-
tivism, which assumes that the world is the sum of those
objects that the scientific observer finds in his experience.*
Our knowledge is confined to the data of experience. The
12 THE AESTHETIC THEORIES OF FRENCH ARTISTS

purpose of science is to collect the instances of the occur-


rence of an event and to generalize them in a formula. No
metaphysical power or substance which might be the cause
of that event discoverable nor needs to be invoked. We
is

disregard what we are unable to verify empirically; " see-


ing is believing " is literally interpreted. Scientific method
brings us to the discovery of general principles which are
thus only descriptions. The mind is a passive end-organ
which receives the data and organizes them, adding noth-
ing of its own construction in the formulation of the
principles.
Realism accepts completely this positivistic outlook.
Painting is the representation purely descriptively of ob-
jects which are visible and tangible, or real and existing.
The duty of the artist, like that of the scientist, is to ob-
serve, analyze, and describe accurately by a detailed re-
production the object of nature as it is seen. Unseen
objects are metaphysical abstractions and have no place
in painting. Beauty is a perceptible quality in seen objects
and the artist simply reproduces this beauty. The im-
agination is a strictly perceiving and reproducing faculty;
it cannot create; it contributes nothing to the beauty of
the work of art. The artist is only the vehicle through
which the language of painting operates, and the more he
effaces himself to become the " slave of his model " ^ the
better is beauty translated from the natural object to the
work of art.
If the aim of art is exact reproduction of natural objects
then the purpose of art can no longer be the one that had
been usually assumed, the embellishment of the world by
ideal beauty. The new justification of art is the declara-
tion that it has a social purpose. When Courbet says that
the artist must represent his own epoch not the past or
the future, and that he wants to make a living form of art
THE REALISM OF COURBET 13

by expressing the ideas and customs of his age, he is again


voicing the sentiment of positivism and nineteenth century
criticism of art.
In 1828 fimile Deschamps had said: " Above all else
one must be of his time." ^ In the field of art criticism
after 1830 a group of writers whom we may roughly call the
democratic school emphasized the social end of art through
their Salons and journals articles. They included fitienne
Arago, Alexander Decamps, Henri Robert, Louis Dessieux,
Charles Blanc, Theophile Thore, and J. A. Castagnary.^
These men had variously held some form of belief that art
is a kind of social documentation or a vehicle for social

betterment. Either directly or indirectly all these had been


influenced by the increased interest in social problems
manifested by the Saint-Simonists or the positivists of
later generations.
In the years following 1815 Saint-Simon and the group
about him set forth what they believed were the essential
aesthetic needs of man and the services the artist can
render to society. They looked upon society as moving
toward an ideal order. Since society is an organic unit
which includes the arts within it, art will have a part in
bringing about this ideal order as well as being a unit in
the present order. In a society the learned, the artist, and
the artisan perform the work of positive utility. When the
physical and moral needs of man are satisfied, he will be
happy. In the new order to come industry will meet our
physical needs and be the basis of society. Art is the moral
expression of a society and its guide. It is the agent of
social reform through its influence in inspiring love of
liberty, or by its power to produce a social revolution.
But a noble art is impossible until after the social reform.
To see that a social order founded on industry will pro-
duce the highest art, we need only look at the flourishing
14 THE AESTHETIC THEORIES OF FRENCH ARTISTS

condition of the arts in the commercial cities and civihza-


tions of the past. Such was the optimism of the Saint-
Simonists.
Comte found a methodological basis for this optimism.
According to him scientific explanation has developed
through three stages. In the first, the theological stage,
phenomena are explained as caused by some divinity or
spirit. This is the stage of primitive and ancient peoples.

In the second, the metaphysical stage, explanation is by


means of abstractions, such as substances, causes, forces.
This is the stage of the Middle Ages. In the third or
positive regime, explanation is description of appearances.
Comte had arranged the sciences according to the histori-
cal order in which they achieved the positive status. The
list he gave was mathematics, astronomy, physics, chem-

istry, biology, and sociology. To him this was an exhaus-


tive classification. Each science is more inclusive than the
ones preceding it and presupposes them. Sociology, the
crowning science, is the study of man in his social setting
and with all his activities. The aesthetic activities of man
gain purpose only when seen in the light of the perfect
positive regime of the future.
According to Comte art an ideal representation of
is

what is. The domain of art like that of science embraces


all reality. Science estimates it; art embellishes it and is a

social necessity. The imagination is subordinated to repro-


duction and description. Politics, art, and philosophy be-
gin in the study of inorganic nature then pass to the study
of man as intellectual and finally as moral. Since the
aesthetic faculty is intermediary between intellectual and
moral faculties, art emanates from philosophy and pre-
pares politics, and rises with least effort to the contempla-
tion of moral values.
The third step in the development of positivistic aes-
THE REALISM OF COURBET 15

thetics found in P. J. Proudhon's book Du principe de


is

posthumously in 1865. It is often credited


Vart, published
with being the complete expression of realist aesthetics.
It is the positivistic apologia for the subordination of art
to politicaland social ends. Proudhon was a good friend
of Courbet and it was he who persuaded Courbet to paint
with some social consciousness.
The aesthetic faculty and its place are Proudhon's first
theoretical considerations. He is against the separation of
art from morals and philosophy, and the theory of art for
art's sake, because he believes both attitudes are founded
on a false theory, namely, that the aesthetic faculty is
equal to the two great faculties of the soul, conscience (or
justice) and science (or truth) to which he considers all
,

others should necessarily be subordinated. Conscience


and science are the poles of the human soul. They are
the two equal and opposing powers from whose operation
arises the artistic ideal of life. The aesthetic faculty cannot
be equal to these two for it is only a sensible faculty. It
is by definition " the faculty in man by which we per-

ceive or discover beauty and ugliness, the agreeable and


the distasteful, the sublime and the trivial in ourselves and
in things and make from this perception new means of en-
joyment for ourselves." ^ Art, which continues the creative
work of nature, is an " idealist representation of nature
and ourselves in view of the physical and moral perfection
of our species." ^ So closely bound up are art, morals, and
society that the taste of a society is the measure of its
moral and intellectual culture. The artist, according to
Proudhon, should be of his time and place but as a critic
of them, for his aim is to be a prophet of humanity. Since,
as a positivist, he believes the perfect society is in the
future, the present society must be condemned. Therefore,
the purpose of art as the expression of a society, is to
16 THE AESTHETIC THEORIES OF FRENCH ARTISTS

comment on that society to the end of destroying it in


order to build a better one.^^
The consideration of positivistic aesthetics as the first

to preach consciously the gospel of the social purposes of


art is not accidental to a discussion of realism. Though

Courbet does not go to the length of talking about the


social obligation of art toward a new order in the future, he
does justify art as a kind of social documentation. And
his realism grew out of his political socialism. It was part
of his rebellion against the Second Empire and its arti-
ficialities.

According to Courbet beauty is a real and visible quality


in things waiting to be discovered by those whose percep-
tion is sensuously attuned to it. It is found in diverse
forms in nature, and artistic expression renders it through
copying the appearance of the model.
The idea of the diversity of beauty was part of the nine-
teenth-century romantic rebellion against idealist art.
Delacroix was declaring that " the sight of the beautiful
works of all times proves that beauty is not always met
under similar conditions " and that " one must see beauty
where the artist has wished to put it." ^^ Baudelaire was
exploring new dimensions of beauty in his poetry. The
grotesque and the ugly, the ordinary and the everyday
were receiving attention as species of the beautiful. The
subject matter of painting was being extended to all the
" ignoble " subjects that the ideahst had spurned. Courbet

was being the child of his time in this respect, too.


Though beauty has a variety of manifestations accord-
ing to Courbet, and though the artist renders it by repro-
duction of appearances, the beauty is not the appearance
itself. It is in the appearance of an object, a quality which

carries its own expression. As such it is a kind of abstract


general quality which to the positivist is a vestige of the
THE REALISM OF COURBET 17

metaphysical stage of culture. Courbet's theory of beauty


is not consistent with his positivism. A reinterpretation
of beauty in concretely verifiable terms is needed. Realist
artists following Courbet will think of a specific quality
of an object such as its lighted surface or talk of the
definite elements of painterly composition such as the
color harmony. The word beauty will not be so important
in their vocabularies.
Courbet's notion of beauty demands that the artist imi-
tate the appearance of his model, for he cannot improve
upon or embellish nature. As painters begin to move away
from such an idea of beauty and to look at art in terms of
the rendition of a character of the object they begin to
feel the inadequacy of the theory of art as exact imitation.
They begin to modify the theory of imitation or rebel
against it completely, moving over to some notion of
expressionism.
Something found in the philosophy of
of this shift can be
art of Taine. He is usually thought of as the aesthetician
par excellence of positivism and realism, but actually in
him we begin to find the transition of realism into its his-
torical consequents. Taine was interested mainly in estab-
lishing an aesthetic theory by which to understand works
of art. Though he defines art as imitation it is not " pure "
imitation. A
work of art exists for a definite purpose;
" The end of a work of art is to manifest some essential,
salient character, consequently some important idea, more
clearly and completely than is obtainable from real
objects." ^^
The one character of natural
artist chooses
objects and renders this predominant by systematically
modifying all others to conform to it. Taine describes the
impressionist procedure.
Eugene Veron carries the rebellion from imitation even
farther. He declares that beauty is an insufficient principle
.

18 THE AESTHETIC THEORIES OF FRENCH ARTISTS

to explain art for beauty leads to the idea of imitation.


But imitation in art not what we admire, but the genius
is

of the artist. The art of the future will be expressive art


wherein the artist will reflect the " measure of his own
sensibility, of his own imagination, and of his own intel-
lect." ^^
Veron points to the artist as expressionist.
Following the pattern of these two aestheticians subse-
quent artists will modify or deny the principle of imitation,
answering as best they may the problems that surround
the changes they make in the realist postulate.

NOTES
* Cf. R. Dumesnil, Le realisme, p. 12; and H. Focillon, La peinture
au XIX^ et XX^ siecles du realisme a nos jours, p. 7.
' Courbet, " Manifeste 1855," in C. Leger, Courbet,
p. 61.
^
Courbet, " Manifeste 1861," in C. Leger, op. cit., pp. 86-88.
* Cf.
G. Mead, Movements of thought in the Nineteenth Century,
p. 459.
Phrase taken from T. Silvestre, Artistes frangaises, vol. II, p. 145.
°

Emile Deschamps (1791-1871), French romantic poet, whose


®

Preface to Etudes frangaises et etrangeres, 1828, is one of the mani-


festoes of romanticism. Quotation from Preface, p. xvi (4th edition,
1829)
^
H. A. Needham, Le developpement de Vesthetique sodologique,
Cf.
pp. 81-83; L. Rosenthal, Du romantisme au realisme, pp. 370-374; G.
Larroumet, " L'art realiste et la critique," in Revue de Deux Mondes,
Dec. 15, 1892 and March 1, 1893.
® P.
J. Proudhon, Du principe de l'art, p. 17.
' Ibid., p. 43.

What better defense than Proudhon might Baron Haussmann
have invoked, for if he did destroy much property which obstructed
the paths of his projected boulevards, if he did seem to waste money
on extending these magnificent avenues out into the fields beyond the
city, was he not changing the present only to make a better future?
Cf. S. Giedion, Space, Time, and Architecture, pp. 465-480.
^^
E. Delacroix, " Questions sur le beau," Revue des Deux Mondes,
July 15, 1854, pp. 307 and 312.
^^
H. Taine, Philosophy of Art, tr. J. Durand, London, 1865, p. 64.
^^
E. Veron, Vesthetique, p. 150.
CHAPTER THREE

FROM REALISM TO NEO-IMPRESSIONISM


FOLLOWING COURBET the succession of painters
from Manet to the impressionists and neo-impression-
ists modified progressively the reahst doctrine, moving
painting away from the idea of imitation and introducing
increasing emphasis on the role of perception.
After the notoriety of the Salon des Refuses in 1863
Manet became had
the leader of the realists as Courbet
been the decade before. During these last years of the
Second Empire Manet was an influential factor in further-
ing among the younger men who met frequently in the
Cafe Guerbois the new ideas that were to issue in impres-
sionism. Whatever aesthetic he had is better found in his
paintings than in written words. However, there are two
brief purported quotations which may be given for what
they are worth:

Conciseness in art is a necessity and an elegance. The concise


person makes one think; the verbose person bores. Always modify in
... In a figure seek the high light and the
the direction of brevity.
deep shadow; the rest work out naturally; it is often a very small
will
matter. Furthermore, cultivate your memory, for nature will never
give you anything more than references. This is like a guard-rail that
keeps you from falling into banality. ... It is necessary always to
remain master and to do that which amuses.^
Color is a matter of taste and sensibility. For example, it is

necessary to have something to say; without that you are nowhere


in your art.^

As aesthetic these statements are only fragmentary.


But they do hint that Manet's realistic concern was
moving from the subject matter to the form. And they
say enough for one to point out that Manet supposes the

19
20 THE AESTHETIC THEORIES OF FRENCH ARTISTS

artistic process to be a choosing of certain elements from


the natural world because they are congenially adapted
to artistic rendition. It is not the meticulous over-detailing
of what appeared to some as the trivialities of Courbet.
This is the first move, consciously so taken by a realist

artist toward the doctrine that since the work of art cannot
imitate its model it had best be a heightened presentation
of chosen characteristics. The admonition to cultivate
memory, since nature gives us only raw data, is an admis-
sion as well to the further truth that the artist must trans-
fuse these data within his own mentality before they can
become artistic material. The positivistic level of experi-
ence alone is insufficient for the artist.
Impressionism has been called an " expression of a
moment of French sensibility, a rejuvenation of painting
not so much by procedures and ideas as by instinct." ^ It
was scarcely a clear-cut style, being more a coincidence of
taste of a group of painters in the seventies and early
eighties who accepted the term impressionism as expres-
sive of their community of interest. They were engaged in
an instinctive search for a style, and they found an aes-
thetic only toward the end of their period of collaborative
work. It is for this reason that we find no theoretical
defence of any principles of impressionism by the prac-
ticing artists. Whatever aesthetic they had remained to be
formulated by critics, Duranty, Duret, Geoffroy, Laforgue,
and others. It is necessary to consider impressionism here
as a link in the general course of the rise of new aesthetic
ideas.
The evident intent of the impressionist was to catch the
subject that he painted in one of the fleeting moments of
its existence. He opened his eyes and looked at the world
before him. He found that whatever objects he saw he
perceived in virtue of the light they receive and as appear-
FROM REALISM TO NEO -IMPRESSION ISM 21

ances conditioned solely by this light. Objects are colored


shapes, but one perceives the shapes only because they are
colored. Hence, if art has any relation to the things of
this world(and what realist of the nineteenth century
would doubt that it had.'^) then the play of sunlight on
,

objects through an envelope of atmosphere is the subject


matter for art.
Again, as with realism, positivistic description is the
ideal. The painter records the perceptual surface of the
world, its lighted, visual surface. To go beneath it would
be to inquire into metaphysical constituents or interior
physical composition.
That impressionism should have chosen light and color
as the proper object of the painter's vision seems appro-
priate if one thinks of the vast amount of work in the fields
of physical and psychological optics at this time. The
complete papers of Augustin Fresnel on diffraction and
polarized light phenomena, unknown or unpublished at the
time of his death in 1827, had been published in 1866. So
it is reasonable to suppose that by ten years later his work

had sufficiently filtered down into popular consciousness


to impress it with the notion that monochromatic light is a
succession of simple vibrations and that color is a matter
of frequency. Or one might recall to mind the importance
of Helmholtz and his researches in sensation which formed
an immediate groundwork for a physiological theory of
perception. Experimental methods when applied to psy-
chology and the science of optics change the character of
perceptual psychology from a philosophical science to an
experimental discipline. Questions of the relation between
the external world and our sensations are approached
through the physical study of the propagation and radia-
tion of light. And since these were the kind of questions
painting had become concerned with by the time of im-
22 THE AESTHETIC THEORIES OF FRENCH ARTISTS

pressionism, in keeping with the practices of its day, it

turned to physical and chemical theories of light and color


to solve them.
Also, beginning in the forties seascapes had become a
popular subject in French painting. The effect of water
reflections opened up to artists a new field, and one which
drew their attention to the intensity of light and play of
color.
The impressionists first thought that in discovering light
they had found all that was necessary to create an art of
painting. Record the visual impression exactly as you see
itwas their dictum. But they soon came to see that this
was not factual recording. The artist must intensify and
transfigure his object to agree with his impression. Light
ismorning light, or that of noon or dusk; it is the light
from a summer sun or of a winter day; it is the light of rain
or cloud or blue sky. If painters are to deal with light
alone they must give us those instants. The cathedral
facade, the haystack, the seascape could be painted again
and again from moment to moment. The subject matter,
the object recorded on the canvas, retreats until it becomes
no longer subject matter, but merely a perceptual motif
over which numerous variations are played. The object is
decomposed through the agency of sunlight. On the canvas
it is recomposed again through the general harmony of

colors. The painters make their analysis for the sake of a


new artistic synthesis. not the old imitation theory
This is

of art. That the impressionists came to realize as they


continued their course.
The method of the impressionists is analogous to that
of the physicists of the seventeenth century. The latter in
defining the world of physics chose those qualities from
their experience of the world which were sufficient and
necessary for that physics: shape, motion, solidity, exten-
.

FROM REALISM TO NEO-IMPRESSIONISM 23

sion, number. The impressionist in making his work of art


chose the quahty of hght which defined for him the
existence of the world. Painting is no longer pure imitation
of appearances, is not selected kinds of subject matter,
but is a way, like physics or mathematics, by which, from
selecting certain elements out of the world of experience,
we build a world that is our own. It is our own in that
we manipulate those elements schematically.
But the primary quality of the impressionists, which
they believed was an actual quality inherent solely in the
world about them, suffered the same fate as did the pri-
mary qualities of Locke when Berkeley pointed out that
they depended upon a perceiving mind as well as did the
secondary qualities. Impressionism came to see that the
quality of light and shade it depicted was far beyond the
simple recording of the light effects of nature. The sensi-
bilities of the painters had outstripped recording. Light
had been intensified past the place where it had been
drawn from nature. It became a formal element usurping
the place of structure and was labored beyond its poten-
tialities as the vehicle for form. The ideas of impressionism

led to their own disintegration and painters began to


search for a new aesthetic. One of the first to rise was
neo-impressionism
By the time of the eighth exhibition of the impres-
sionists in 1886 impressionism was giving way to the neo-
impressionism of Seurat and Sisley. Their work showed
that the influence of scientific studies on color and light
was more thoroughgoing, but it was also marked by the
intrusion of an unashamed rationalism.* Among the older
impressionists Pissarro embraced the new ideas for several
years. Each of the three men mentioned gave in brief or
extended written form his aesthetic.
Pissarro's theory is found in a letter to Durand-Ruel,
24 THE AESTHETIC THEORIES OF FRENCH ARTISTS

dated November 6, 1886, in answer to one from the latter


asking for complete notes on the painter's life and theory
of art, which were desired for an exhibition of French
painting to be held in New York. Pissarro's reply is

scarcely an adequate theory but it authenticates the ex-


position of the doctrine given by Felix Feneon and recog-
nizes Seurat as the moving force in the crystallization of
the theory. He says:

My dear M. Durand-Ruel:
I send you here enclosed the notice which you have asked me to
make about my new artistic doctrines.
Would you please complete it by consulting the pamphlet of M.
Felix Feneon which recently appeared under the title of Les im- '

pressionnistes en 1886,' for sale at Soiret's in Montmartre and at the


principal booksellers.
If your son brings out a publication on this subject I should want
him to understand thoroughly that it is M. Seurat, a most able artist,
who was the have the idea and to apply the scientific theory
first to
after having studied thoroughly. I have only followed, as have my
it

other confreres, Signac, Dubois-Pillet, the example given by Seurat.


I hope your son will be willing to render me this service for which I
shall be truly thankful.
Theory
Seek the modern synthesis through scientific means, which will be
based on the theory of colors discovered by M. Chevreul,' and accord-
ing to the experiments of Maxwell and the measurements of N. O.
Rood.«
Substitute the optic mixture for the pigmentary mixture, in other
words, the breaking up of a color tone into its component elements,
for the optic mixture creates much more intense luminosities than the
pigmentary mixture.
As for execution, we regard it as of no importance, at least as of
very little importance. Art has nothing to do with it, according to us.
The sole originality consists in the character of the drawing and in the
particular vision of each artist.^

The important thing for a neo-impressionist is that a


work of art is a synthesis of the color values composing the
FROM REALISM TO NEO-IMPRESSIONISM 25

canvas. Though Pissarro had turned back from the tech-


nique of divisionism characteristic of neo-impressionism
by 1890, another statement of his in 1904 shows how com-
pletely hewas still imbued with the search for harmony:
I see only spots. When I commence a picture the first thing that
I try to fix is the harmony (I'accord) . Among this sky, this land, and
this water there is a necessary harmonious relation, and that is the
great difiiculty with painting. What interests me less and less in my
art is the material side of painting, the lines. The big problem to
resolve is how to pull back everything, even the smallest details of a
picture, into the harmony of the ensemble, into full agreement.^

Pissarro never thoroughly assimilated the neo-impres-


sionist style.He continued to attempt to realize his syn-
thesis through color concord. Seurat's theory sets down
more fully how the desired harmony is to be brought
about, including more than Pissarro's color harmony.
The year that Seurat exhibited Chahut, 1890, a short
study on his life and work by Jules Christophe was issued
in Paris. The conclusion of the essay presents Seurat's
theory on painting as based on a letter written by the artist
to Christophe himself. Apparently the text received from
Seurat was shortened or edited and the painter was not
completely satisfied with the statement as given. In any
event we find a new exposition of his doctrines in a letter
to the writer, Maurice Beaubourg, dated August 28,
apparently 1890:

In conclusion, I shall record for you the aesthetic and technical


note which terminates M. Christophe 's work and which is from my
hand. I modify it a little, not having well condensed it for the printer.

Aesthetic:

Art is harmony.
Harmony is the analogy of contraries, the analogy of similarities in
tone (ton) , and line considered under the aspect of the
color (teinte) ,

dominant one and under the influence of lighting in gay, calm, or sad
combinations. The contraries are:
,

26 THE AESTHETIC THEORIES OF FRENCH ARTISTS

in tone, a lighter, more luminous one in place of a darker; in


color, complementaries, that is, a certain red opposed to its

complementary, etc. red-green, orange-blue, yellow-violet; in line,


those making a right angle.

Gaity, in tone is [gotten] through the use of dominant luminosity; in


color, of prevailing warmth; in line, those above the horizontal. ^^
Calmness, in tone is the equality of light and dark; in color, of warm
and cool; and the horizontal for line.

Sadness, in tone is a prevailing dark; in color, a prevailing cool one;


and in line, directions downward from the horizontal, a
Technique:
When we admit the phenomena of the duration of luminous impres-
sions on the retina, synthesis imposes itself as a result. The means for
expression is the optic mixture of tones and colors (according to the
placing and the way the colors are lighted, by sun, oil lamp, gas, etc.)

that is to say, the mixture of lights and their reactions (shadows)


following the laws of contrast, diminution, and irradiation. The frame
is in harmony opposed to that of the tones, colors, and lines of the
picture.^

Though Seurat calls a work of art a synthesis, it seems


apparent from what he has written that it is a sum. To
understand it one isolates the elements of which it is com-
posed, much as a chemist would set to work. For the work
of art these elements are concrete, visible existents, color,
lines, light and shade. One does not prate about beauty
or sublimity as the idealists do, for these are metaphysical
essences which the scientific aesthetician cannot find. Take
the elements which analysis yields, study them, learn how
to manipulate them in isolation and in combination and
you have the formula for the creation of a work of art.
If Courbet had adopted the point of view of positivism,
Seurat has adopted its method.
Paul Signac in his book, D' Eugene Delacroix au neo-
impressionnisme, defines the neo-impressionist painters as
" those who have revived and since 1886 developed the
.

FROM REALISM TO NEO -IMPRESSIONISM Ql

technique called divisionism in employing the optic mix-


ture of tones and colors as a manner of expression." ^°
These painters have been drawn to this manner of painting
because through it they can obtain greater luminosity than
by any other means. On this supposition the book is an
exposition of painting as a scientific technique. Signac is

concerned to show that the method of the neo-impres-


sionists is a realization of the theories of Delacroix and the
impressionists.
The neo-impressionists do not paint in dots, they " do
not *
pointille but divide," employing the technique of
'

little daubs of color because of the necessity of doing so


from scientific principles. Divisionism assures them lumi-
nous, colorful, harmonious effects through

1 The optic mixture of pigments ....


2 The separation of diverse elements, color, form, light, and their
reactions ....
3 The equilibrium of these elements and their proportion (accord-
ing to the laws of contrast, diminution, and irradiation)
4 The choice of touch proportional to the dimensions of the
picture.^^

Signac does not like the word pointillism. It does not


have the overtone of correct meaning. Divisionism, he
thinks, has, meaning " a complex system of harmony, an
aesthetic more than a technique," for " the basis of division
is contrast, and is not contrast art itself "
^^ .^^

With the thrill of the artist who finds pleasure in his


artistry he tells us in one beautiful passage that the painter
takes joy in his instrument of color as the musician in his
orchestration, that he composes with the seven notes of
the color scale as the musician does with the scale of
sound.^^
Long portions of the work are given over to showing how
the neo-impressionist aesthetic is in the tradition of Dela-
28 THE AESTHETIC THEORIES OF FRENCH ARTISTS

croix. Passages from Delacroix's Journal are quoted to


emphasize the comparisons. Impressionist theory is dis-
cussed for the same purpose. The aim of Delacroix, the im-
pressionists and the neo-impressionists is " to give color
the most brilliance possible." ^*
Among the three the
methods vary and consequently the results. The third
group alone fully realized the common aim. " By the
suppression of dull pigmentary mixture, by employing
exclusively the optic mixture of pure colors, by a methodic
divisionism and the observation of the scientific theory
of colors, [they] guaranteed a maximum luminosity, col-
oration, and harmony, which had never before been
^'
obtained."
Divisionism is further establishedby appeal to author-
ity. The scientific work of Rood is quoted and the theories
of Ruskin on stippling are called upon to support it.
There are two further ideas which should be noticed.
The first concerns beauty. Pissarro and Seurat do not
mention the word, Signac only twice. He calls the rules for
the harmony and accord of the elements in a painting the
rules and principles of beauty. Knowing these rules, he
argues, does not make one too technically learned to be
an knowledge of the principles of relationships
artist, for
does not make artists less capable of being moved emo-
tionally or less able to move us.^^ Beauty is not a meta-
physical quality which touches the soul, but a quality
inherent in physical relationships among various elements.
It has no mystery clinging to it and may be presented
through measured quantities. A new path is being followed
here. An embarrassing term of realist aesthetics is given
a scientific definition.
The second idea more important for it touches upon
is

the relation of the work of art to nature, of the picture to


the model. Signac refuses to make the artist the slave of
his model:
FROM REALISM TO NEO-IMPRESSIONISM 29

The divided touches of the neo-impressionists are the artifices these


painters use to express their particular vision of nature. Does a. . .

painter render a more beautiful homage to nature in striving, as the


neo-impressionists do, to re-establish upon the canvas her principal
essential, light, or in servilely copying the least blade of grass or the
^^
smallest pebble?

The artist does not depart from nature, but he strives to


give her fundamental aspect. For the neo-impressionist as
for the impressionist this is the sensory aspect of light.
The three statements of theory presented here are
clearly of unequal merit. Pissarro's letter of 1886 is too
sketchy and general. Signac deals almost exclusively with
divisionism, treating it as a scientific technique and being
concerned with establishing it as a traditional method.
Both men look to Seurat as their leader. His exposition
is the clearest.
It is also easy to note outstanding variations among the
three statements. Pissarro in 1904 admitted his interest in
the material side of his art, the lines, had decreased.
Signac makes mention of linear harmony, referring
little

only to the laws of line which are obviously copied from


Seurat. Seurat gives as much attention to line as to color.
Pissarro depreciates execution; Seurat presents the aes-
thetic and technique as interdependent.
These differences are minor. Neo-impressionism as a
single theory is clearly demonstrated by the fact that all
three of its apologists rest their theory on three funda-
mental suppositions:
1 The aesthetic value of a work of art lies in its internal
harmony of simple elements.
2 This internal harmony can be made explicit by an
analytic aesthetic.
3 The painter may achieve this harmony by the ap-
plication of a scientific technique.
30 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Good painting to the neo-impressionist is the bringing


together of disparate elements to fuse them into a har-
monious unit. Its process is twofold. The painter first

subdivides and classifies the elementary parts of his pic-


torial composition. The subject to be painted is analyzed
as composed of colors, lines, and tones.These elements
are then joined into a pattern in a painting, and the con-
textual unity of these parts gives the aesthetic quality of
the work, beauty to which we normally respond. This
its

is so because would seem that the quality of the elements


it

themselves as well as the manner in which they are used


have each corresponding emotional states constantly and
universally associated with them. There are definitely dis-
coverable laws of association between colors, lines, and
tones on the one hand and psychological reactions on the
other. For instance, gaiety, calmness, and sadness are
expressed by a dominating use of lines above the horizon-
tal, along it, or below it respectively. Now these seem-

ingly dogmatic statements were probably arrived at quite


empirically through experimentation in the studio with
certain color and line patterns for certain desired results.
Psychological studies substantiate the contention that cer-
tain colors and lines give rise to certain responses within
statistical limits.^^
Yet for all this, to tell what harmony is by giving the
conditions under which it is brought about does not make
the proposition that the aesthetic value of a work of art
lies in its harmony any less a gratuitous assumption. To
state* the objection in another manner, though we may
learn through our analysis that a good work of art may
have the harmony and contrast of line, tone, and color that
Seurat has formulated for us, we cannot be sure a work is
a good work of art because it has those qualities and rela-
tions. The proposition is not convertible. You may en-
FROM REALISM TO NEO-IMPRESSIONISM 31

lighten me about all the technical problems which have


been solved in a painting, you may explain to me the
relationship of lines, colors, volumes in it, moreover, we
may even agree that the particular painting has appeal;
but your demonstration cannot show me that the appeal is
the result of those solved problems and those relations.
It is possible to attribute the appeal to other reasons, and
to declare also that other paintings in which the harmony
is equally well got make no appeal. Many good painters

have done bad pictures. Aesthetic value is not the re-


sultant of good painting technically speaking.
The error here is one that was general in the nine-
teenth century. It is the error of associational psychology
and the representative theory of knowledge, which tried
to explain our mental processes and knowing activities by
analyzing them as the sum of certain units of sensation.
It was an error that William James pointed out in psy-
chology. No sum of elements ever yields us a qualitative
totality. The quantitative whole is the sum of its parts,
but the quality of the whole is not the sum of the qualities
of the parts. Water is chemically hydrogen and oxygen,
but the wetness of water is not the sum of the properties
of hydrogen and oxygen. Water is wet and puts out a fire;
hydrogen and oxygen are not wet, and the one is necessary,
the other an aid to combustion. The situation is similar
when we turn to meaning-situations. A sentence is the
sum of the words that make it up, physically speaking,
but the meaning of the sentence is not the sum of the mean-
ing of each word. The sentence has its own meaning. And
frequently, too, the meaning of a context will change or
alter the meanings of words within it. A work of art also
with its elements an analogous situation with refer-
is in
But in the nineteenth century
ence to emotional responses.
analysis was the recognized method of science, and the
32 THE AESTHETIC THEORIES OF FRENCH ARTISTS

method of science had borne fruit wherever it had been


appHed. It was reasonable to expect it would do so for
intensive qualities as well, and the new aesthetic analysis
was diligently pursued by Seurat and those who followed
the neo-impressionist beliefs.
With the supplanted by
neo-impressionists aesthetics is

the science of art. Aesthetics had approached art through


speculation on the nature of the beautiful. The science of
art examines the object of art and the elements contained
in it to arrive at generalized laws of the nature of art.
German theorists like Zimmermann and Fiedler had made
the science of art a study of schemes of pure visibility.
The neo-impressionists explained art as a visual scheme of
formal relations. The analysis in both cases is positivistic.

Taine had also worked along the same line when he made
his aesthetic study a philosophy of art. The difference was
only in his explanatory methodology. He had undertaken
his analysis on the basis of the elements that contribute to
the artist's make-up, the trinity of time, place, and race.
But he had failed to limit his study to elements visible
within the work itself and this was where the neo-impres-
sionist aesthetic was more positivistic than his. The new
aesthetic coming to the conviction that the sum of what
is

we can know about art is in the work of art itself. We


stand at the threshold of the autonomy of art.
The neo-impressionist is well aware of this. In his
theory the relation between art and nature begins to break.
If nature does not present itself to our sight in divided
touche*s,^^ where do we make the transition from nature to
SiTt? The principal quality in nature that painting can
render is light. The artist starts with his vision of nature

but to create a work of art he must proceed thenceforth


according to the rules of art itself. " Never in nature could
the artist discover its rigorously integrated design; nature
FROM REALISM TO NEO-IMPRESSIONISM 33

to him merely furnishes the data from which to select and


invent." Color arrangements and line compositions are

coming to be valued for themselves. The free beauty of


formal design is coming into its own. Signac correctly
senses the affinity of painting to music.
So, within forty years after the inception of realism
painting had advanced along the general path that Cour-
bet had marked out in following scientific procedures. But
in so doing it began to doubt the realist theses, for it
found the relation of nature to art a problem not a solu-
tion, and one that was to be crucial to the artists of the
next fifty years.

NOTES
^ E. Moreau-Nelaton, Manet raconte par lui-meme, vol. II, p. 95;
originally in the Grande Revue, August 10, 1907.
' Ibid., p. 96.
' H. Focillon, La peinturedu X/X* et du XX*' siecles, du realisme
a nos jours, p. 200. The same general point of view is expressed by
L. Venturi in Les archives de Vimpressionnisme.
* D' Eugene Delacroix au neo-impressionnisme, p. 59:
Cf. P. Signac,
" It is in 1886 ....
that .... there appeared works painted ....
by a reasoned method."
^Michel-Eugene Chevreul (1786-1889), French chemist, did ex-
tensive work on the theory of color as Director of the Gobelin
Tapestry Works. He was the author of an important work, De la hi
du contraste simultane des couleurs, 1839.
* Ogden Nicholas Rood (1831-1902) American physicist, made ex-
,

tensive studies on the quantitative analysis of color contrasts. His


principal work is Modern Chromatics, 1881.
'
Given in Venturi, op. cit., vol. II, p. 24.
® Robert de la Villeherve in the Havre-Eclair, Sept. 25, 1904; in
Venturi, op. cit., vol. I, p. 101.
* R. Rey, La renaissance du sentiment classique, reproduction facing
pp. 132-133; also in G. Coquiot, Des peintres maudits, pp. 132-134
with minor exceptions. Cf. Rey, op. cit., L. Cousturier, Seurat, and
J. Rewald, Seurat, for Seurat 's manifesto.
34 THE AESTHETIC THEORIES OF FRENCH ARTISTS

^°P. Signac, op. cit., p. 1.


" Ibid., pp. 4-5.
^'
Ibid., p. 79.
" Ibid., p. 82.
" Ibid., p. 87.
" Ibid., p. 88.
" /bzVi., pp. 106 and 107.
" Ibid., pp. 83-84.
^*
K. Hevner, " Experimental Studies of the Affective
Fic^e, e. g.,

Value of Colors and Lines," Journal of Applied Psychology, Aug.


1935, pp. 385-398.
^^
Yet in the theory of Helmholtz nature does so present itself to us.
He distinguishes three kinds of nerve fibres, each responding to one
of the pure colors, red, green, and violet. The sensation of a color,
therefore, depends upon the changes occasioned in these fibres by the
external stimulus.
of " La Grande
^*
D. C. Rich, Seurat and the Evolution Jatte,"
p. 15.
CHAPTER FOUR

THE GREAT INDEPENDENTS: RENOIR AND


CEZANNE
PAINTING was not to follow the ways of scientific
realism without a vigorous denial from some artists.

This denial could take two forms. In the first place, one
could repudiate any connection between science and art
and disregard the problem of the relation of the work of
art to the natural object. In the second place, one could
go on to a deeper probing that would modify the posi-
tivistic viewpoint and result in a new conception of the
relation between the world of art and the natural world.
And would mark a new point of departure in modern
it

aesthetics. The first was the way of Renoir. The second


was that of Cezanne. Both men were great enough
painters to pursue their own individual ways. And in their
theories they never descended to the dogmatic limits which
define a school.Renoir steadily fought against schools.
Cezanne was too possessed with the analysis of his own
aesthetic to fall into such a groove.
The first of Renoir's writings that we have are two
articles which appeared in Riviere's Uimpressionniste in
1877, signed un Peintre. Why he asks in the first of
is it,

these, that the architecture of today does not have the


beauty of the older buildings.'^ The old Louvre is beautiful
the new one is not. He ventures the answer that it is
because the ornamentation on the new is heavy, banal, and
the work of laborers, that on the old is light and the work
of artists.^
His answer betrays a belief that beauty is a kind of
decoration superadded to life. This is in keeping with the

35
36 THE AESTHETIC THEORIES OF FRENCH ARTISTS

impressionist concern with surface appearances only. The


beauty of architecture is not its functional value, but is the
decorative arabesque that plays over fagade and walls.
" The architect should be an artist more than a savant " ^

is the burden of the second article. U


6 cole des Beaux Arts
education has made copyists of architects, sculptors, and
painters. Such a school should be suppressed. Until a
new generation of architects comes to reverse affairs with a
new education, the direction of architectural monuments
should be given to painters, he claims.
Though French painters had fulminated against the
academic teaching of the popularly accepted schools of the
nineteenth century, beaux arts architecture was not sub-
jected to widespread criticism until later. The eclecticism
of the schools was generally accepted, or at least domi-
nated architectural practice, until the end of the century.
True there were men like Labrouste, who thought of archi-
tecture as building in forms that met the new social
demands and employed new materials, but their work was
the rare exception. Renoir is making a plea for the archi-
tect as a creator, not as a copyist of styles. It is significant
that he is among the earliest of those who fought the false
training of the schools. The very fact that these articles
were the work of a painter and not an architect demon-
strates the healthy influence of the new revolt in painting
on the other arts. Furthermore, they show that the revolt
against beaux arts architecture had been in gestation since
the early days of the Third Republic.
In 1884 Renoir drew up the program of irregularity,
opposing the traditions of academic painting and the
stereotype art which the public was accepting. Its greatest
virtue is its simple statement of what seems obvious:

Nature abhors a vacuum, say the physicists. They should com-


plete their axiom by adding that it has no less a horror of regularity.
THE GREAT INDEPENDENTS 37

Observers know in effect that, in spite of the apparent simplicity


of the laws which preside at their formation, the works of nature are
infinitely varied, from the most important to the least, of whatever
type or kind they are. The two eyes of the most beautiful face will
always be the least bit dissimilar; even the nose is not found exactly
placed over the center of the mouth. The sections of an orange, the
leaves of a tree, the petals of a flower are never identical. It seems
that the beauties of every order draw their charms from diversity.
Upon examining from this point of view the most renowned plastic
or architectural reproductions one can easily see that the great artists
who have created them, careful to proceed like nature of whom they
are always respectful pupils, are on their guard never to transgress
its fundamental law of irregularity. It is even proved that works
.... don't present one line of perfect
based on geometric principles
exactness, and what round, square, or oval figures there are which
could easily have been made exact, are never so. Thus one can
without fearing error affirm that every really artistic production has
been conceived and executed according to the principle of irregularity.
In a word serving us as a neologism to express more completely our
thought [we may say] that it is always the work of an irregularist.
At this time when our French art, [which was] so full of penetrating
charm and exquisite fantasy at the beginning of this century, is going
to decay because of regularity, dryness, and the mania of false perfec-
tion that is tending to make the unadorned cleanness of the engineer
into an ideal, we think it useful to react promptly against the moral
doctrines which threaten to annihilate it. We also think it is the duty

of all sensitive people and men of taste to join together without


delay regardless of what may be their repugnance to struggle and
protestations.^

An association of artists dedicated to the principle of


irregularity is necessary, he continues, and the condition of
admission to it for architects would be that they render

their ornamental motifs according to nature and without


precision instruments.
Such a program divorces art from the mechanical crafts-
manship that had become the popular ideal in the nine-
teenth century and from repetitious formulas. It makes
freehand spontaneity the basis for art. That art should
38 THE AESTHETIC THEORIES OF FRENCH ARTISTS

strive to be irregular in this sense may seem an innocent


statement. But it must be recalled that from the time of
Plato there had been the tradition that true beauty is the
beauty of geometrical figures.* In the eighteenth century
the love for natural landscape, the cult of the picturesque,
and the Gothic revival indicate a change to an interest in
variation. This change is found in the love for the exotic,
the great admiration for Chinese art, architecture, and
gardens. This is the taste that Renoir was bred in when
he decorated fans in his youth.
The functionalism whose bridges, fac-
of the engineer,
tories, expositions buildings, and department stores were
becoming numerous at the time Renoir wrote, was " regu-
larist " and contrary to his taste. The unadorned, clean
beauty of such buildings repelled him. He did not see that
in the crude, machinelike buildings of the engineer was the
new force that would sweep away the old beaux arts tradi-
tions and clear the ground for a new architecture.
Renoir was concerned only with a very small problem
raised by the conditions about him. The end of the nine-
teenth century saw a general bankruptcy of taste and the
substituting of moralizing and electicism to justify the
opulence of life and the acceptance of middle class values.
To the average taste of the time any work of art was good
by virtue of its applied ornament and its decorative appeal.
The quality of the decoration was not questioned. We
must at least admit that Renoir raised his voice in favor of
good craftsmanship and good taste in decorative design.
Renoir's attitude was that of those who helped to bring
about the new style in decoration at the turn of the cen-
tury known as Vart nouveau. Inspired by the crafts move-
ment of William Morris in England who revolted against
the poor taste of his day, the arts of design experienced a
minor renaissance. There was a revived interest in good
THE GREAT INDEPENDENTS 39

craftsmanship that is bearing admirable fruit today.


Going to nature for their models the artists of Vart
nouveau employed graceful curves, swirling spirals, plant
and animal forms with supposed imaginative freedom in
furniture,metal work, jewelry, architectural accessories,
and painted decoration. L'art nouveau was the embodi-
ment of Renoir's principle of irregularity. But the move-
ment was doomed to an early death for the artistic value
of a work can scarcely lie in so weak a principle.
Renoir also wrote a preface to an edition of Cennino
Cennini's Livre de Vart.^ This preface has been called
" Renoir's testament " which affirms " on each page his

faith in the classic tradition," ^ a statement that is cer-


tainly exaggeration.
We are liable to forget the past, Renoir tells us, because
so many recent marvelous events and discoveries turn our
eyes from what happened prior to them. But a glance
back at this past is not a useless procedure for us today,

and in a man like Cennini we are amply rewarded. Besides


being a technical manual, the treatise of Cennini is a
history from which we can learn much about art. The
artistic greatness of a nation not measured by the few
is

works of a man of genius but by all the works of all its


artists. It was to all artists of his time that Cennini spoke,
his purpose to make good artisans."^
Renoir reiterates the proposition of the realists: " One
would not know how to live outside of his time. ." ® . .

The times of Cennini and our own, Renoir says, differ in


two important respects which make our own artistically
inferior. Painters could then engage in collectivistic activi-
ties, that is, several might decorate one church and yet

among their individual works there would be a unity for


" the painters would possess the same metier." ^ Painters

then learned their trade through apprenticeship, through


40 THE AESTHETIC THEORIES OF FRENCH ARTISTS

initiation little by little into the actual practice of their


art. Yet such apprenticeship did not kill the originality
of those painters: Raphael was no less himself for being
the pupil of Perugino. This gone
spirit of the atelier is
from art today. The second factor which explains the
great value of past art is religious sentiment, that har-
monizing agent which moved among all men and the age
in which they lived. Though religious sentiment has
steadily grown weaker the rules established under its in-
fluence have kept art on a high plane where Catholic
culture has prevailed. Even in the time of the greatest
power of the Church the artists were relatively free.
" Faith was the regulator of their fantasy which could
nourish itself without fear from profane sources." ^^ Today
we are not in that happy situation:

Gods and gods are necessary to our imagina-


are no longer wanted,
tion. It must be confessed, modern rationalism, if it is able to satisfy
the learned, is a manner of thinking incompatible with any conception
of art.ii

Beside these two factors there is a secondary cause for the


decadence of art today. Formerly one worker would carry
one object to completion himself, and thus the work would
be a product of his own interest and intellectual joy in the
solution of the problems it presented. But today the
division of labor, servile repetition, machinisme has sup-
pressed this joy. The has forgotten his ideal: " The
artist
most skilful hand is always only the servant of thought." ^^
One may learn to make clever works in the modern art
school, but he will never make a work of art until he can
work with this great ideal of craftsmanship
enliven his
which makes the painter subject to the same rules of
metier as the carpenter or the iron worker.
Here, as in 1877 and 1884, the quarrel of Renoir is with
THE GREAT INDEPENDENTS 41

inadequate craftsmanship. To work by rule is machinisme


but art not encompassed by precept and logic. What is
is

made by rule does not have spontaneity; it is a copy. And


that which can be wholly known by reason is not art. The
lesson of the history of art is that works of art cannot be
produced by the methods of the schools for that which
makes the old art great is precisely what we can never
know by rules. It is the human quality found in the
religious sentiment of the painter of the Renaissance, or
in the joy he took in his work. Great art is not imitation,
and being itself indescribable, is inimitable. ^^ The artistic
quality in a work of art is something inherent in each work.
Irregularity is the principle of vitality as opposed to mech-
anism. No theory is capable of describing what a work
of art is. No rules can tell us how to make such a work.
One may say that the painter must strive to perfect his
craft and that he must put himself in the tradition of great
art, but specifically how he is to do this cannot be told. It
is a matter of the feelings or intuition of the artist.
This is the admission that art is inexplicable. " If they
could explain a picture ^*
it Renoir is
wouldn't be art,"
reported to have said. Courbet had made artistry some-
thing definite, copying the beauty of nature, the plainness
of homely objects simply intuited and recorded. The neo-
impressionists had reduced art to a formula of harmonic
relationships. There was no mystery which need be con-
cealed under the word " intuition " for all was open to
analysis. With Renoir painting becomes instinctive again.
Good artistry is akin to the craftsmanship of the Renais-
sance painters. One
paints with a feeling of inspiration like
a religious sentiment, and one takes joy and deep interest
in his performance. And the works which one produces
reflect this gaiety and exuberance. Teaching can-
lifelike
not give this quality to a work of art and theory will never
42 THE AESTHETIC THEORIES OF FRENCH ARTISTS

describe it. If ever any artist believed the painter should


have no program, no conscious theory, it was Renoir. Art
is instinct, and categorical theorizing cannot describe

instinct.
The philosophical parallel of Renoir's theory
is the Berg-

sonian metaphysics. In the division of reality between


time and space, that which is living is found in the first
category, that which is inorganic in the second. Life has
a dynamic urge, an elan vital, which propels it through
time and which brings to fruition all the multiple varia-
tions of living things, the irregularities of nature which
Renoir talks about. Reason is the power by which we
know the world of space; intuition, our inner feeling, is the
principle by which we perceive the world of time with its
living movement. Reason gives us relative knowledge,
intuition gives absolute knowledge. The first is symboli-
cal, linguistic description, presenting only a limited phase

of an object. It is never complete. Intuition, on the other


hand, is immediate and direct. We sympathize with the
object and know it " from within." Symbolical knowledge
is analytic, it breaks up the object; intuition gives knowl-

edge of the object as a unit. Symbolical knowledge is


abstract, scientific. In intuition we identify ourselves with
the object of knowledge by imaginative effort. It is pos-
sible to pass from intuition to analysis but not the other
way. Intuition is the intellectual sympathy by which one
knows the inexpressible, the meaning of a poem, the beauty
of art. Art belongs to the world of time, and reason is
dumb before it. Theory can not explain art, for theory
is the work of reason and a work of art is the spume of the

intuition. Art is indescribable, so reason cannot reach it;


it is inimitable, so reproduction will not give the copy that

quality which belongs to aesthetic sensibility.


In art man makes use of intelligence because he makes
THE GREAT INDEPENDENTS 43

choicesand works towards purposes. In artistic technique


the hand becomes the servant of thought, he settles those
problems of his craft which arise from the difficulties of his
medium by the application of rules. Yet in all this the
intelligence kept subordinated to intuitive desires. " I
is

arrange my subject as I want it, then I go ahead and paint


it like a child. I want a red to be sonorous — to sound like
a bell; if it doesn't turn out that way I put on more reds
or other colors till I get it." ^^
When art belongs to Berg-
sonian intuition how else could one explain his painting.^^

The painter must seize the rich multiplicity of natural


forms and translate the life of the organic world to his
canvas. The flesh on his canvas must be alive.
In his concern with spontaneity Renoir never once con-
siders the realist's problem of the relation of the picture to
the model. It is a false problem to one who denies that the
artist should follow the suppositions and techniques of the
positivist aesthetic. The world of the intuition has nothing
to do with art as an imitation of outward natural forms.
Cezanne, however, faced the problem and gave it thought,

hence became the fountainhead of early twentieth century


developments. He, too, did not have a clear-cut theory,
but that is because his ideas suggest how much a theory
of painting depends upon psychological and epistemologi-
cal foundations.
For many years Cezanne worked in retirement trying
to realize his conceptions, hiswork unknown at public
exhibitions and seen only by the artists who frequented
the shop of Tanguy, dealer in oils. In a letter to Octave
Maus, leader of the Belgian Vingt, Cezanne voiced the
reason for his retirement:

I have resolved to work in silence until the day when I shall feel
myself capable of defending theoretically the result of my attempts.^®
44 THE AESTHETIC THEORIES OF FRENCH ARTISTS

By a theoretical defense he never meant dogmatic rules


of art. He wanted an explanation of what he was doing
that would satisfy himself and explain his art to others.
He wanted a statement of his personal way of painting
when that manner had crystaUized to his satisfaction, not
a formalistic theory. That such is the case may explain
why his theory was never fully set forth in a single docu-
ment, but is found in scattered passages in his correspon-
dence during the last four years of his life.
In 1878 he had written to his friend, Zola, that painting
was a means of expressing sensation. ^^ This was a hfelong
belief to which many a letter bears witness. ^^ In sensation
we make our contact with nature, and in nature art finds
its subject matter. Cezanne started with the position of

the impressionists to whose group he once belonged.


Yet one must also recognize that there is the problem of
the technical adjustment of the natural model to the
demands of the artistic expression:

In art above all everything is theory developed and applied on


contact with nature. ^^
Now if a strong sensation of nature —and certainly I have a lively
one — is the necessary basis of all artistic conception, and the basis
on which the grandeur and beauty of the future work reposes, knowl-
edge of the means of expressing our emotion is no less essential and
is acquired only through a very long experience. ^^

But sensation is not a simple thing, as the impressionists


had thought. Nature is a diversified and complex model
and one cannot be merely submissive before it.

I me as very com-
proceed very slowly, nature showing herself to
plex; and the progress to be made [in understanding her] is endless.
One must see one's model very well and feel it very forcefully, and
still more, express it with distinction and force. The real and . . .

stupendous study to be ventured is that of the diversity of nature's


picture.21
.

THE GREAT INDEPENDENTS 45

Cezanne is never the slave of his model, but goes to


nature to depict her according to his own artistic vision:

But I always come back to this: the painter ought to dedicate


himself entirely to studying nature, and should endeavor to produce
pictures which are a lesson .... One can't be too scrupulous, too
sincere, or too submissive to nature, but one ought to be more or
less master of his model and above all of his means of expression.
Penetrate what is before you, and express yourself as logically as
possible.^^

Cezanne denies that the visual sensations are primarily of


the luminous surface of bodies. It is not light and color
we perceive; rather, these are the forms under which we
perceive receding planes. The painter is concerned with
depth, the plasticity of nature:

I wish to say that in an orange, an apple, a bowl, a head there is


a culminating point, and this point is always —
in spite of the terrible
effect: light and shade, colored sensations —drawn closest to the eye
[when] the borders of objects flee towards a center placed at our
horizon. 23
An optic sensation is produced in our visual organ, which makes us
class by light, half-tone or quarter-tone, the planes represented by
colored sensations, (light does not exist, therefore, for the painter)
Insofar as you go forcibly from black to white, the first of these
abstractions being like a fulcrum as much for the eye as for the
mind, we daub and do not succeed in having our freedom, in being
masters of ourselves. 2*
Now the colored sensations which give light are the cause of
abstractions which prevent me, an old man about seventy, from
covering my
canvas and from pursuing the delimitation of objects
when the points of contact are tenuous and delicate, from which it
is implied that my image or picture is incomplete. In another way
the planes one upon the other, from which neo-impression-
fall

ism arises, which circumscribes the contours with a black stroke,


a fault which it is necessary to combat with all force. Now nature,
when we consult it, gives us the ways for attaining this purpose.^^

Truly Cezanne has made impressionism something solid.


46 THE AESTHETIC THEORIES OF FRENCH ARTISTS

The principle under which the artist orders his perceptions


is recession from the eye. From this he perceives volumes
in depth, overlapping planes, and the blue effect of aerial
perspective. These are the elements he can adjust on his
pictorial surface. In a famous passage he announces this:

Permit me what I said to you: treat nature by the


to repeat
cylinder, the sphere, and the cone; the whole placed in perspective,
let each side of an object or a plane be directed toward a central

point. Lines parallel to the horizontal give extent, whether it be a


section of nature, or if you prefer, some spectacle which the Omni-
potent Father, Holy God, spreads before your eyes. Lines perpen-
dicular to this horizon give depth. Indeed, nature for men is more a
thing of depth than of surface, whence comes the necessity of intro-
ducing in our vibrations of light, represented by the reds and the
yellows, a sufficient sum of blues for the air to be felt.^^

These passages show quite clearly how Cezanne was


reacting to the neo-impressionist aesthetic and turning to
the western tradition of perspective, both linear and aerial,
to correct its errors. They also show him grappling with
our recurrent problem of the relation between the natural
and the pictorial world. To him
nature and art are con-
nected but not in the model-copy relationship.
strict
Nature supplies the grounds for art, whose business it is to
comment on the natural world within the limits of its
own possibilities, not to copy outward appearances. The
basis of such a procedure is sensation. The positivistic
point of view of realism and impressionism is still the
determining influence here, but new factors have entered.
The* realist preoccupation with exact imitation of nature
raised the question of what is nature. One may say, " open
your eyes and look,'* but the course of painting from 1855
to 1889 had shown that nature may be several things. It
may be rural and social simplicity and that part of the
world which the average person calls unpleasant. It may
THE GREAT INDEPENDENTS 47

be the physical surface of objects in their iridescence. It


may be any organized, arranged pattern, or it may be
irregularity. The data of sensation are in the natural
world, but for the artist these data are selected, determined
by the principle which governs his conception of what
painting can or should do. The realists, the impressionists,
Renoir, had conceived of the natural world which art
all

presents as some form of surface quality. They had not


completely broken with the superficiality of the July Mon-
archy and the Second Empire, with satisfaction in orna-
ment, with the positivistic point of view. Cezanne was
trying to make the nature which art describes something
more essential, something closer to physical reality. He
was trying to establish for that nature a principle which
did not derive from the perception of accidental qualities.
He was trying to find the primary qualities for the nature
which art knows as the Renaissance philosophers had tried
to find the primary qualities of the nature which physics
knows when they gave up sensible qualities in favor of
mathematical qualities. And these primary qualities which
nature displays for painting, Cezanne declared were
volumes and the recessive juxtaposition of planes as given
in linear and aerial perspective. All the means open to
the painter are subordinated to the purpose of giving more
adequate expression to the density of nature. Color is a
means of representing volumes; light and shade also. The
nature of the painter is as much an abstraction as the
nature of the physicist. Now when sensation yields the
world of volumes expressed in painting, the positivist posi-
tion would appear to be injected with an element of the
intellectual. Plastic organization cannot come from simple
passive sense perception, but only from a reorganization
of nature, treating it according to certain hypothetical
constants (receding planes, volumes, atmospheric perspec-
48 THE AESTHETIC THEORIES OF FRENCH ARTISTS

live) . The precept about the cyHnder, the sphere, and


the cone is not a rule for the representation of objects, as
it was later tortured by the cubists to signify. It specified
the manner, the presuppositions under which the artist
was to approach nature.
Cezanne had raised and settled to his own satisfaction,
at least insofar as it affected art, the ever recurring philo-
sophical question of how much of what is given in sense
perception is to be attributed to the external world and
how much to our own intellectualizations. It has been
said that the theory of Cezanne lets " vision appear in its
sensible purity." But what is the sensible purity of
^^

visual sensation.? Things do not look to us as they do


solely because of what they are. We do not passively re-
ceive simple sensations which, by separate mental actions
of interpretation, are changed into perceptions. The ap-
pearance of objects depends upon the total visual field in
which they are found, and upon definite attitudes and
experiences, in short, upon the total organization of the
observer. Now our visual environment may consist, let us
say, of chairs, tables, and apples, or it may consist of
cubes, rectangles, and spheres. To whichever class the ob-
jects of our environment belong depends upon their rela-
tionship to our activities. The objects are " functional
characters " of our designs upon them. They are chairs if
we wish to sit in them, apples if we wish to eat them. The

objects that we look upon as objects to eat are also decora-


tions if we place them along with other fruit in a dish on a
sideboard; they are something which secure our personal
tenuous economic well-being if we have grown them to sell,
and we immediately translate them into monetary terms;
they are colored spheres if we wish to paint them. Which-

ever way we " see "them depends upon our assumed


mental attitude toward them resulting from some behavior
THE GREAT INDEPENDENTS 49

we wish to take regarding them.'^^ The hungry man, the


farmer, the artist, etc. each has his own " categories "
under which he subsumes these objects, and these cate-
gories are each dependent upon the other. This is the
assumption Cezanne is making when he talks of treating
nature as volumes and planes. Hence when he insists that
the artist should approach nature through sensation, this
sensation is predetermined as one which will yield him the
object under the assumed artistic category. It cannot be
claimed that sensation is pure or can be considered apart
from abstraction. Sensation by its very nature and pur-
poses includes the second, and is not something prior upon
which abstraction works. The distinction between the two
is mental. Even if sensation were a pure and simple thing

it could not account for the difference between the aes-

thetic object and the natural object, or the way in which


the natural model is adjusted to the demands of two-
dimensional representation.
The aesthetic object is a hypothetic object. It arises as
relative to a naturalmodel and to a painter's intent as that
is by his medium. In the adjustment of the
qualified
natural model to the pictorial demands there seems to be
an assertion of the autonomy of art. In the continual in-
sistence on a reference back to nature Cezanne qualifies
that assertion. In his aesthetic the perceptions and ab-
stractions of the artist are placed on an equal level with
mathematical thought and logico-scientific procedures,
methods to which alone the world had given priority from
the seventeenth to the nineteenth centuries.
The theory of Cezanne like his painting was destined to
be developed in two directions, one way by the cubists and
intellectualists, the other by the fauves and expressionists.
The analogy of Cezanne's idea that artistic creation
depends upon the acceptance of an ordering principle of
50 THE AESTHETIC THEORIES OF FRENCH ARTISTS

sensation and Poincare's demonstration of the reason for


hypotheses in scientific method is a striking example of
similar trends of thought occurring in disparate fields.
Both began on a positivistic basis, Cezanne assuming that
the artist limits his art to the expression of his experience
of nature; Poincare, the physicist, recognizing as legitimate
explanation only the descriptions of empirical sciences.
Yet each affirmed that within the rendition of his experi-
ence there was a subjective element. To Cezanne nature
was so diversified that the artist must have some principle
by which he orders his sensations of it. This principle was
depth, and it determined what the artist was to see in
nature. And to Poincare the scientist adopts his hypo-
theses in view of the principle of greater simplicity and
consistent application. Such a principle is not only practi-
cal but is also aesthetic; the hypothesis adopted meets the
scientist's taste for neatness. His principle enables him to
schematize the materials of knowledge as the artist's
enables him to organize his complex sensations.
The formulation of Einstein's relativity hypothesis, by
which simultaneity depends upon an assumed point of
reference; the rise of polytonality in music, in which two
keys are employed simultaneously but the harmony is
dependent upon a principle of overtone relationship among
the tones occurring at any moment; the development of
symbolic logic, which formalizes the old logic of class-
inclusion and the logic of propositions, are further evi-
dences of a general trend to new schematizations in the
sciences and the arts at this time.
With Cezanne the practice of art is raised from a con-
cern with realistic rendition to an intellectually and sensi-
bly modified transcription of the natural world. And the
aesthetics of painting assumes the attitude of a critical
positivism which is careful of its own methodology and is
THE GREAT INDEPENDENTS 51

aware of the problems of psychology and epistemology


within itself.

I NOTES
^ Articles quoted in Venturi, Les archives de Vimpressionnisme, vol.
II, pp. 321-322, 326-329.
' Ibid,, p. 328.
^
Ibid., yo\. I, pp. 127-129.
*
Cf. Plato, Philebus, 51c.
^
Cennino Cennini, Florentine artist, wrote a technical treatise on
painting about 1400. A modern French translation appeared under
the title, Livre de I'art, prefaced by a letter from Renoir to the
translator's son, Henry Mottez, who published this edition. The date
of Renoir's letter is doubtful, though probably in his last year. The
letter is given in R. Rey, La renaissance du sentiment classique, pp.

I v-xii. Letter is hereinafter referred to as Lettre a H. M. Translations

are quoted from this letter by the kind permission of Art Catholique,
copyright holders of Victor Mottez' edition of C. Cennini, Livre de
Vart.
" Rey, op. cit., p. 58.
' Cf Lettre a H.
. M„ p. vi.
* Ibid., p. vii.
* Ibid., p. viii.
^'Ibid.,p.x.
^^
Loc. cit.
^*
Ibid., p. xii.
^'^
Cf. W. Pach, " Pierre Auguste Renoir," Scribner's Magazine,
I
May 1929.
" Ibid.
''Ibid.
^*P. Cezanne, Correspondance, ed. J. Rewald, p. 214, letter dated
11-27-89.
''Ibid., p. 153, to Zola 11-20-78.
" Cf. ibid., p. 227, fragment, 1896; p. 288 to his son 9-8-06; p. 297,
to his son 10-13-06 and 10-15-06, and in following passages quoted.
" Ibid., p. 253, to Chas. Camoin 2-22,03.
''Ubid., p. 257, to Louis Aurenche 1-23-04.
''Ibid., p. 261, to Bernard 1905, also p. 238, to his son 9-8-06.
" Ibid., p. 262, to Bernard 5-26-04.
52 THE AESTHETIC THEORIES OF FRENCH ARTISTS
" Ibid., p. 265, toBernard 7-25-04.
**
Bernard 12-23-04.
Ibid., p. 269, to
" Ibid., pp. 276-277, to Bernard 10-23-04.
'"
Ibid., p. 259, to Bernard 4-15-04.
" L. Venturi, Cezanne, vol. I, p. 53.
*"
For the complete presentation of the psychological argument here
briefed cf. K. Koffka, Principles of Gestalt Psychology, especially pp.

75-105, 129-164, 392-394.


CHAPTER FIVE

SYMBOLISM AND FAUVISM: GAUGUIN, DENIS,


MATISSE
REPUDIATING the validity of a connection be-
INtween science and art Renoir had Hmited his critique
to machinism, and had declared that art is the result of
sensibility. An attitude that completely divorced art and
science was voiced by those men who talked of primi-
tivism, the intuition, and expression.
In rebellion against the analytic aesthetic of neo-impres-
sionism, Paul Gauguin and a group of friends initiated
about 1888 the style of painting later to be called sym-
bolism. The superficial characteristic of the style was a
complete change from broken brushwork to smooth and
unvarying application of color to a definite area of canvas,
much like a block print in effect.^ Behind this technique
there was a developing aesthetic. Gauguin and his friends
exhibited as a group at the Paris Exposition of 1889 in a
cafeon the grounds known as the Cafe Volpini. The pic-
tures were in white frames, and the announcement pro-
claimed they were painted by a group calling themselves
the symbolists or synthetists. Later the painters, Emile
Bernard and Paul Serusier, separated from the group and
from that time Gauguin spoke of these only as the sym-
bolists.^ Similar groups later came into being with different
variations on the original aesthetic, the nabis with Maurice
Denis, and the Rose Croix painters. However, their theo-
ries of art are similar enough to warrant using the term
symbolism as generic for all of them.
In the many journals and letters written by Gauguin
there are numerous allusions to art.^ Here one finds in-
cipient the principles of symbolism in Gauguin's identifica-

53
54 THE AESTHETIC THEORIES OF FRENCH ARTISTS

tion of his exotic art with his escape to the exotic Hfe. The
first is the symbol of the second. To Strindberg he writes:

is what you suffer from; to me barbarism is a rejuvena-


Civilization
tion. Perhaps the memories of your selection have evoked [for you]
a painful past when confronted with the Eve of my choice, which I
have painted in the forms and colors of another world. This world,
which neither a Cuvier nor a botanist would possibly recognize, will
be a paradise of which I alone have made the rough draft. And from
the rough draft to the realization of this dream is far. But what
matter! To catch a glimpse of blessedness, is not that a foretaste of
nirvana? *

Such was his reply to the great dramatist who with realist
concern had doubted the barbarity, the torturedness, and
the anti-literal painting of such works as The Spirit of the
Dead Watching.
Gauguin fled to the islands of the Pacific because he
believed that in the simple life of the primitive of his day
one finds again man's true nature which has been lost
through civilization. This yearning to return to a " natural
state " where man once had his bliss is the stuff of all
primitivisms. It has existed in a thousand forms since the
beginning of history, for nature like beauty is not a single
thing. Each race has its myth, each society its interpreta-
tion. There is the legend of the natural state of man in
the Garden before the fall. There is the legend of a past
golden age before troubles and evils came into the world.
Nature has meant the state of the " noble savage," or the
Petit Trianon and a milk-maid's costume. A romantic love
for the exotic confuses that which is primitive in place with
that which is primitive in time. It is assumed that those

peoples who live untrammeled by the economic and social


artificialities of occidental civilization have retained that
desired quality of the childhood of the race. This quality
they display in their works of art, and whoever would
SYMBOLISM AND FAUVISM 55

strip from his own art the inhibiting accretions of civiliza-


tion must make work Hke theirs, not only in appear-
his
ance but in the strong feeling for the simple which is
incorporated in them. One who is a primitive in life is

thus the better able to be a primitivist in art. The works


of art he gives us will correspond to the emotions and
joys he experiences in his adopted life.

For Gauguin the work of art was also the symbol of his
own emotions: " The work of art for him who can see is

a mirror wherein is reflected the state of soul of the


artist." ^ He carried into his art all the pathetic emotion-
alism that led him to attribute to the hfe of the Poly-
nesians a happiness not to be found in Europe. In this he
was distinctive. True, artists before him had found in-
spiration in far places and distant times. The eighteenth-
century interest in Chinese art had been reflected in Eng-
lish gardens and Chippendale furniture. Delacroix had
found new motifs in Algeria. Whistler and the impres-
sionists admired and copied Japanese color prints. David
and Corot had looked at Rome with a certain nostalgia.
The Pre-Raphaelites had set their eyes on the Italian
primitives. But all these attempted to translate the past
into their own Gauguin
times, or to imitate the strange.
had reversed the process: he retreated from the milieu in
which he was born to an elsewhere that he thought
incorporated the past.
Since the primitivist sees perfection in the past he is
thereby opposed to the positivist who believes in an evolu-
tionary progression to a perfect state of the future. The
former is thus opposed to modern scientism, and in paint-
ing he disparages realistic techniques. Gauguin states the
point of difference in his criticism of the impressionists:

They pursued their searches in accordance with the eye and not
toward the mysterious center of thought, and consequently fell into
scientific rationalizations . . .
.*
56 THE AESTHETIC THEORIES OF FRENCH ARTISTS

The mysterious center of thought or state of the artist's


soul is inexpHcable. It is only through works of art that
one experiences it:

There an impression resulting from any certain arrangement of


is

colors, lights,and shadows. It is what is called the music of the


picture. Even before knowing what the picture represents [as when] —
you enter a cathedral and find yourself at too great a distance to

make out the picture frequently you are seized by this magic accord.
Here is the true superiority of painting over other forms of art, for
this emotion addresses itself to the most intimate part of the soul.^

In art color symbolizes this interior emotion:

Color, being an enigmatic thing in the sensations it gives us, can


logically only employed, not
be used enigmatically, every time it is

for drawing but for giving the musical sensations which proceed out
of its own nature, its own interior, mysterious, enigmatic force.®

The aesthetic of Gauguin, as of the other artists, is

inadequately stated in his writings. In the few quotations


cited above he has expressed his position on the important
issues. In Avant et apres he describes in a parable of some
length his own technique; but nothing new to his aesthetic
would be contributed by an exposition of it.®
The mysterious center of thought whose representation
is possible only in symbol is a reinterpretation of the

Schellingian Absolute. To Schelling nature and mind are


the dual manifestations of an underlying reality. What
this is we know only obscurely and we approach it best
through aesthetic experience. Scientific explanations are
superficial, for science describes only the symbol itself. Art
is the highest form of human activity, for through it one

experiences this deeper reality. To Gauguin the work of


art and the feelings of the artist are both expressions of a
common center. We know this center within us more
clearly as it becomes objectified in a work of art. True
SYMBOLISM AND FAUVISM 57

art is not a technical tour de force but a symbol of the


state of the artist's soul. So for Gauguin, too, art as a way
to reach the source of all experience is able to go beyond
science. Certainly the artist's theory echoes the romantic
poetry of Schelling's philosophy.
Elaborations on these aesthetic ideas suggested by
Gauguin were written by Maurice Denis, beginning with
a manifesto in 1890 entitled Definition du neo-tradition-
nisme and extending through more than thirty years of
critical writings of diminishing importance. ^° The name,
neo-traditionism, thus used to describe the movement, was
indicative of the religious interpretations which Denis put
upon symbolism, attempting, in looking back toward the
sacred art of the Renaissance, to make it a kind of glorified
Pre-Raphaelitism. The basic supposition upon which he
built his aesthetic, however, at least raised it above the
puerility of these earlier English painters. The much
quoted opening sentence of his manifesto is ample evidence
of this:

It —
must be recalled that a picture ^before it is a picture of battle-
horse, —
nude woman, or some anecdote is essentially a plane surface
covered by colors arranged in a certain order.^^

Denis opens his essay with this sentence as though it


were a truth we must recall and a self-evident proposition.
After all, a painting as an object is precisely a surface of
colors in orderly arrangement. This seems so obvious to us
today that we are apt to overlook the importance of such
a statement in 1890. To the public at that time this
remark was revolutionary. One still looked at a painting
as a picture of something, and to be told that subject
matter is only secondary demands a complete re-education
of one's artistic vision. Comparison of a painting with a
natural object is no longer the basis for an aesthetic judg-
ment, but the spectator must adjust his vision to the
58 THE AESTHETIC THEORIES OF FRENCH ARTISTS

intentions of the artist. The consequences of this astonish-


ing statement were twofold. In the first place, the artist
was no longer practically bound by a relationship between
his picture and any natural object in respect to form and
color. These qualities could become variable in the in-
terests of expressive power. That such was the program of
symboHsm is undeniable; one need only look at Gauguin's
painting to verify In the second place, theoretically a
it.

new basis of relationship had to be sought between the


work of art and its model.
Denis tried to clarify this new relationship by seeking a
re-definition of nature for painting. If we accept the cur-
rent theory of art, he continues, that nature is the total
of optic sensations, we have not given just place to the
power of cerebral habits Does not Signac set out
on vision.
to prove that chromatic perception is necessary, and does
not academic Bouguereau believe he copies nature.^^ ^^ By
our own will we see nature in art, and the painter sees art
in nature:

You will see in each picture whatever you will have wished to see
there. . . . And the converse is true. There is an ineluctable tendency
among painters to carry back to the aspects of a picture already seen
appearances perceived in reality. ^^

As a result of this practice modern artists have formed the


habit of interpreting their sensations. Litterateurs call this
habit temperament. When one says art is " nature seen

through a temperament " ^^


the definition is just because
it is vague, but it leaves uncertain the important point,

namely, what is the criterion of temperament. The defini-


tion covers the painting of those who strive for literary
effects by failing to go beyond a superficial vision. Nature
as surface qualities is le trompe-Voeil of the crowd, the

grapes of antique painting which the birds came down to


peck, the battle scene of Detaille, where one doubts
SYMBOLISM AND FAUVISM 59

whether the figures are on canvas.^^ Nature itself is


real or
constantly changing. and cerebral habits made
Our desires
the nature which is the subject of our sensations only an
occasion for artistic creation. Nature for the artist, as
Denis was definitely to state years later, is " a state of his
own subjectivity." ^^
The parallelism in art is not to be
found between the work of art and the subject matter,
but between the work of art and the aesthetic emotions of
the painter. The meanness of tromfe'Voeil is a literary
effect.
The value of a work of art arises in the artist himself,
not in his subject: " From the state of soul of the artist
comes, unconsciously or almost so, all the feeling of a work
of art." ^^ The artist himself brings into being the beauty
of the work of art. This beauty is the expressive form of
the work itself. It is the Byzantine Christ, which is a
symbol for the religious feeling of its maker, in contrast to
the modern copied, literary Christ. " In all periods of
decadence the plastic arts have shed their leaves in literary
affectations and naturalistic negations." ^^ The expressive
quality of a work of art lies in the manner in which the
subject presented, not in the subject itself. Before a
is

Calvary scene " our superior impression of the moral


order . . . could not arise from the represented motif or
motifs of nature, but from the representation itself, from
the forms and colorings." The work of art is an arrange-
^^

ment of colors which corresponds to the emotion of the


artist. It is a symbol:

Art is the sanctification of nature, of the nature of the whole world


which is What is great art, that which is called
content with living.
decorative, but the disguise of everyday sensations and natural
objects, in sacred, hermetic, imposing icons.^o

This conception of the correspondence between the emo-


tion of the artist and the plastic representation is the
60 THE AESTHETIC THEORIES OF FRENCH ARTISTS

distinguishing point between symbolism and other theories


of art. It is a felt correspondence, is intuitive, existing
between the whole work of art in its entirety and the feel-
ing of the artist. The correspondence sought by the neo-
impressionist was an objectively determined correspon-
dence between a specific emotional reaction and the con-
dition of a particular element in the work. The work of
art is calculated to convince and is carefully laid out like
an argument in syllogistic form. The symbolist work of
art is an emotionalized burst, and the verity of its cor-
respondence with the feeling of the artist finds its justifica-
tion precisely in the feeling which determined its own
embodiment. This notion of art as symbolic Denis set
forth repeatedly.^^ It is also found in slightly modified
form in Matisse's defense of fauvism.
Fauvism, like impressionism, is best described as a co-
incidence of taste among a group of artists for a more or
less indefinite period of time. It received its impetus from
the free use of color and the attraction for primitive art
which symbolism introduced. The painters of the group
first called themselves les incoherents and were the princi-

pal representatives of the newly formed Salon d'Automne


of 1903. In 1905 the name by which they were subse-
quently known was given them by the critic, Louis Vaux-
celles, who described the gallery in which they exhibited in

the autumn salon of that year as a cage aux jauves, derid-


ing the wild quality of their color work. In three more
years most of the group had moved on to other styles of
painting. The group was Matisse.
spiritual leader of the
Theoretically their point of departure was the opening
sentence of Denis' Definition of 1890, that a painting is a
pattern of colors.
Matisse published the theory of fauvism in a lengthy
article entitled Notes d'un peintre in 1908 ^^ and supple-
SYMBOLISM AND FAUVISM 61

I men ted his ideas in two short interviews in 1929.^^ The


cabn theoretical presentation of the 1908 article seems in
no way analogous to his markedly rhythmical painting.
But both show the Bergsonian anti-intellectualism that
was so strong in the fields of practical action in the decade
preceding the First World War.^*
His heritage from Gauguin is immediately visible:

What I seek to get above everything else is expression. ... I


cannot distinguish between the feehng I have for hfe and the artistic

technique with which I translate it. . . . The expression is the whole


disposition of my picture. The place the volumes occupy, the space
about them, the proportions, all have their part.^^

The vehicle of expression in painting is the compositional


arrangement of the picture. The painter concentrates his
feelings by it. But certain feelings are fleeting, such as
charm, lightness, and the momentary interpretations
caught by the impressionists. If his subject is in motion,
for instance, he must not isolate in thought the present
moment from that which precedes or follows. He must
be conscious of the entire action and endeavor to render its
quality .^^ Expressive composition makes its own demands
upon the artist:

It is not possible forme to copy nature servilely, because I am


forced to interpret and to subject it to the spirit of the picture.
it

When I have found all my harmonies of tones there ought to result


a complete accord of living colors, a harmony analogous to that of
musical composition. ^^

Color has an important function in composition:

The dominant tendency of color is to serve expression as best it

can. . The expressive side


. . of colors is imposed on me in a purely
instinctive way.
The choice of my colors does not repose on any scientific theory, it
is based on observation, on sentiment, on my own sensibility. ^^
62 THE AESTHETIC THEORIES OF FRENCH ARTISTS

The instinctive quality of art is constantly emphasized


through all this. The painter does not proceed by thought
so much as by sensibility. Consequently he does not work
by established rules. Rules belong to his own nature:
" Rules have no existence outside of individuals." ^^ His

instinctive use of color as a compositional element may


even lead the painter to distort the figures in his work.
Natural appearances must always give way to composi-
tional arrangements. The principle of distortion is of
prime importance in expressionist aesthetics. Painters
who proceed upon the imitation theory strive to depict
objects as they are perceived. On the other hand, those
painters who attempt to make art a vehicle for express-
ing their consciousness of life mold their subject as they
translate it from natural space to the canvas until they
are satisfied it embodies some meaning which any other
form of rendition would not convey. They are not
guided by perception but by ** aesthetic intuition." For
them a work of art has reference not only to the object
which it represents but to a feeling on the part of the
artist. As such it is a symbol. If one refers the work to

the object then it falsifies that object. The principal refer-


ence of the work is to the meaning which the artist gave it,
and the work does not falsify that meaning (insofar as the
artist's technique was able to express it adequately) .^^
Intuition as contrasted to reason is opposed to discursive

modes of thought. It seems more in harmony with the


imaginative life of the child or the primitive. And so the
expressionist painter finds inspiration in the distortions of
primitive art. In fairness to Matisse must be observed
it

that however much his theory has seemed purely expres-


sionistic,he does state that neither the imitative nor the
intuitive principle should be preached to the exclusion of
the other .^^
SYMBOLISM AND FAUVISM 63

Since the work of art is a symbol whose principal refer-


ence is to the feeling of the artist, the spectator must make
the reference to the subject subordinate to this more im-
portant one. And this being a reference instinctively made,
the ideal work of art would open its meaning to us without
intellectual effort:

When I see the frescoes of Giotto at Padua I am not concerned to


know what scene from the hfe of Christ I have before me, but im-
mediately I understand the feeUng which it conveys for it is in the
lines, the composition, the color, and the title only confirms my
impression.
What I dream of is an art of equilibrium, purity, and tranquillity,
without disquieting or disturbing subjects, which could be for the
mental worker, the business man, and the man of letters too, for
example, a mental refreshment and relaxation, something analogous to
a good easy chair in which one rests from his physical fatigue. ^^

The fauves cast the perceptual world into the back-


ground an intuited world. This intuition is a
in favor of
consciousness of which is of the nature of a religious
life

attitude. They make no distinction between their con-


sciousness of life and their manner of expressing it. The
work of art is thus a sample of the feeling it expresses. It
is the only rendition of that feeling in concrete form. The
work of art is therefore a symbol in a special sense. We
usually think of a symbol as that which represents an
object in virtue of an idea in the mind of the one using it,

without which the connection would not exist .^^ That is,

we say 3 is a symbol for the number three. But there is


nothing in the shape of this figure itself which suggests
three; by recognized convention the symbol has come to
represent the number. This interpretation of the symbol
will not hold when we come to the theory of the symbolic
function of a work of art as presented above. Here a trans-
formation has taken place by which the work of art is at
the same time the objectified feelings of the artist. The
64 THE AESTHETIC THEORIES OF FRENCH ARTISTS

mind moves from the symbol to the symbohzed of its


own accord. There is no mediating ideational rule or
convention.
Croce's aesthetic is similar. His philosophical problem
being to examine human activity,he recognizes two such
types of activity, knowledge and practice. Knowledge he
further divides into two varieties, a pre-perceptual kind
obtained through the intuition, and intellectual knowledge.
The artistic activity is of the first, for it is concerned with
the expression of our interior feelings. As a matter of fact
intuition is precisely expression. Croce further agrees with
symbolist and fauve theory in denying that beauty or ugli-
We cannot
ness inheres in certain forms or certain colors.
weigh or measure these elements as embodying any aes-
thetic value in themselves, as the neo-impressionists had
done. Denis and Matisse state that one finds the feelings
of the artist expressed through the colors, the forms, or
the composition, but they do not attribute that feeling as
arising from any particular arrangement of elements. The
artist's sensibility guides his arrangements and tells him
when they give a satisfactory rendition of his emotions.
As noted in considering the aesthetic of Renoir, Bergson,
too, was famous for making the appreciation of art in-
tuitive. Aesthetically intuition is a sympathetic feeling for
human emotional content in an object. It is the power
par excellence of the artist.
In symbolist and fauve theory the value of art lies in
its expression of human feelings. The painting is a symbol
of the painter's psychic state. The spectator is immedi-
ately aware of this emotional content in the work. Denis
says there is a correspondence between the arrangement
of colors in a painting and the emotion of the artist.
Matisse says that the composition of a painting must give
a discharge of feeling. In both cases it would seem that
SYMBOLISM AND FAUVISM 65

the value belongs to the formal elements themselves. The


important thing is that whatever value the color or com-
position has owes to the emotional content which the
it

artist has put there. Yet the content is itself this color and
composition. The painting is what it means and means
what it is. Also it seems clear that this symbolic meaning
which is the feeling of the artist can be experienced by the
spectator. He simply responds aesthetically to the picture
by tuning to the artist's emotional wave length, so to
speak. Being painters and not aestheticians these men who
have recorded their theories of art have stopped short of
the crucial question which their theories bring up, or fail
to see the tenuous and controversial positions into which
they are led. Is the emotional reaction of a spectator
before a painting the same as the artist's.? Is there no
place for individual reaction to works of art.? If all spec-
tators and the an emotional coincidence before
artist find
a work of art, then
not art a language, a thing of public
is

experience.? If this is so, then is not an attempt to make


a calculus of that language or to explore it semantically,
as the neo-impressionists tried to do, a justifiable and
necessary labor.? Or are we back to a mystical interpreta-
tion of art which says much but explains little, telling us
finally that art is inexplicable.?These painters fail to see
their theory opens up numerous questions.^*
When a painter establishes a correspondence between
the work and his feelings, as these painters do, he is
attributing to the object his own emotions and kinaesthetic
sensations. In this respect symbolism and fauvism are
empathy theories. In the last quarter of the nineteenth
century and the first quarter of the present one Einfiihlung
occurred in various forms. As first used by Robert Vischer
in 1873 the term Einfiihlung was descriptive of the act by
which one projects into the object before him his own
66 THE AESTHETIC THEORIES OF FRENCH ARTISTS

psychic feelings. He
emphasizes or " feels-into " the object
quaUties which properly belong to himself. As such the
word isa psychological, descriptive term for the childish,
primitivistic, or artistic habit of animism. The aesthetic
experience is explained by the various exponents of Ein-
fiihlung as a transfer of one's own kinaesthetic sensations,
muscular adjustments, or emotional reactions from oneself
to the work of art or natural scene. In the classical for-
mulation of the doctrine made by Lipps a study is made
of the relations of various formal elements such as design
or color to particular emotions. As a result laws governing
these formal elements of all aesthetic objects can be formu-
lated. In this respect the neo-impressionist theory of
Seurat in its definite statement of the relations between
emotions and line, color, and shade is also an example of
empathy theory. A critique of Einfiihlung is not here in
order. The theory merely supplies a name describing the
habit of transferring one's emotions from subject to object,
an attitude which is fundamental to symbolist and fauve
theories of art. The belief that certain forms and colors
have an expressive value in themselves does not belong
exclusively to the painter of this time. The introduction
of the " new " experimental psychology, begun with
Fechner and continued particularly in Germany, had
brought the techniques of the laboratory to bear on the
problem of hedonics. From this work a psychological
aesthetic has resulted, wherein a statistical definition of a
pleasing object is attempted. Studies in the affective
quality of various simple shapes and colors, and an aes-
theticsbased on experimental investigations are now
important branches of psychological inquiry.
The projection of feeling into a work of art by artistand
spectator also finds expression in formal aesthetics. George
Santayana only a few years previously had explained the
.

SYMBOLISM AND FAUVISM 67

perception of beauty as objectified pleasure, noting this


transfer of quality to the object.^^
The sentiment in art was part of a
rise of religious
definite current in French thought at the time. The renais-
sance of Catholic philosophy in Maurice Blondel, Peguy,
Laberthonniere, and others at the turn of the century was
as strong as traditionalism had been almost a century
before. It was probably this new traditionalism that
accounted for some of the symbolists moving in the
direction of religious painting exclusively.

NOTES
^ The Vogue for Japanese color prints was largely responsible for
bringing about this change in style.
' Cf. R. H. Wilenski, Modern French Painters, pp. 98, 137, 142.
' For Gauguin's writings see bibliography. Those passages which
bear on art and sum up his aesthetic have been collected by Jean de
Rotonchamp, Paul Gauguin (Chapter X) Where possible references
.

will be made to this book as the most convenient and practicable

source since I have been unable to obtain the original sources in


several instances.
* J. de Rotonchamp, Paul Gauguin, p. 152.
" Ibid., p. 245.
• Ibid., p. 243.
' Ibid., pp. 246-247.
" Ibid., p. 244.
' Gauguin, Avant et apres, pp. 55-59; also in Rotonchamp, op. cit.,

pp. 247-250.
^"Maurice Denis, b. 1870, painter associated with the Nabis and
the symbolist movement and an important theoretician. Journeys to
Germany and Italy resulted in his being deeply influenced by the
Nazarenes and the Italian primitives. His religious inclination is re-
flected in his preference for sacred art, and in his founding, in 1919
with Georges Desvallieres the Atelier d'Art Sacre. His principal works
are church and public building decorations. His critical essays have
been collected in Theories, 1890-1910, and in Nouvelles theories, etc.
(cf. bibliography)
68 THE AESTHETIC THEORIES OF FRENCH ARTISTS
^^ " Definition
de neo-traditionnisme," first published in Art et
Critique, August 23-30, 1890, reprinted in Maurice Denis Theories,
1890-1910, pp. 1-13, to which edition page references are here made.
Quotations by the kind permission of Art CathoHque. Above passage
p. 1.
^'Ibid.,p.l.
" Ibid., p. 2.
^*
The Emile Zola gave in Mes Haines,
definition p. 24 in the
Bernouard edition of his Oeuvres Completes.
" Theories, pp. 2-9.
^^
" De Gauguin et de Van Gogh au classicism," 1909, Theories,
p. 267.
" " Definition," Theories, p. 9.
" Ibid., p. 8.
" Ibid., p. 10.
''
Ibid., p. 12.
" Vide in Theories, pp. 41, 180, 259-260, 263.
^^
Published in the Grande Revue, December 25, 1908.
^^
E. Teriade, " Interview with Matisse," in L'intransigeant, Jan.
14 and 22, 1929.
^*
Vide, e. g., L. Quintanilla, Bergsonisme et politique, Ph. D. dis-
sertation, The Johns Hopkins University, 1936,

H. Matisse, " Notes d'un peintre," Grande Revue, Dec. 25, 1908,
p. 733.
" Cf . ibid., pp. 732-736.
'"
Ibid., p. 739.
'"
Ibid., pp. 739-740.
" Ibid., p. 744.
Cf. W. M. Urban, Language and
''°
Reality, pp. 423-426.
" Cf. Matisse, op. cit., p. 742.
'"
/6ic?., pp. 741-742.
" Cf. C. S. Peirce, Collected Papers, vol. 11, par. 299.
'*
Cf. L. Stein, The A-B-C of Aesthetics, especially chapters IX,
X. XI.
'"
G. Santayana, The Sense of Beauty.
CHAPTER SIX

EARLY CUBISM
ANOTHER variant of realist aesthetics was possible,
^ljL and that was to declare that the artist's business is

to paint the real object but to reinterpret what is meant


by a denying that reality is in perceptual
real object,
appearance but is a deeper essential quality in an object.
This was the way of Cubism.
The same year that Matisse published his Notes d'un
peintre he gave the name cubism to a new manner of
painting in which his former confrere, Braque, had ren-
dered a landscape. He shook his head with genuine con-
cern and spoke of les petits cubes. Influenced by Negro
sculpture and Cezanne's later work and encouraged by the
latter's remarks about treating nature as cones, spheres
and cylinders, the cubist movement had solidified by 1911
when the first exhibition was held at the Salon des Inde-
pendents. Guillaume Apollinaire was the critical champion
of the group.
The fortunes of cubism have been long and varied.
Various stylistic changes have been duly noted and
labeled.^ Its influence has been far reaching. Construc-
tivism, neo-plasticism, purism are variations on it.^ It
has given birth to sporadic schismatic movements like
orphism.^ From its inception it flourished more than
twenty years.* The immediate impetus of the movement
was in its early phase until the First World War and the
present examination will be limited to this phase. A suffi-

cient insight can be gained within this limitation.


An essay on the doctrines of cubism was published in
1912 by Albert Gleizes and Jean Metzinger under the
title of Du cubisme. This small book represents the first
70 THE AESTHETIC THEORIES OF FRENCH ARTISTS

important manifesto issued by any of the artists of the


movement.
Preceding the theoretical discussion an important sec-
is

tion on the history of painting since Courbet to set in rehef


the fundamental assumptions of cubism. Courbet, the
authots say, accepted the retinal world without any intel-
lectual control. His principal concern was to render the
subject exactly as it appeared to him. But the world of
the retinal image is a vitiated world, and a true realist
would not be one who limited himself to it but takes it
as already controlled by his artistic intent:

The visible world does not become the real world except through
the operation of thought, and . . . the objects which strike us with
the most force are not always those whose existence is the richest in
plastic truths.^

What this real world is unfolds as we move through the


preliminary section. After Manet the realist tendency
diverged. There was the new superficial realism of impres-
sionism where the paintings were productions from purely
retinal stimulations and in which forms atrophy and dis-
appear. And there was the profound realism of Cezanne
where painting is an art that gives plastic consciousness
to instinct, where painting searches for reality.^
Objects have plastic truth when their volumes and
dimensions can be satisfactorily rendered in the medium
of painting. Plastic truth is the result of reason. Artistic
space and the space of things in sensation are at most only
analogous. This distinction between the real world and the
visible world, presented as the difference between Cezanne
on the one hand and the impressionists on the other, is
fundamental to cubist doctrine.
Plastic truth is found only in anti-decorative art. Deco-
rative art, that which exists as having its raison d*etre
outside of itself, " only becomes a living work in virtue of
EARLY CUBISM 71

the relations which are estabUshed between it and de-


terminate objects."^ The relationship may be that be-
tween a picture and its surroundings. For instance a
Resurrection scene is appropriate to the church in which
one finds it. In anti-decorative art a " picture carries in
itself its reason for being." ® It may be carried away from
church to drawing room, to gallery, and it always remains
complete, satisfying in itself. " It is in accord with every-
thing, the whole universe; it is an organism." ^ Or the rela-
tionship may be that between the pictorial world and the
visual world of the model. Both may be a decorative
pattern, or both may have deeper aspects:

Oil painting today permits the expression of notions [heretofore]


deemed inexpressible, those of depth, density, and duration, and it
incites us to represent according to a complex rhythm and in a
restricted space a veritable fusion of objects. ^°

Non-decorative art expresses the permanence of objects


in substantial physical form, and deals with extended
matter in its unchanging aspects: extension, density, dura-
tion. These are qualities of an object which are known by
analysis and by inference. It is such an object that pro-
found art strives to depict. A work of art of this profound
manner is in agreement with all other things of the uni-
verse. It has plastic truth.
This attempt to replace in painting the visible, retinal
world by a rational world which is declared to be more real
is a direct parallel in aesthetics to the critique of scientific
methodology made by Meyerson. The first edition of
Identite et realite appeared in 1908, the same year that
Cubism began to attract attention, and the second was
issued in the same year as Du cuhisme, 1912. Meyerson
distinguished positivistic legalism and causality. Legalism
can only give laws which must be accepted. A causal
explanation yields an equality whereby cause and effect
72 THE AESTHETIC THEORIES OF FRENCH ARTISTS

are identified. This demands that we must make mental


simpHfications and strive to get rid of sensory qualities,
for sensory quahties change and we can get no identity
with them. Scientific method must, and historically did,
proceed by finding the permanence behind change. We say
that we have explained something when observable effects
are rationally equated with a cause. For example, the
symptoms of a disease announce to the physician that in
the body of the sick person there are, for instance, a
number of particular germs above a given amount which
have caused this illness. We say that the disease is the
presence of these germs in the body, but it is also the
abnormal condition of that body, the symptoms that are
observable. The disease is identified with the effect and
with the cause. An explanation is made when thought
operates on observation.
Superficial realism, the cubist might say, yields only a
legalistic artistic expression on the level of appearance; the
new profound art of Cezanne and the cubists is a form of
expression on the level of reality.
The second section of Du cuhisme deals with what is
called " the integration of plastic consciousness." This is
a term which covers both artistic sensation and the pro-
cedures of artistry. Plastic consciousness " discerns a
"
form and then proceeds to render perceptible the relation
between the world and the thought of the artist. ^^
To discern a form implies, aside from the visual function and the
faculty of being moved, a certain development of the spirit. The
exterior world amorphous to the eyes of the majority.
is

To discern a form is to verify it by a preexistent idea, an act which


no one, unless he is a man we call an artist, does without exterior
aid.i2

The form is the permanent qualities of the world before


us. Its recognition, not being an act of sensation, which
EARLY CUBISM 73

catches only the fleeting aspects, is a kind of mental aware-


ness. The word discernment seems a happy choice. It
suggests recognition and ideation. The artist's discern-
ment is superior to other men's because he has a capacity
of " spirit " which enables him to seize the forms of things
without immediate referenda. The child, it is said, must
compare his sensations to the picture-book; the man com-
pares his to works of art to coordinate them and render
them subject to mental control." The child recognizes the
animal in the zoo from his picture-book drawings. The
grown person sees the Grand Canal through Guardi and
Turner and Westminster Bridge through Monet. The
artist sees the world directly and without external aid.

The artist, having discerned a form, prefers that form insofar as it

presents a certain intensity of analogy to his preexistent idea and


consequently — for we impose our own preferences on others —he exerts
himself to enclose the quality of that form (the immeasurable sum of
affinities glimpsed between the visible manifestation and the tendency
of his spirit) in a sign proper to affect others. . . . The painter, care-
ful to create, rejects the natural image as soon as it has served his
purpose.^*

The must reject the natural image because a work


artist
of art which has an application to two categories, the
natural and the artistic, is a confused result.^^ We ought
not judge art by similarity to nature. A landscape scene in
art, for example, should be something of entirely different
spatial relations from that landscape in nature. The di-
mensions of art are not those of observation. If the dimen-
sions and proportions of the natural scene are relatively
intact in a painting, the genius of the painter is not thereby
exhibited. The picture isnot aesthetically effective. The
mission of the artist is not to imitate but to translate.
Yet " the reminiscence of natural forms could not be abso-

lutely banished. . . . the cubist painters know this."
74 THE AESTHETIC THEORIES OF FRENCH ARTISTS

The abstraction is not complete. The object as translated

by art is still a " reminiscence " of the natural object.


The between visual space and pictorial space
difference
dictates the difference between the natural image and the
artistic rendition.
Visual space depends upon the " sensation of conver-
gence." In art this convergence is represented by perspec-
tive. But need not depend upon perspective.
pictorial space
Infractions of the rules of perspective do not compromise
spatiality. In Chinese painting there is space without
perspective."

In order to establish pictorial space it is necessary to have recourse


to tactile ^^ and motor sensations and to all our faculties. Our whole
personality, contracting or expanding, transforms the plane of the
picture. When this plane in reaction is reflected on the understanding
of the spectator pictorial space is defined: a sensible passage between
two subjective spaces. ^^

Pictorial space is the space which embodies discerned


forms, tactual, motor, and ideational, besides purely visual
ones. It is the context in which these forms are rendered
for the perception of others.
Ideational forms and visual shapes are in constant inter-
play. An ellipse may become a circle when inscribed in a
polygon. One emphatic form may govern a picture; one
leaf of a tree may be imitated and thereby the whole tree
seem to be represented. " The eye quickly makes the mind
^° The mind continually plays with
interested in its ways."
the nuances of shape. " Composing, constructing, design-
ing come back to this: regulating through our own
all

activity the dynamism of form."^^ Shapes are only


" nuances of the notion of plenitude," " which plenitude

is form. In the design of a picture there is a relationship


between straight lines and curves. A work of art moves us
when it embodies this relationship as an indefinite diver-
EARLY CUBISM 75

sity (not all straight lines or curves, or an equal balance


between them) .^^ Exterior shapes are constantly being
modified by one another but our mind sees the permanent
form beneath them.
The artist tries to symbolize the quality of this form.
He does this by catching the weight, density, and bulk, the
visual, kinaesthetic, and intellectual aspects of his model.
In this manner he makes the work of art an organism
in whigh all qualities are coalesced in a kind of vital
conjunction.
Nuances of form and nuances of color also ajffect each
other:

All inflection of form is heightened by a modification of color; all

modification of color engenders a form. 2*

Ideational and sensory content constantly interplay.


This rational interplay is the only reality:

There is nothing real outside us; there is nothing real but the coinci-
dence of a sensation and an individual mental direction. It is not our
intention to cast in doubt the existence of objects which affect our
senses;but there is reasonable certitude only in regard to the image
which these call forth in our minds. ^^

Since forms are ideational there is no single one belong-


ing to each object:

An object hasn't just one absolute form; it has many. It has as


many as there are planes in the domain of signification. ^^
There are as many images of an object
as there are eyes to see it;
there are asmany essential images of an object as there are minds
to understand it.^^

The object the painter knows is different from the object


the scientist knows; the object as seen from one position
is not the object as seen from another. No one of these
aspects alone is the object; it is the fusion of all these
76 THE AESTHETIC THEORIES OF FRENCH ARTISTS

mental appearances. This is the complete object that the


cubist tries to crowd into pictorial space:

We are sure that the least sagacious will soon acknowledge that
the pretention of making the weight of a body apparent and the time
spent in numbering its diverse aspects is as legitimate as that of
imitating the daylight by the shock of a blue and an orange. Then
the fact that an object moved about in order to have its many
is

successive appearances caught, when these appearances blend into a


single image and reconstitute the object in duration, will no longer
rouse the indignation of reasonable people. ^^

We are warned, though, that this desire to get down in


pictorial form such an aggregate object must not be taken
too literally. The artist should not painfully juxtapose the
six faces of a cube or the two ears of a model seen in
profile.^®

This notion of a conceptual object as a system of


appearances also occurs in formal philosophical work at
about the same time. Bertrand Russell, for example, under
the admitted influence of A. N. Whitehead began to treat
" things and the whole conception of the world of physics

as a construction rather than an inference.'' ^° The appear-


ance of objects changes with the " point of view " of the
observer. The perspective, that is appearance, of one
observer is not that of another observer. Unperceived
perspectives are also conceivable. The object itself is not
the cause of these appearances but is the whole system of
them. " Given an object in one perspective, form the
system of all the objects correlated with it in all the per-
spectives; ." ^^ (this sounds like a program for cubist
. . .

painting) When the time element is considered the object


.

becomes an event spread out in space-time. If it is located


at any place it is located at the point of intersection of all

its appearances. But that is the only place at which no


perspective is possible, so that the object is everywhere
but in the position we attribute to it. We must go all
EARLY CUBISM 77

around it to find it. It is a " logical construction " of this


system of aspects. A cubist picture is the pictorial image
of such an object.
The parallel between cubist theory and new realist
philosophy is surprising at first, but undeniable. Whether
the relationship was accidental coincidence or one in which
there are traceable connections is a question that conjec-
ture will not settle.
Gleizes and Metzinger's theory of cubism makes paint-
ing a rationalized enterprize. The business of the artist is

no longer to reproduce the surface appearance of natural


objects, or the world as he emotionally experiences it. The
object he paints is a logical formalization of the natural
model. Things in the natural world are measured by their
correspondence to an idea. The categories of painting and
the natural world are separate, but the painter still tries to
move from one to the other. He has not yet given up his
model completely. He views the world through the mind
and molds it to his view.

NOTES
^ Cf. the distinction between analytic and synthetic
cubism made
by A. H. Barr, Jr., Cubism and Abstract Art; the fourfold distinction
made by Apolhnaire, Les peintres cubistes, and the classification of
Wilenski, The Modern Movement in Art.
^ Constructivism:
Russian art movement, 1913-1922 being the
years of its greatest activity, which constructed from glass, wood,
plaster, metal, etc. models of abstract design.
Neo-Plasticism: style of painting of the Dutch artist, Piet
Mondrian, in which the canvas is divided into rectangles of color and
greys by thick black lines, cf. manifesto of Modrian, Le neo-plasti-
cisme, 1920.
Purism: French theory of painting which flourished in the decade
following the war reacting against what it considered decorative
tendencies in cubism and drawing its inspiration from the efficient and
functional appearance of machines. Cf. Ozenfant and Jeanneret,
Apres le cubisme and La peinture Moderne.
78 THE AESTHETIC THEORIES OF FRENCH ARTISTS
' Orphism: manner of painting employed about 1912 by Robert
Delaunay and Frank Kupka, consisting in the use of abstract color
designs for their own sake. It was influential in bringing subsequent
cubism to a freer use of color than it had previously employed. Cf.
letter of Robert Delaunay in G. Coquiot, Les independents, 188 Jf-
1920, pp. 181-182, " everything is color and movement."
*
For above information concerning cubism and related movements
cf. A. H. Barr, Jr., Cubism and Abstract Art.
^
Gleizes and Metzinger, Du cubisme, p. 6.
" Ibid., pp. 8-10.
'/6ic^., p. 11.
® Loc. cit.
"^
Ibid., p. 12.
'"
Loc. cit.
" Ibid., pp. 13-15.
''
Ibid., p. 13.
^^
Loc. cit. Does not, for example, a European's picture of a land-
scape seem queer to a Chinese who has become accustomed to seeing
landscape through his native art?
'*
Ibid., pp. 13-14.
^'
Ibid., pp. 14-15.
''Ibid., p. n.
''
Ibid., p. 18.
^*
The importance of tactile sensations in painting had been em-
phasized just prior to this time by Berenson in his Florentine Painters
of the Renaissance.
'^
Gleizes and Metzinger, op. cit., p. 18.
'"
Ibid., p. 19.
^^
Loc. cit.
" Ibid., p. 20.
"Ibid., pp. 20-21.
'*
Ibid., p. 29.
" Ibid., p. 30.
""/bicZ., pp. 30-31.
" Ibid., p. 32.
''Ibid.,p.Se.
^'
Picasso, however, does a similar thing in his later cubist work,
showing two eyes in a profile face in his Seated Woman in 1927.

B. Russell, Our Knowledge of the External World, 1914, p. vi.
" Ibid., p. 89, Lecture III, pp. 63-97.
CHAPTER SEVEN

SURREALISM
THE FINAL
tion of an
phase of this history is the considera-
aesthetic which denies all the theses of
realism and expands upon the symbolist and expressionist
doctrines until they are scarcely recognizable. Surrealism
has led an uninhibited and boisterous existence among us
for nearly twenty-five years. As the exploration of a par-
ticular point of view which is in revolt against the accepted
standards of art and criticism it has faced the snobbism
of aesthetician and critic. Since it has come of age as a
taste whose influence bids fair to become as far-reaching
as its antecedents are numerous, whose rise and impor-
tance in interwar culture is interesting and astonishing, it
demands sympathetic study.
Though not born until eight years later surrealism was
conceived in 1916. It received its name from its literary
godfather, Guillaume Apollinaire, who designated as sur-
realiste adrama which he wrote in that year entitled Les
mammelles de Tiresias. The word was later adopted by
Andre Breton to describe the attitude of dream thought
and psychic automatism that was engaging his attention.
Its mother was dada, from whose body it was born.
Dada made its debut during the First World War in
1916 at the Cabaret Voltaire in Zurich, and its nihilistic

attitude quickly spread among the disillusioned artists of


Europe. It was so named, as one story goes, because dada
was the first word which met the eye of Tristan Tzara
when he inserted a paperknife in a dictionary at random
and opened the book to that place. If its adherents ever
had any illusions about art they abandoned them in dada,
for dada was the display of nonsense for its own sake.

79
80 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Manifestoes were read by the score to damn art and


thought and manifestoes and dada itself:

I write a manifesto and I don't want anything. Yet I say certain


things, and I am by principle against manifestoes as I am also against
principles. ... I write this manifesto to show that one can take
opposite actions at the same time in one cool breath. I am against
action, in favor of continual contradiction and for affirmation also.
I am neither for nor against and I do not explain for I hate good
sense.

Such was the turn of mind out of which surreahsm came.


Surreahsm was formulated and given a program by
Andre Breton, one of the Paris dadaists. A medical stu-
dent as well as a poet, he proposed to investigate the
possibilities offered by literary activity freed from rational
control as a method of psychoanalysis. In his Manifeste
du surrealisme of 1924, he first attempts a definition, de-
scribes it, background. Though not the
and indicates its

work of a painter this manifesto must be considered to


understand the point of view of surrealism, which is more
than a theory of art.
Breton first describes the turn of thought distinctive of
surrealism in contrast to the conventional, realist attitude.
This attitude refers to an absolute rationalism which has
the fixed limits of discursive reason, is always in agreement
with common sense, and confines itself to the tautological
possibilities of logic:

We still live under the reign of logic. . . . But nowadays logical


processes are applied only to resolving problems of secondary interest.
[True] logical ends escape us, however. . . . Under color of civiliza-
tion,under pretext of progress all that rightly or wrongly can be
charged to superstition or fancy has been banned from the mind, all
manner of searching for truth which does not conform to usage has
been proscribed.

On the other hand is the materialist attitude (for so he


SURREALISM 81

which opens up the domain of the imagination and


calls it)
shows the true psychic life hidden beneath rationalization:
The imagination is perhaps on the point of acquiring its rights.
If the depths of our souls conceal strange forces capable of augment-
ing those on the surface or of struggling victoriously against them
it is to our best interests to exploit them. . .
.^

The imagination is identified with a subconsciousness


that is made apparent to us through dreams. Breton
acknowledges the debt of the surrealist to Freud. By
affirming the unconscious surrealism begins to move away
from dada and pure negation, to erect a positive structure.
Accepting the authority of Freud that the subconscious is
most easily apprehended in the dream state which " is con-
tinuous and carries trace of organization " whereas the ,

waking state is only " a phenomenon of interference " *


obeying unconscious suggestions, the surrealist examines
this dream state with a new intent:

I believe in the future resolution of these two states which in


appearance are so contradictory, that of the dream and that of reality,
into a kind of absolute reality, a surreality, if one may call it such.^

By dialectic the two contradictory states, the dream and


reality, are absorbed by a deeper realm. This is the philo-
sophic position of surrealism. The things of the outer
world, though real in the sense that they have their own
independent existence, lose this reality in our thought and
enter into new relationships which are psychical, not
physical. To the surrealist " a tomato is also a child's
balloon " and word like is " sup-
in this relationship the
pressed." ^ Certainly in such a program our suspicions of
the marvelous are put in disrepute and reason is surclasse.
We ask at this point, perhaps, for a description of this
surreality. But shows we misunderstand. The surreal
this
is the last dark continent of the mind; it exists but we
82 THE AESTHETIC THEORIES OF FRENCH ARTISTS

cannot yet describe it for it has never been explored and


its borders have barely been touched at times. It is here

surrealist artistic techniques enter, for they are the


methods by which the poet and the painter can explore
the unconscious as does the psychologist. Surrealism, the
philosophic position, posits the surreal; surrealism, the
method, helps to describe it. The duality is apparent in
Breton's definitions:

Surrealism, N. m. Pure psychic automatism by which it is pro-


posed to express, either verbally, in writing, or by any other means,
the real functioning of thought. A dictation of thinking, in the
absence of all control exercised by reason and outside of all aesthetic
or moral preoccupations.
Encycl. Philos. Surrealism rests on the belief in the superior
reality of certain forms of association neglected before its time, in the
omnipotence of dreams, in the distinterested play of thought. It tends
to ruin the definitiveness of all the other psychic mechanisms and to
substitute itself for them in resolving the principal problems of life.'^

Breton describes the first surrealist method whereby


our activity is freed from rational control, psychic auto-

matism. This is the attempt to record the stream of unin-


hibited verbal imagery for oneself as a psychoanalyst
would that of a patient. In as passive a state as possible
one writes down rapidly his flow of thought. Or one may
draw at random in a kind of doodling.
The first manifesto clarifies the ideological basis of
surrealism and describes one possible method for it.

In the Second manifeste du surrealisme of 1930, Breton


declares that it is the purpose of surrealism to provoke a
" crisis of consciousness " by exposing the falseness of the
old antinomies which have so long prevented mankind
from finding the common point of reference.^ The exposure
can best be made through a study of the problem of human
expression for:
SURREALISM 83

No one, in expressing himself, does better than to adjust himself to


knew he
the possibility of the very obscure reconciliation of what he
had to say on a subject with that he didn't know he had to say but
has said.^

The two -most valuable means of expression are dream


accounts and automatic texts (whether written, painted,
or drawn makes no difference as their automaticity is
evident in any one) They are valuable because they are
.

the means to which one may turn when the object of


artistic activity is not the production of works of art but
the revealing of the psychic life. In artistic inspiration
there occurs a kind of short circuit between an idea and
its response/^ For that reason art is valuable to the
surrealist.
This gives rise to the all important question which
brings us to the heart of the surrealist position. Ifwe do
have such an unconscious self as described why should we
want to know about it and exercise it? The answer is
found in the significance which is attached to the surreal.
Down in this inner psychic life of each one of us we come
to the human crucible itself, the overindividual state
wherein exist the universal truth, beauty, and reality that
all of us can comprehend. Here is not only the man him-

self but mankind. Here is the foundation upon which we

must build our morals, our art and ways of thought, our
life and actions:

The products of psychic activity offer a key capable of


. . .

opening indefinitely that box of multiple depths called man.^^

In Le surrealisme et la peinture Breton presents the pur-


poses and problems of surrealist activity in the plastic arts.
Because mankind has felt the need of fixing his visual
images, a plastic language has grown up corresponding to
the various degrees of his sensations, the things we all see
84 THE AESTHETIC THEORIES OF FRENCH ARTISTS

and remember whether we will to or not, the things we do


not dare see, that we see differently, that we cannot see.
This plastic language is a reality as powerful in itself as

the physical world, forwhen one turns to a picture the


world that surrounds him is no more.^^ It is as easy to put
the natural world in question as it is the world of art.
The attitude of the realist, who takes his model from the
natural world, is the artistic equivalent of the discursive
thinking of the realist in philosophy which the surrealist
hates:

A very strict conception of imitation, which has been given as the


purpose of art, is the root of a grave misunderstanding which we see
persisting even until our time. On the belief that one is capable of
reproducing with more or only the image of that which
less success

have shown far too much timidity in


affects the sensations, painters
the choice of their models. The error they made was in thinking the
model could only be drawn from the exterior world, or even in think-
ing one could be taken. True, human sensibility can confer on the
most common subject quite an unforseen distinction; but it is no less
true that to make the magic power of figuration which some so
happily possesses an aid to the conservation and reenforcement of
what would exist without them is to make a beggarly use of that
power. That is an inexcusable abdication. At all events it is impos-
sible in our present state of thought, above all when the exterior world
appears to be of a nature more and more suspect, to consent still
to such a sacrifice. The plastic work, in order to respond to the
necessity for the absolute revision of real values which all minds
today agree is necessary, will refer to a purely interior model or will
no longer exist. ^^

For the natural model surrealism substitutes the interior


model, the mental world. The expression of this model
reduces itself to the employment of iconographical devices
which are personal to each painter. But do these various
personal pictorial means have any public significance.?
Will the fragments from my subconscious speak to your
subconscious simply because those portions of our minds
both harbor similar repressions and erotic urges.?
SURREALISM 85

The aim of the surreahst is a complete reversal of critical


values. In casting aside the artist's preoccupation with
the rendering of natural objects, he makes such a complete
rejection of this practice that he not only condemns real-
istic reproductions but also rejects any aesthetic based on
the manipulation and expressive value of formal relation-
ships. The surrealist work of art is a vehicle by which
artist and spectator are brought up against the center of
the world where thought and things meet. The work is a
plastic sign of the surreal center and a fragment of it as
well. The beauty of surrealist art is a convulsive beauty
that creates " a physical disturbance in a spectator like a
^*
cutting wind on his face on a chilly day."
Louis Aragon, the poet, in talking about collage art, had
said that was a method more akin to magical operations
it

than to painting. The collage piece, which is an object in


its own right, is put into a larger context which contains

other pieces that are unrelated to the first and to one


another. The result is that the pieces are transformed by
the chance association. An electric bulb by itself is an
electric bulb, but when put in a picture it has entered into
a new context and may become for the artist a young girl.^^
The surrealist sees a new significance. Max Ernst, the
German painter, defines it as " the miracle of the total
transfiguration of any object either with or without modi-
fication of its physical or anatomical appearance." ^^
Two
mutually distinct realities coalesce by meeting through
chance in a subconscious plane. ^^ The collage is an ex-
cellent example of surrealism's non-aesthetic values in
painting.
For Salvador Dali the whole ambition of the artist is to
" materialize the images of concrete irrationality." He
must record the interior model as faithfully and as clearly
as any realist would copy his exterior model. If the artist
86 THE AESTHETIC THEORIES OF FRENCH ARTISTS

does not understand his pictures as he paints them it is

because " their meaning is so profound . . . that it es-


^^
capes the most simple analysis of logical intuition."
The shift of values and overindividual content of the
surreal Dali well expresses:

The subconscious has a symbolic language that is truly a universal


language, for it . . . speaks with the vocabulary of the great vital
constants, sexual instinct, feeling of death, physical notion of the

enigma of space these vital constants are universally echoed in every
human. To understand an aesthetic picture, training in appreciation
is necessary, cultural and intellectual preparation. For surrealism the

only requisite is a receptive and intuitive human being.^^

The mysteries before which man is dumb, the questions


which he voices and hazards answers to, are those of him-
self. The marvelous is within him and speaks to him
through the vital constants.
Dali adds another kind of surrealist activity to those
already mentioned, " paranoiac-critical activity." The
images of reality are susceptible of false interpretations in
terms of some mental delusion. The paranoiac does not
deny that the object which others believe to be his hat is
his hat to them; but he does deny it is such to him. For
him it is an object which has entirely different relations
to his mind because of the shifted viewpoint under which
he sees it. For him it is really something else. Similarly,
a picture of a horse may be seen as that of a woman also,
or even further as a lion. Which of these it may be and
how many images an individual may see depend upon his
degree of paranoiac capacity. The images of reality de-
pend upon this capacity as well, for reality may be as
easily dissociated and put in question as illusion. ^° As
Breton has said, one may look at a tomato and see a child's
balloon. Knowledge for one who does this is not knowl-
edge of things as they are in the world, but knowledge of
SURREALISM 87

things as viewed through one's individual desires. The


world of such a one is ego-centered, and however much
anyone may claim the objective and subjective are tele-
scoped, it is the subjective element which is in the as-
cendency. The rational brought into line with the
is

irrational, the world of common sense with that of illusion.


When a work of art is a paranoiac phenomenon it is no
longer an aesthetic object but is purely a psychiatrical
index of one's unconscious activity.
This turbulent refusal of the mind to accept the given,
ever substituting its own interpretations, has been called
by Gaston Bachelard surrationalism. This term indicates
the whole general state of mind in which the mind is sup-
posedly liberated from the restrictions of logic by its own
caprice. Surrationalism traces its origin to Lobatchewsky's
discovery of a non-Euclidean geometry. For centuries a
hardened rationalism had repeated the axioms and postu-
lates of Euclidean geometry, without seeing that they may
be freely rejected as they were freely accepted. Lobatchew-
sky dialecticized geometry and in so doing cast doubt on
all fundamental issues, challenging the human mind to rise

above them and free itself by its own imprudence.^^


Such are the ideas of surrealism, the climax of twentieth-
century anti-intellectualism. Its sources in fantastic art
and erotic literature have been often pointed out. One
could also mention the great wealth of fancy and imagery
one meets in English literature, in never-never land, the
looking-glass country, midsummernight dells, in the wealth
of children's literature of which French literature is almost
entirely devoid. But these are antecedents of the fanciful
only.
The theoretical backgrounds of surrealism are found in
the principle of dialectic which it inherited indirectly from
Hegel, and in psychoanalytic method which it drew
directly from Freud.
88 THE AESTHETIC THEORIES OF FRENCH ARTISTS

But more a word than a dialectic.


surrealism's dialectic is

It is not the dialectic of Hegel for reality and rationality


do not logically generate their antithesis. There is no
necessary movement from a given thesis to its opposite,
and given both of them there seems no reason for sup-
posing that there is possible any synthesis between them.
The synthesis is one in name only; it is only a new label
Though the rational and irrational, the mental
for reality.
and material are identified in surreality, all supposedly
aufgehoben by the magic of a name, all that we can find
there are the irrational, the marvelous, the illogical, the
unreal. Though the whole theoretical polemic of surrealism
debunks the intelligence and discursive reason, it fails to
distinguish on the other hand between surreality and the
irrational. If the surreal were a truly dialectical synthesis
it would not be coextensive with the antithesis from which

it is not dialecticizing
proceeded. Liquidating the rational
it. Furthermore, the surreal is always something which is
assumed to inhere in mental life and consequently cannot
be the bridge between the mental and actual worlds.
Surrationalism is not dialecticizing for it is nothing
more than a misinterpretation of a very rational process.
What is called dialecticizing geometry is the creation of
systems of geometry which are mutually contradictory,
by denying one system a postulate affirmed in another.
in
When Lobatchewskian geometry was thus " dialecticized "
from Euclidean, the latter system was not thereby ruined,
and from the two systems no synthesis results. There
simply remain two different geometries, corresponding
proposition for proposition. And when a third is estab-
lished, all three exist side by side. Thought is not neces-
sarily in any turbulent agression.
In examining surrealist activity the theoretical back-
ground is found entirely in psychological theory, for the
SURREALISM 89

activities are altogether psychical and mainly subconscious.


These activities may eventuate in material objects, pic-
poems, and the like, but the existence of these as
tures,
material objectsis always disregarded by the surrealists.

They are looked upon simply as signs of the mental life


being explored.
The immediate psychological background is Freud and
Breton admits his indebtedness to him. The Viennese
physician had specifically stated that the basis of our
psychic life is the unconscious. The unconscious is no less
real for not always being present. We know the uncon-
scious through the conscious just as we know the external
world through the senses, but in both cases the data sup-
plied by our means of knowledge are insufficient to dis-
close the full nature of their reality." Psychoanalytic
theory presents the idea which surrealism interprets dia-
lectically, that the unconscious includes the conscious. It
is only a short step to the conception of the unconscious
surreal as the container of the real state as well as the
dream state in which it is more directly revealed. This is

especially true since psychoanalysis based upon a belief


is

in an antagonism between the conscious and the uncon-


scious parts of the mind, the former repressing those desires
of the libido which are socially or morally reprehensible,
but which, nevertheless, escape from the unconscious
through sublimation. Surrealist poetry, painting, and ob-
jects frequently express either directly or symbolically
their sexual associations. Surrealist freedom in artistic ex-
pression is the removal of the censorship which prevents
the psychic unconscious from rising into consciousness.
Breton's recipes for the production of automatic texts
correspond in purpose to the analyst's desire to reject
conscious directing ideas so that unconscious ones may
determine our flow of thought.^^
90 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Freud was also important to the surrealists for suggest-


ing to them a theme which they further capitalized, that
ordinary lifein its errors and dreams is not so very differ-
ent, other than in degree, from the symptoms which one
meets in psychosis and neurosis.^* The relation of sur-
realism to madness is discussed by Breton,^^ and the intro-
duction and acceptance of the paranoiac-critical method is

based on the supposition that paranoia is prevalent in


everyone to some extent at least.
Finally, the general correlation which Freud makes be-
tween these symptoms in civilized life and in pagan and
primitive practices ^^ might suggest the supposition of the
surrealists that the unconscious surreal is not only the true
" psyche " of the individual but is a center conunon to

all men.
This theory of the surreal as the superindividual, how-
ever, is more closely akin to that of the " collective un-
conscious " put forth by C. G. Jung and the Zurich school
of psychoanalysis, a schism from the original movement.
The important characteristic of the unconscious for Jung
is that it contains the heritage of all races:

[Men] love their unconscious, that is, that remnant of ancient


humanity and the centuries-old past in all people, namely, the
common property left behind from all development. But in . . .

loving this inheritance they love that which is common to all. Thus
they turn back to the mother of humanity, that is to say, to the
spirit of the race, and regain in this way something of that connection
and of that mysterious and irresistable power which is imparted by
the feeling of belonging to the herd.^^

The surrealist distinction between rational thought and


irrational activity is found exactly in Dr. Jung's distinction
between " directed thinking " and " phantasy thinking."
The first of these forms of thinking is logical thinking in
which the images in our mind flow in a causal sequence
SURREALISM 91

corresponding to exterior historical events. This type of


thinking is directed to the outside world. Its peculiarity is

that one and hence can take place only for short
it tires

periods of time. Its material is language. In the second


kind of thinking the thoughts are not directed but rise and
fall by their own weight in a stream of associations. This

thinking does not tire us. In it images crowd upon images


and we leave the world of the present for the past, the
future, the would-be world of dreams. The source from
which phantasies dra*w their materials is the same as that
from which myth and folk-belief arose. The dream phan-
tasy is the last remnant of archaic thought, like a vestigial
organ of our physical evolution. It is for this reason that
phantasy thinking brings us back to the common mankind
of the past. The products of dream thought are 'day
dreams, the dreams of sleep, and the phantasies of those
with dissociated personalities.^^ The surrealist activity of
psychic automatism gives us the products of dreams; the
paranoiac-critical method is employed by those who have
a dissociated personality and can see woman, horse, and
lion in the same object. The surrealist theory of the de-
pendence of the rational upon an irrational foundation is

the parallel of Jung's contention that directed thinking is

connected with the very foundations of the human mind


through phantasy thinking. Phantasy thinking, which is
so largely unconscious (and Freud, remember, had in-
cluded consciousness in the unconscious) is the surrealist
activity which brings us down to the surreality, the collec-
tive unconscious. The surrealist believes that in eliminat-
ing from his thought and his art all rational controls or
moral and aesthetic preoccupations he is eliminating those
elements which hide what is common to all as a race.
Intellectualism is an added structure to the common
minimum of subconscious life. By stripping it away we
92 THE AESTHETIC THEORIES OF FRENCH ARTISTS

get down to the surreality where the vital constants


wallow. The value of an exquisite corpse^® is its importance
as tangible evidence of the possibility that several persons
together may explore the world of the irrational. By
reducing the unconscious to the overindividual both Jung
and the surrealists believe they are escaping the predica-
ment of the mental giving us only a private world.
Now when the surrealist says that unless one can see the
tomato as a balloon he is a cretin, or that one's paranoiac
capacity accounts for the number of multiple images he
may find in a picture, he has not removed the balloon or
the image from the private world of the spectator, and he
has just as much, on the other hand, left the tomato and
picture as public property. The tomato has become a
balloon for the individual who sees it as such, and that
individual is not concerned whether the object is a tomato
or a balloon to anyone else. It is a balloon in his imagina-
tion. The rational world is denied as far as he is concerned,
and his own mental constructions are put in its place. The
surrealist artist may conceive of his tomato as a balloon; a
spectator may conceive the tomato as something very
different. Each is within his own imaginative world and
the two are independent of each other. If the unconscious
were the source of what is common to mankind, and if

surrealist activity explores that unconscious, it is difficult

to understand how so individual an activity as that by


which a tomato becomes a balloon to one who looks at it
can open up to him the overindividual depth of his nature.
If a spectator ever feels any kinship with others before a
work of art it is insofar as he can rationally interpret that
work. Dali's Persistence of Memory, for example, is capa-
ble of such interpretation. The lifeless watches devoured
by ants and the scene without horizon depict memory as
the death of time. If a work were not open to such in-
SURREALISM 93

terpretation it would be hard to find in it any point of


contact open to all minds.
Surrealist activity recognizes only the free play of the
imagination. In this the artist becomes a passive instru-
ment. This notion, too, of course, is not new with them.
Plato noted the passivity of the artist in his Ion where he
speaks of the poet as the inspired instrument of the gods,
mysteriously directed by them in his work. In nineteenth-
century France Delacroix voiced the belief in the artist as
the servant of his art which inspires him and uses him for
her own purpose.
An anti-rationalism which exalts the imagination is also
not new in philosophy. Coleridge's distinction between
fancy and the imagination is said to have corresponded to
and proceeded from Hartley's descriptions of the differ-
ence between delirium and madness. ^° Fichte's belief that
the imagination is the finite power which corresponds to
the creative act of the Absolute makes the world a product
of voluntaristic action. Among the other German roman-
tics there is some similarity to surrealism's conception of
imagination. For Novalis the philosopher becomes identi-
fied with the poet, and the source of poetic inspiration is
the life of dreams. In Schelling one finds the combination
of romantic idealism and dialectical struggle in nature
which occurs in surrealism.
The aesthetician and the art critic, to say nothing of
the moralist and the philosopher, are presented with a
problem by this strange and heterogeneous theory which
comes into their field with its destructive doctrine and
perplexing works of art. They may scoff at it and deny its
entrance into their world, turning it over to the psychi-
atrist as a case of mental abnormality, or they may accept
its artistic activity, its works of art,and its theory as
phenomena to be absorbed within a more general aesthetic.
94 THE AESTHETIC THEORIES OF FRENCH ARTISTS

It seems that the important place which surreahsm has


occupied in the world for the last two decades and the
influence which it has had on traditional forms of art and
on many artists who do not accept its principles, justify

its consideration at least as a special case within a general


theory of aesthetics and philosophy of art. It is the busi-
ness of the aesthetician to systematize or explain as far as
he can the material which is presented to him by artistic

activityand not to prejudge this material on principles


drawn up prior to the intrusion of the new material. Wild
data must be admitted in aesthetics as much as in episte-
mology. If it does nothing else for the aesthetician sur-
realism should make him look to the grounds of his science.
Surrealism has gone back to a pre-positivistic romanti-
cism for its theory and to a psychiatric theory which
reverts to a new mythology to explain the human mind.
With it our study has come full circle to the denial of the
naivete of realism by the substitution of a new fiction.

NOTES
^ From " Manifeste dada 1918 " by Tristan Tzara, in Tzara, Sept
manifestes dada. p. 18.
*
A. Breton, Manifeste du surrealisme, p. 21.
' Ihid., pp. 21-22.

^Ihid.,Y>V- 24-26.
' Ihid., pp. 27-28.
'A. Breton, "Exhibition X..Y...," What is Surrealism? tr. D.
Gascoyne, p. 25.
"^
A. Breton, Manifeste, p. 46.
^ A. Breton, Second Manifeste du surrealisme, pp. 9-10.
'Ibid., pp. 51-52.
"/6z(Z., pp. 54-55.
^^
Loc. cit.
^'
A. Breton, Le surrealisme et la peinture, pp. 10-11.
^'Ibid., pp. 12-13.
SURREALISM 95

**
A. Breton, " Beauty Will be Convulsive." What is Surrealism?
p. 37.
^^L. Aragon, La peinture au defi. Collage is the method of con-
structing a picture by pasting pieces of various materials, glass, paper,
wood, etc., on a background.
^*
Dictionnaire abrege du surrealisme, p. 7.
" M. Ernst, " Inspiration to Order," This QuaHer, Sept., 1932, pp.
79-80.
^^
S. Dali, The Conquest of the Irrational, pp. 12-13.
"S. DaU, Address dehvered at M. M. A., 1934, from J. Levy,
Surrealism, p. 7.

Dali, The conquest of the Irrational, pp. 17-18; also vide
S.

"The Stinking Ass," This Quarter, Sept., 1932, pp. 49-51. Paranoia
is a chronic form of mental disease in which the sufferer experiences
delusions of persecution or grandeur. He interprets falsely the atti-
tude of others toward him or accounts for the outer world in terms
of his delusions. His delusion is a systematic, rational explanation,
but isformed from false convictions or beliefs.
^^
Cf. G. Bachelard, " Surrationalism," in J. Levy, Surrealism, pp.
186-189.
Cf. S. Freud, The Interpretation of Dreams, tr. A. A. Brill in
'^^

The Basic Writings of Sigmund Freud, p. 542.


" Cf ibid., p. 482.
.

^*
Cf. S. Freud, Psychopathology in Everyday Life, in op. cit.
" A. Breton, " Surrealism and Madness," This Quarter, Sept., 1932.
However, it is interesting to note that drawings by surrealists and
those by the insane are quite different, the former retaining a trace
of conventionality, artistically speaking, the latter being more infan-
tile. Cf. H. Prinzhom, Bildnerei der Geisteskranken.
^"
S. Freud, Totem and Taboo, in op. cit.
*' Reprinted by permission of Dodd, Mead and Co., Inc. from C.

G. Jung, Psychology of the Unconscious, tr. by B. Hinkle, pp. 200-201.


'^Ibid., pp. 12-21 and 32-37.
^*
Exquisite corpse, a composite picture drawn by several people
one after the other, each one continuing where the former one has left
off but with the previous part of the picture concealed from him.

Hence the irrationality of the outcome and the surrealist's belief the
outcome shows the common unconscious substratum among the
creators.
""
Cf. H. S. Davies, " SurreaHsm at This Time and Place," in H.
Read, Surrealism, particularly pp. 138-148.
CHAPTER EIGHT

CODA
STARTING from realism it was possible to trace the
various theories of art as mutations or denials of
realist suppositions, and to see how certain crucial ques-
tions came into the aesthetics of painting which broadened
and deepened that subject.
The ideas of realism were three: 1) beauty is a quality
in objects, 2) the labor of the painter is exa-ct copying of
his model, 3) the work of art is an imitation of its model.

As time progressed the first was discarded because it was


seen to be in contradiction with the positivistic notion
that metaphysics must be scrapped, and beauty was too
loaded with the tradition of a metaphysical quality. It
also lost out as denotative of aesthetic quality, for the
tendency was more and more to locate aesthetic quality in
the work of art as a character peculiar to it, something
which the painter puts into his subject when he paints it
which was not in the object. This latter notion arose when
the painter saw that his labor was not merely copying, for
in reproduction he did not always find so easily a beauty
which made his work worthwhile. Hence, he was led to
doubt the second idea of realism and to reexamine the
question of what artistry is. Here two paths were open to
him. Either he could continue in the way set by realism,
following the ideal of science but accepting more modern
qualifications of scientific methodology and giving up the
early positivistic notion of science as mere description. Or
he could take an entirely new tack and claim that artistry
is the rendition of his own sensibilities and feelings or of

his private mental life. In either case he is brought to

96
CODA 97

doubt the third idea of reaUsm. Here the crucial questions


enter.
Those following the scientific ideal could ask: if painting
is the rendering of what one perceives, what is the object
as seen by the artist and is its nature affected by his per-
ceptual activity? In brief the painter is faced with episte-
mological questions. And from
came a second ques-
this
tion: how much does the artist's manner of rendering his
model demand a reorganization of that model to comply
with the form in which it is expressed.? If the model is
luminous surface, as for the impressionists or neo-impres-
sionists, light is what he must record and on the canvas
the model becomes a shimmer of divided strokes of pig-
ment. Already we are started on the road to abstraction
and the autonomy of art. In the theory of Cezanne the
whole problem was how to render his sensations. In solv-
ing this the artist recognized that the painter's experiences
in the natural world are distinctive and determined by his
artistic activity, just as the scientist's are determined by
his activity. Cubism distinguished a natural space and a
pictorial space, the second governing the renditions of the
artist.

Those turning away from the scientific ideal could ask:


if the object seen is not the real model what is the subject
of a painting.? Or also, what does my painting tell me
about the new subject I render.? Renoir turned from the
scientific ideal but never faced these questions. The sym-
bolists made their subject the feelings of the artist and
weakened the reference to the natural object. The sur-
realists reintroduced a traditional subject matter after the
pure abstractionists had about gotten rid of it, but their
subject matter was only a referent to an inner model.
With these painters the problems were those of psychology.
The first group recognized the weakness of imitation
98 THE AESTHETIC THEORIES OF FRENCH ARTISTS

as a theory of art and sought to quahfy it. The second


group moved more wholeheartedly in the direction of
expressionism.
Each aesthetic has a philosophical background that it

reflects. When Courbet says beauty is in an object he is

voicing the popularly accepted notion of the old realist


theory in philosophy that qualities are inherent in objects.
On the other hand, when he says it is the business of the
artist to copy the object exactly he is reflecting the new
positivistic point of view of his age that description is the
sum of what we can understand about things. In impres-
sionism's preoccupation with surface appearance one sees
a definite turning away from any attempt to paint an
essence or a typical example; the attitude of the painter
has become completely non-metaphysical. The attempts
of the neo-impressionists to lay down scientific schemes by
which the painter may attain the harmony of formal
elements are also perfect examples of positivistic explana-
tion through description. Renoir, the symbolists, and the
fauves find the artistic power in intuition, rebelling against
any theories which reduce this faculty to intellect. This
tendency among painters is a kind of Bergsonism. The
distinction between the intellectual and the intuitive is
fundamental to the philosophies of Bergson and Croce who
both place the artistic on the intuitive side. The new
positivism of Poincare and the new rationalism of Meyer-
son find their counterpart in the theories of Cezanne and
cubism. The knowing mind exerts its own influence in the
process which begins in the artist's sensation of the natural
world and terminates with his rendition of part of that
world in a work of art. Cezanne's emphasis that the vision
of the artist before nature is not the vision of the scientist
or of the average man hints at a viewpoint that was later
to come into great prominence in psychology and episte-
CODA 99

mology. Cubism, similar to new realism in philosophy and


the new scientific thought, reduces the known object to a
rationally conceived structure, the result of a mental con-
struction. Finally, surrealism brings in the suppositions of
Freudian psychology.
The problems of the painter are the problems of the
philosopher, the enigmas of our knowledge of the external
world.
BIBLIOGRAPHY
All titles mentioned in the text and works consulted are listed.

The edition or translation given is the one used.

Alberti, Leon Battista. Of Painting. Translated into Italian by


Cosimo Bartoli, now into English by James Leoni. London,
1726.
ApoUinaire, Guillaume. Les peintres cubistes. Paris: Figuiere, 1913.
Aragon, Louis. La peinture au defi. Paris: Jose Corti, 1930.
Aurier, Albert. " Le symbolisme en peinture." Mercure de France.
II: 155-165. 1891.
Barnes, Albert C. and Violette de Mazia. The Art of Cezanne. New
York: Harcourt, Brace, 1939.
Barr, A. H., Jr. Cubism and Abstract Art. New York: Museum of
Modem Art, 1931.
. Fantastic Art, Dada, Surrealism. New York: Museum of
Modern Art, 1936.
. Henri Matisse Retrospective Exhibition. New York:
Museum of Modern Art, 1931.
Baudelaire, Charles. Oeuvres completes. Vols. II and III. Paris:
C. Levy, n. d.
. Varietes critiques. Paris: Cres, 1924.
Benrubi, Isaac. Contemporary in France. New York:
Thought
Knopf, 1926.
. Les sources et les courants de la philosophic contemporaine
en France. Vol. I. Paris: Alcan, 1933.
Berenson, Bernard. Florentine Painters of the Renaissance. New
York: Putnam's, 1901.
Bergson, Henri. L'evolution creatrice. Paris: Alcan, 1908.
. New York: Putnam's, 1912.
Introduction to Metaphysics.
Boas, George (ed.) Courbet and the Naturalistic Movement. Balti-
.

more: The Johns Hopkins University, 1938.


. French Philosophies of the Romantic Period. Baltimore:
The Johns Hopkins University, 1925.
Bosanquet, Bernard. History of Aesthetic. London: Macmillan, 1892.
Bougot, Auguste. Essai sur la critique d'art. Paris: Hachette, 1877.
Bouvier, Emile. La bataille realiste. Paris: Fontemoing, 1914.
Brehier, fimile. Histoire de la philosophic. Vol. 11, Parts 3 and 4.

Paris: Alcan, 1932.


100
BIBLIOGRAPHY 101

Breton, Andre (ed.) . Le surrealisme au service de la revolution.


May 15, 1933.
. Le surrealisme et la peinture. Paris: Gallemard, 1928.
. Manifeste du surrealisme, poisson soluble. Paris: Editions
Kra, 1924.
. Position politique du surrealisme. Paris: ]feditions du Sagit-
taire, 1935.
. Second manifeste du surrealisme. Paris: Editions Kra, 1930.
. " Surrealism: Yesterday, Today and Tomorrow." Translated

by ?. This Quarter. V: 7-44. September, 1932.


. What is Surrealism? Six essays translated by David Gas-
coyne (Criterion Miscellany no. 43) London: Faber & Faber, .

1936.
Brett, George Sidney. A History of Psychology. Vol. HI. New York:
Allen & Unwin, 1921.
Bulletin of the Museum of Modem Art. TV. November-December,
1936.
Cassagne, Albert. La theorie de Vart pour Vart en France. Paris:
Hachette, 1906.
Cennini, Cennino. Le Livre de Vart. Translated by Victor Mottez.
Paris: J. Watelin, n. d.
Cezanne, Paul. Correspondance. Edited by John Rewald. Paris:
Grasset, 1937.
Champfleury. Le realisme. Paris: Michel Levy, 1857.
Cheney, Sheldon. The Story of Modem Art. New York: Viking,
1941.
Cherfils, M. L'esthetique positiviste. Paris: A. Messien, 1909.
Cherniss, Ruth Meyer. A Precursor of Nineteenth Century French
Realism: Champfleury. Abstract of thesis for Ph.D. Cornell,
1933.
Chevreul, Michel-Eugene. De la loi du contraste simultane des
couleurs. Paris: Pitois-Levrault, 1839.
. Des couleurs et de leurs applications aux arts indu^triels.
Paris: BailHere, 1864.
Comte, x\uguste. Cours de philosophic positive. Vol. V. Paris: Bail-
liere, 1864.
Coquiot, Gustave. Cubistes, futuristes, passeistes. Paris: Ollendorff,
1923.
. Des peintres maudits. Paris: Delpeuch, 1924.
. Les independents, 1884-1920. Paris: Ollendorff, 1920.
102 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Courbet, Gustave. " Manifestes de 1855 et de 1861." In Charles


Leger. Courbet. Paris: Cres, 1929.
Cousturier, Lucie. Seurat. Paris: Cres, 1926.
Couture, Thomas. Methode et entretiens d'atelier. Translated as
Conversations on Art Methods by S. E. Stewart. New York:
Putnam *s, 1879.
Croce, Benedetto. Aesthetic as a Science of Expression and General
Linguistic. Translated by Douglas Ainslie. London: Mac-
millan, 1909.
Dali, Salvador. Conquest of the Irrational. New York: J. Levy, 1935.
. " The Object as Revealed in Surrealist Experiment." This
Quarter. V: 197-207. September, 1932.
. " The Stinking Ass." This Quarter. V: 49-54. September,
1932.
Dampier-Whetham, WiUiam Cecil. History of Science. Cambridge
University, 1929.
David-Sauvageot, A. Le realisme et le naturalisme dans la litterature
et dans I'art. Paris: C. Levy, 1889.
Deffoux, Leon. Le naturalisme: oeuvres representatives. Paris: Les
Oeuvres Representatives, 1909.
Degas, Hilaire, Lettres. Edited by Marcel Guerin. Paris: Grasset,
1931.
Delacroix, Eugene, Journal. 3 vols. Paris: Plot-Nourrit, 1926.
. " Questions sur le beau." Revue des Deux Mondes, July 15,
1854.
Denis, Maurice. De Gauguin et de Van Gogh au classicisme, theories:
1890-1900. Paris: Watelin, 1913.
. Nouvelles theories. Paris: Watelin, 1922.
Deschamps, Emile. Preface des etudes frangaises et etrangeres. 4*

ed. Paris: Levasseur et Canel, 1829.


Dictionnaire abrege du surrealisme. Paris: Galerie Beaux-Arts, 1938.
Ducasse, Curt J. The Philosophy of Art. New York: Dial Press,
1929.
Dumesnil, Rene. Le realisme. Paris: Gigord, 1936.
Duret, Theodore. Manet and the French Impressionists. Translated
by J. E. C. Flitch. Philadelphia: Lippincott, 1910.
Ernst, Max. " Inspiration to Order." This Quarter. V: 79-84. Sep-
tember, 1932.
Feneon, Felix. Les impressionistes en 1886. Paris: Publications de La
Vogue, 1886.
BIBLIOGRAPHY 103

Ferran, Andre. L'esthetique de Baudelaire. Paris: Hachette, 1933.


Feidler, Konrad. Schrijten uher Kunst. 2 vols. Munich: Piper, 1913-
1914.
Focillon, Henri. La peinture aux XIX^ et XX^ siecles. Paris: Libraire
Renouard, 1928.
Fontaines, Andre. Histoire de la peinture jrangaise. Paris: Mercure
de France, 1922.
Fouillee, Alfred. Le mouvement positiviste. Paris: Alcan, 1896.
Freud, Sigmund. The Basic Writings of Sigmund Freud. Edited and
translated by A. A. Brill. New York: The Modem Library,
1938.
Fuller, B. A. G. History of Philosophy. New York: Holt, 1937.
Gascoyne, David. A Short Survey of Surrealism. London: Cobden
Sanderson, 1935.
Gauguin, Paul. Avant et apres. Paris: Cres, 1923.
Diverses choses, 1896-1897. MS of M. Daniel de Monfreid.
.

. Letters a Georges-Daniel de Monfried. Paris: Cres, 1918.


. Noa-noa. With Charles Morice. Paris: Cres, 1925.
. Racontars d'un rapin. 1902.
GeofProy, Gustave. Claude Monet. Paris: Cres, 1922.
Giedion, Sigfried. Space, Time, and Architecture. Cambridge: Har-
vard University, 1941.
Gilbert, Katharine and Helmut Kuhn. A History of Aesthetics. New
York: Macmillan, 1939.
Gleizes, Albert and Jean Metzinger. Du cubisme. Paris: Figuiere,
1912.
Goldwater, Robert J. Primitivism in Modem Painting. New York:
Harper's, 1938.
Guerard, A. L. French Civilization of the Nineteenth Century. New
York: Century, 1914.
Gunn, J. Alexander. Modern French Philosophy. London: T. Fisher
Unwin, 1922.
New York:
Hendricks, Ives. Facts and Theories of Psychoanalysis.
Knopf, 1939.
Hammond, William A. A Bibliography of Aesthetics and of the
Philosophy of Art from 1900 to 1932. New York: Longmans,
Green, 1934.
Hevner, Kate. " Experimental Studies of the Affective Value of
Colors and Lines." Journal of Applied Psychology. XIX:
385-398. August, 1935.
104 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Hoffding, Harald. A History of Philosophy. Vol. II, translated by


B. E. Meyer. London: Macmillan, 1900.
Hogarth, William. The Analysis of Beauty. Pittsfield, Mass., 1909.
Hugnet, Georges. " L'esprit dada dans la peinture." Cahiers d'art
VII & IX. 1932 and 1934.
Huyghe, Rene. Les conterrvporains. Paris: Tisne, 1939.
Jenkins, Iredell. " Hippolyte Taine and the Background of Modem
Aesthetics." Modern Schoolman. XX.
Jung, C. G. Psychology of the Unconscious. Translated by Beatrice
Hinkle. New York: Dodd, Mead and Co., 1916.
Koffka, K. Principles of Gestalt Psychology. New York: Harcourt,
Brace, 1915.
Konig, Rene. Die naturalistische dsthetik in Frankreich und und ihre
Auflosung. Borna-Leipzig: R. Noske, 1931.
Larroumet, Gustave. " L'art realiste et la critique." Revue des Deux
Mondes. December 15, 1892 and March 1, 1893.
Laver, James. French Painting in the Nineteenth Century. London:
Botsford, 1937.
Lecomte, Georges. Camille Pissarro. Paris: Bernheim-Jeune, 1922.
Leger, Charles. Courbet. Paris: Cres, 1929.
. Courbet selon les caricatures et les images. Paris: Rosenberg,
1920.
Lemaitre, Georges. Cubism and Surrealism in French Literature.
Cambridge: Harvard University, 1941.
Lenoir, P. Histoire du realisme. Paris: Quantin, 1889.
Leonardo da Vinci. Literary Work. 2 vols. Compiled and edited by
J. P. Richter and I. A. Richter. London: Oxford, 1939.

Levy, Julien. Surrealism. New York: Black Sun, 1936.


Levy-Bruhl, Lucien. History of Modem Philosophy in France. Trans-
lated by G. Coblence. Chicago: Open Court, 1899.
.The Philosophy of Auguste Comte. Translated by Kathleen
de Beaumont-Klein. London: Sonnenschein, 1903.
Lipps, Theodor. Zur Einfiihlung. Leipzig: Engelmann, 1913.
Listowel, The Earl of. A Critical History of Modem Aesthetics.
London: Allen & Unwin, 1933.
Marriott, Charles. A Key to Modem Painting. London: Blackie,
1938.
Martineau, Harriett. Comte s Positive Philosophy. Vol. m. Lon-
don: G. Bell & Sons, 1896.
Martino, Pierre. Pamasse et symbolisme. Paris: Colin, 1935.
BIBLIOGRAPHY 105

Matisse, Henri. " Notes d'un Peintre." Grande Revue. December


25, 1908.
. Interview to Guillaume AppoUinaire. La Phalange. Decem-
ber 15, 1907.
. Interview to E. Teriade. L'intransigeant. January, 1929.
Mead, George. Movements of Thought in the Nineteenth Century.
Chicago: University of Chicago, 1936.
Merz, John T. History of European Thought in the Nineteenth
Century. 4 vols. Edinburgh and London: W. Blackwood &
Sons, 1903-04.
Meyerson, Emile. Identite et realite. Paris: Alcan, 1912.
Michel, Andre. Histoire d'art. Vol. VIII, part 2. Paris: A. Colin,
1926.
Mondrian, Piet. Neo-plasticisme. Paris: Effort Moderne, 1920.
Moreau-Nelaton, Etienne. Manet raconte par lui-meme. 2 vols.
Paris: Laurens, 1926.
Morice, Charles. Paul Gauguin. Paris: Floury, 1920.
Mustoxidi, T. M. Historie de Vesthetique frangaise 1700-1900. Paris:
Champion, 1920.
Needham, H. A. Le developpement de Vesthetique sociologique.
Paris: Champion, 1925.
Ozenfant, Amadee, and Charles-Edouard Jeanneret. Apres le cubisme.
Paris: Editions des Commentaires, 1918.
Ozenfant, Amadee, and Charles-Edouard Jeanneret. La peinture
moderne. Paris: Povolsky, c. 1920.
Pach, Walter. " Pierre-August Renoir." Scribner's Magazine. May,
1912.
Peirce, Charles Sanders. Collected Papers, Vol. II. Cambridge:
Harvard University, 1932.
Plato. Dialogues. Translated by B. Jowett. Oxford University, 1892.
Poincare, Henri. Science and Hypothesis. Translated by George
Bruce. New York: Science Press, 1905.
Praz, Mario. The Romantic Agony. Translated by Davidson Angus.
Oxford University, 1933.
Prinzhorn, Hans. Bilderei der Geisteskranken. Berlin: J. Springer,
1923.
Proudhon, P. J. Du principe de I'art. Paris: Gamier Freres, 1865.
Quintanilla, Luis. Bergsonisme et politique. Ph. D. thesis. The Johns
Hopkins University, 1938.
Rader, Melvin. A Modem Book of Aesthetics. New York: Holt, 1933.
106 THE AESTHETIC THEORIES OF FRENCH ARTISTS

Raphael, Max. Proudhon, Marx, Picasso. Paris: Editions Excelsior,


1933.
Raymond, Marcel. De Baudelaire au Surrealisme. Paris: Correa,
1933.
Raynal, Maurice. Modem French Painters. New York: Brentano,
1928.
Read, Herbert (ed.) Surrealism. New York: Harcourt, Brace, 1937.
.

Redon, Odilon. Memoires a soi-meme. Paris: Floury, 1922.


Reid, Louis A. A Study in Aesthetics. London: Allen & Unwin, 1931.
Renoir, Auguste. " La Societe des Irregularis tes." In Venturi, Les
Archives de rimpressionnisme. Paris: Durand-Ruel, 1939.
Lettre a Henry Mottez. In Cennini, Livre de I'art.
.

Rewald, John. Gauguin. Paris: Hyperion, 1938.


Seurat. New York: Wittenbom, 1943.
.

The History o] Impressionism. New York: Simon & Schuster,


.

1946.
Rey, Robert. La renaissance du sentiment classique. Paris: Beaux
Arts, 1921.
Reynolds, Joshua. Discourses on Art. London: J. Carpenter, 1842.
Rhodes, Solomon. The Cult of Beauty in Charles Baudelaire. 2 vols.
New York: Columbia University, 1929.
Ribot, Theodule. Essai sur I'imagination creatrice. Paris: Alcan, 1921.
Rich, Daniel C. Seurat and the Evolution of " La Grande Jatte.**

Chicago: University of Chicago, 1935.


Rood, Ogden Nicholas. Modern Chromatics with Applications to Art
and Industry. New York: Appleton, 1879.
Rosenthal, Leon. Du romantisme au realisme. Paris: Laurens, 1914.
Rothschild, E. F. Meaning of Unintelligibility in Modem Art.
Chicago: University of Chicago, 1934.
Rotonchamp, Jean de.Paul Gauguin 1848-1903. Paris: Cres, 1925.
Russell, Bertrand. Our Knowledge of the External World. London:
Open Court, 1914.
Salon de 1863. Paris: Morgues, 1863.
Santayana, George. The Sense of Beauty. New York: Scribners,
1896.
Schinz, Albert. " Dada, poignee de documents." Smith College
Studies in Modern Languages. October, 1923.
Serouya, Henri. Initiation a la peinture d'aujourd'hui. Paris: Renais-
sance du livre, n. d.

Signac, Paul, D' Eugene Delacroix au neo-impressionnisme, 3« ed.


Paris: Floury, 1921.
BIBLIOGRAPHY 107

. " Le neo-impressionnisme: documents." Gazette des Beaux


Arts. Per. 6, XI: 49-59. January, 1934.
Silvestre, Theophile. Artistes frangais. 2 vols. Paris: Cres, 1926.
Stein, Leo. The A-B-C of Aesthetics. New York: Boni & Liveright,
1927.
Sweeney, James J. Plastic Redirections in the Twentieth Century.
Chicago. University of Chicago, 1934.
Taine, Hippolyte. Philosophy of Art. Translated by John Durand.
New York: Holt, 1875.
Topass, Jan. La pensee en revolte. Brussels: Henriquez, 1935.
Tzara, Tristan. Sept. manifestes dada. Paris: Budry, 1921.
Urban, Wilber. Language and Reality. London: Allen & Unwin, 1939.
Van Gogh, Vincent. Letters to His Brother. 3 vols. London: Con-
stable, 1927-1929.
. Letters to Simile Bernard. Translated by Douglas Lord.
London: Cressit, 1938.
Venturi, Lionello. Cezanne. 2 vols. Paris: Rosenberg, 1936.
. History of Art Criticism. Translated by Charles Marriott.
New York: Dutton, 1936.
. Les archives de Vimpressionnisme. 2 vols. Paris: Durand-
Ruel, 1939.
Veron, Eugene. L'esthetique. Paris: C. Reinwald, 1878.
Vischer, Robert. Aesthetik. 6 vols. Munich: Meyer & Jens, 1922-
1923.
Vlaminck, Maurice de. Polement. Paris: Stock, 1931.
VoUard, Ambrose. Renoir, an Intimate Record. Translated by Ran-
dolph T. Weaver. New York: Knopf, 1925.
Whistler, James M. The Gentle Art of Making Enemies. New York:
Putnam's, n. d.
Wilenski, Reginald. French Painting. Boston: Hale, Cushman &
Flint, 1931.
. Modern French Painters. New York: Reynald and Hitch-
cock, 1940.
. The Modern Movement in Art. London: Faber and Guyer,
1928.
Windelband, Wilhelm. History of Philosophy. Translated by James
Tufts. New York: Macmillan, 1901.
Zimmermann, Robert. Aesthetik. Vienna: Braumiiller, 1858-1865.
Zola, fimile. Mes haines. In Oeuvres completes. Vol. XL. Paris:
Bernouard, 1927-1929.
INDEX
Aesthetic value 29, SI Braque, G. 69
Alberti, L. B. 3 Breton, A. 79-84
Analysis of Beauty (Hogarth) S
Anti-intellectualism 61 Cabaret Voltaire 79
and surrealism 87 Cafe Guerbois 19
see Instinct, Intuition Cafe Volpini 53
Anti-rationalism in philosophy 93 Castagnary, J. 13
Apollinaire, G. 69, 79 Cennini, C. 39, 51 note 5
Arago, E. 13 Cezanne, P. 3, 7, 43-51, 69, 70, 97, 98
Aragon, L. 84 Chahut (Seurat) 25 ,

Architecture 35-36, 38 Champfleury, H. 9


Aristotle 4 Chevreul, M. 24, 33 note 5
Art Children's literature and surrealism 87
and craftsmanship 37-41 Christophe, J. 25
and feeling 56-57, 61-62, 64 Coleridge, S. 93
and idealism 11 Collage 85
and instinct see Instinct Color 19, 21, 24-27, 29-30, 33, 45-47,
and nature 19-20, 22, 28-29, 33, 35, 56, 61, 75-76
44-47, 57-58, 62, 70, 73-76, 96, 98 Comte, A. 14
and philosophy 4-8, 11-18, 42-43, Constructivism 69, 77 note 2
47-49, 56, 64, 66-67, 76-77, 87- Corot, J. 55
88, 98-99 C orrespondance (Cezanne) 43-46
and psychology 21, 31, 48-49, 65- Courbet, G. 6, 7, Ch. 2, 19, 20, 26,
66, 87-93, 96. 97 33, 41, 70, 98
and religion 40 Courrier du Dimanche 9
and science 3-4, 11, 14, 22-23, 31, Croce, B. 64, 98
96, 97 Cubism Ch. 6, 97, 98
as autonomous 32 Cubists 7, 49
as expression 7, 61-62, 98 Cuvier, G. 54
as harmony 25, 28, 29-31
as imitation see Imitation Dada 79-80, 81
as intuition see Intuition Dali, S. 85-87, 92
as symbolic Ch. 5 David, J. 55
social purpose of 12-16 Decamp, A. 13
Automatic texts 83 Definition de neo-traditionnisme
Autonomy of art 32 (Denis) 57-60
Avant et apres (Gauguin) 56 Delacroix, E. 3, 16, 27, 28, 55, 93
Denis, M. 53, 57-60, 64, 67 note 10
Bachelard, G. 87 Deschamps, E. 13
Baudelaire, C. 16 Dessieux, L. 13
Beaubourg, M. 25 Detaille, J. 58
Beauty 10, 12, 16-18, 26, 28, 32, 35, D' Eugene Delacroix au neo-impres-
59, 96 sionnisme (Signac) 26-29
Beaux arts, tradition of 36, 38 Discourses on Art (Reynolds) 3
Berenson, B. 78 Divisionism 24, 25, 27-28
Bergson, H. 7, 42-43, 61, 64, 98 Dubois-Pillet 24
Berkeley, G. 23 Ducasse, C. 5-6
Bernard, E. 53 Du cubisme (Gleizes and Metzinger)
letters of Cezanne to 44-46 69-76
Blanc, C. 13 Du principe de Vart (Proudhon) 15-
Blondel. M. 67 16
Bouguereau, A. 58 Durand-Ruel 23
109
no INDEX
Duranty 20 L'art nouveau 38-39
Duret, T. 20 Laforgue, J. 20
L'J&co/e des Beaux Arts ^Q
Einfiihlung 65-66 Leonardo 3
Einstein, A. 50 Les mammelles de Tiresias (Apolli-
Empathy 65-66 naire) 79
Ernst, M. 85 Le surrealisme et la peinture (Bre-
Euclid 4 ton) 83-83
Expressionism 7, 61-62, 98 Light 20-23, 45-47
Exquisite corpse 92, 95 note 29 L'impressionniste (Riviere) 35
Line 29, 30, 33
Fauvism 7, 60-67, 98 Lipps, T. 66
Fechner, G. 66 Livre de l'art (Cennini) 39
Feneon, F. 24 Lobatchewsky, N. 87, 88
Fichte, J. 93 Locke, J. 23
Fiedler, K. 32 Logic and imagination 80-81
Form 45-47, 72-75 and symbolism 50
Fresnel, A. 21 Louvre 35
Freud, S. 81, 88, 89-90, 99
Functionalism 38 Machinism 40-41
Gauguin, P. 3, 53-57 Manet, E. 19-20, 70
GeofTroy, G. 20 Manifeste dada (Tzara) 80
Giotto 63 Manifeste du surrealisme (Breton)
Gleizes, A. 69-77 80-82
Gothic revival 38 Manifeste 18S5 (Coubert) 9
Guardi, F. 73 Manifeste 1861 (Courbet) 10-11
Manifestoes 3
Hartley, D. 93 use of 4-5
Haussmann, G. 18 note 10 value doubted 5
Hegel, G. 87-88 value upheld 6
Helmholtz, H. 21, 34 note 19 Matisse, H. 60-64
Hogarth, W. 3 Maus, O. 43
Maxwell, J. 24
Idealism 11, 26 Meaning 31
Identite et Realite (Meyerson) 71- of painting to Matisse 65
72 Metzinger, J. 69-77
Imagination, Coubet on 10, 14 Meyerson E. 71-72, 98
Imitation 12, 18, 22-23, 28, 43-46, Monet, C. 73
84, 96 Morris, W. 38
Impressionism 7, Ch. 3, 36, 44, 47, Music and painting 27, 33
55-56
Instinct 7, 41, 42-43, 61-62 Nabis 53, 67 note 10
Intuition 7, 41, 42-43, 62, 64 Naive attitude toward art 3-4
Ion (Plato) 93 Nature see Art and Nature
Irrational, the 85 Negro sculpture 69
Irregularity 36-38, 39, 41 Neo-impressionism 7, 23-33, 45-46, 65
Neo-plasticism 69, 77 note 2
James, W. 31 Neo-traditionalism 68
Jomrnal (Delacroix) 3, 28 Notebooks (Leonardo) 3
Journal (Redon) 3 Notes d'un peintre (Matisse) 60-64
Journal (Whistler) 3 Novalis 93
Jung, C. 90-92
Optic mixture 24, 27, 28
Laberthonniere, L. 67 Optics 21
Labrouste, H. 36 Orphism 78 note 3
INDEX 111

Painting defined by Courbet 10 Scientific aesthetic 26, 28


defined by Denis 57 Scientific technique 29
Painting and music 27, 33 Seascape 22
Paranoia 95 note 20 Seated Woman (Picasso) 78 note 29
Paranoiac-critical activity 86, 90 Second Empire, art world of 9, 11,
Peguy C. 67 16, 19
Persistence of Memory
(Dali) 92 Second manifeste (Breton) 82-83
Philosophy and art 4-8, 11-18, 42- Sensations 21, 45
43, 47-49, 56, 64, 66-67, 76-77, Serusier, P. 53
87-88, 98-99 Seurat, G. 23, 25-30, 32
Pictorial space 74 Signac, P. 26-29, 33, 58
Picturesque, the 38 Sisley, A. 23

Pissarro, C. 23-25, 28, 29 Snapshop art 4, 8, note 1


Plato 38, 93 Social purpose of art 12-16
Poincare, H. 50, 98 Spirit of the Dead Watching (Gau-
Point illism 27, see Divisionism guin) 54
Polytonality 50 Strindberg, A. 54
Positivism 7, 11-12, 14, 21, 26, 32, 47 Subconscious, the 81, 84
Pre-Raphaelites 55, 57 Surrationalism 87, 88
Primitivism 54-55 Surreal, the 81, 83, 99
Proudhon, P. 15-16 Surrealism 7, Ch. 7
Psychology 21, 31, 48-49, 65-66, 87- defined 82
93, 96, 97 Symbolic knowledge, 42
Pure visibility 32, 48 Symbolism 7, Ch. 5. 98

Purism 69, 77 note 2 Synthetism see Symbolism

Ravaisson, J. 7 Taine, H. 17, 32


Realism 8, Ch. 2, 39, 46, 47, 84, 96 languy, Pere 43
Realistic aesthetics 28 Third Republic, art world of 36
Redon, O. 3 Thore, T. 13
Religion and art 40 Tone 25-26, 30
Renoir, A. 7, 35-43, 47, 53, 64, 97 Treatise (Alberti) 3
Reynolds, 3 J. Trompe I'oeil 58-59
Riviere, G. 35 Turner, J. 73
Robert, H. 13 Tzara, T. 80
Romanticism 9, 94
Rood, N. 24, 28, 33 note 6 Van Gogh, V. 3
Rose Croix painters 53 Vauxcelles, L. 60
Ruskin J. 28 Veron, E. 17-18
Russell, B. 76 Vingt, the 43
Vischer, R. 65
Saint-Simon, C. 13-14 Visual space. Cubist definition of 74
Salon d'Automne (1903) 60
Salon des Independents (1911) 69
Whistler, J. 3, 55
Santayana, G. 66-67
AYhitehead, A. 76
Schelling, F. 56-57, 93
Science and art 3-4, 11, 14, 22-23,
41, 96, 97 Zimmermann, R. 32
and Cezanne 47, 50 Zola, E. 44
Science of art 32 definition of art 58
PHILOSOPHY, ART

Including the exponents of realism, impressionism,


neoim-
pressionism, eubism, surrealism, and such artists as Renoir,
Cezanne, Gauguin, and Matisse, the author presents
the aes-
thetic theories of the principal French artists
since 1855.
'The book is excellently organized and carefully indexed.
It is filled with well-translated quotations from the most im-
portant writings of leading artists of the period. It has a large
and excellent bibliography of items indispensable for anyone
who wishes to understand late nineteenth and twentieth century
dexelopments in art. For these reasons, if for no others, the
book will be of value to everyone interested in esthetics as
propounded by artists.' —College Art Journal
"Books on art such as Professor Gauss's are important because
they break down the barriers which specialization has built up
between the sciences."— Times LiteTary Supplement
'The double aim of this book seems a fascinating one: to
study the history of ideas concerning art from 1855 to the
present in the writing of French artists, and to indicate roots
or parallels for these ideas in the contemporary philosophies."—
Magazine oi Art

CHARLES EDWARD GAUSS was Elton Professor of Philos-


ophy George Washington University and assistant to the
at
president at Nova University of Advanced Technolog\-, Fort
Lauderdale, Florida. He is now retired.

THE JOHNS HOPKINS PRESS


Baltimore and London

ISBN 0-80 18-02 16-4

You might also like