Skip to main content

Full text of "On Art And Connoisseurship"

See other formats


701 F91o 60-0558^ 

Friedlander 

On art and connoisseur ship 





ON ART 
AND CONNOISSEURSHIP 

tr 
MAX J. FRIEDLANDER 



With 40 Illustrations 



Beacon Press Beacon Hill Boston 



Translated from the author's manuscript by 

TANCRED BORENIUS 
First published in 1942 by Bruno Cassirer, Ltd. 

First published as a Beacon Paperback in 1960 by permission 
of Bruno Cassirer, Ltd. 

-* J l ^ '" 
Printed in the 'United States of America 



CONTENTS 



PACE 



INTRODUCTION. By TANCRED BORENIUS 9 

PREFACE 1 3 

I. SEEING, PERCEIVING, PLEASURABLE CONTEMPLA- 
TION 19 

II. EXISTENCE, APPEARANCE, OBJECTIVE INTEREST 

IN THINGS 32 

III. ART AND SYMBOL 39 

IV, FORM, COLOUR, TONALITY, LIGHT, GOLD 43 
V. THE CONCEPT OF 'PICTORIAL* 3 

VI. SIZE AND SCALE, DISTANT VIEW AND NEAR VIEW $8 

VH. ON LINEAR PERSPECTIVE " 64 

Vffl. MOVEMENT 69 

DC. TRUTH TO NATURE, ARTISTIC VALUE AND STYLE 75- 

X. INDIVIDUALITY AND TYPE 84 

XI. ON BEAUTY 87 

XII. ON COMPOSITION 91 

XIII. ON THE PICTURE CATEGORIES 97 
XTV. RELIGIOUS AND SECULAR HISTORY IN PAINTING 100 

XV. THE NUDE 104 

XVI. GENRE PAINTING 108 

XVII. LANDSCAPE 113 

XVffl. PORTRAITURE 124 

XK. STILL LIFE W v-V.t I3 , 

S 



CONTENTS 



PAGE 



XX. THE ARTIST: GENIUS AND TALENT 1 34 

XXI, ART AND ERUDITION 143 

XXn. THE STANDPOINT OF THE SPECTATOR 1 55 

XXffl. ON THE VALUE OF THE DETERMINATION OF 

AUTHORSHIP 160 

XXIV. ON THE OBJECTIVE CRITERIA OF AUTHORSHIP 163 

XXV. ON INTUITION AND THE FIRST IMPRESSION 172 

XXVI. PROBLEMS OF CONNOISSEURSHIP 179 

XXVIt. THE ANALYTICAL EXAMINATION OF PICTURES 1 84 

XXVUI. ON THE USE OF PHOTOGRAPHY 197 

XXIX. ON PERSONALITY AND ITS DEVELOPMENT 200 

XXX. ON THE ANONYMOUS MASTERS, THE MEDIUM 

MASTERS AND THE LESSER MASTERS 213 

XXXI. THE STUDY OF DRAWINGS 218 

XXXII. INFLUENCE 222 

XXXHI. ARTISTIC QUALITY: ORIGINAL AND COPY 230 

XXXIV. DEDUCTIONS A POSTERIORI FROM COPIES 

REGARDING LOST ORIGINALS 246 

XXXV. WORKSHOP PRODUCTION 2^0 

XXXVI. ON FORGERIES 2^8 

XXXVH. ON RESTORATIONS 267 

XXXVni. ON ART LITERATURE 273 

INDEX 281 



LIST OF ILLUSTRATIONS 

HATE FACING PAGI 

i . HANS MEMIING. PORTRAIT OF A MAN IN AN ATTITUDE 

OF PRAYER. Lugano, Castle Rohoncz Collection frontispiece 
*2. MATTHIAS GRUNEWALD. THE CRUCIFIXION. Colmar 

Museum 24 

*3. PAUL CEZANNE. AUVERS-SUR-OISE 24- 

4. MASTER OF ALKMAAR. PANEL FROM THE SERIES OF THE 

WORKS OF MERCY. Amsterdam, Rijksmuseum 48 

$. DIRK BOUTS. THE LAST JUDGMENT (Detail). Lille, 

Museum 9 2 

*6. ADOLPH VON MENZEL. SCENE FROM THE LIFE OF 

FREDERICK THE GREAT (Woodcut) 93 

7. ALBRECHT DURER. A NUDE WOMAN (Drawing). Bay- 

onne, Museum 104 

8. ALBRECHT DURER. ADAM AND EVE (Drawing). New 

York, Morgan Library 105- 

9. JAN VAN EYCK. THE ROLLIN MADONNA (Detail). 

Paris, Louvre 1 1 2 

10. JOACHIM PATINIR. LANDSCAPE WITH THE RIVER OF 

DEATH. Madrid, Prado r 1 3 

11. LUCAS CRANACH. REST ON THE FLIGHT INTO EGYPT. 

Berlin Picture Gallery 1 1 6 
ii. WOLF HUBER. THE MONDSEE WITH THE SCHAFBERG 

(Drawing). Nuremberg, Germanisches Museum 1 1 7 

13. RUELAND FRUEAUF THE YOUNGER. PANEL FROM THE 

ALTARPIECE OF ST. LEOPOLD. Monastery of Kloster- 
neuburg 120 

14. LUCAS CRANACH. PORTRAIT OF JOHANN CUSPINIAN. 

Winterthur, Collection of Dr. O . Reinhart 1 2 1 

i j. LUCAS CRANACH. PORTRAIT OF THE WIFE OF JOHANN 
CUSPINIAN. Winterthur, Collection of Dr. O. Rein- 
hart 1 24 

1 6. LUCAS VAN LEYDEN. PORTRAIT OF THE ARTIST (Detail). 

Brunswick Museum 1 2 

17. HANS MEMLING. STILL LIFE, Lugano, Castle Rohoncz 

Collection 13 

1 8. MARINUS VAN REYMERSWAELE. ST. JEROME IN HIS 

STUDY. Madrid, Prado 1 3 i 



ILLUSTRATIONS 

PLATE FACING PAG* 

*i9. HUGO VAN DER GOES. ADORATION OF THE MAGI (*THE 

MONFORTE ALTARPIECE*). Berlin Picture Gallery 148 

*20. HUGO VAN DER GOES. ADORATION OF THE SHEPHERDS 

('THE PORTINARI ALTARPIECE'). Florence, Uffizi 192 

*2i. HANS HOLBEIN. MADONNA OF THE BURGO-J 

MASTER MEYER. Darmstadt, Grand Ducal Castle | BETWEEN 
*22. AFTER HANS HOLBEIN. MADONNA OF THE [PAGES 232-3 

BURGOMASTER MEYER. Dresden Gallery J 

23. JAN VAN EYCK. CANON VAN DE PAELE (Detail of the 

Altarpiece in the Bruges Museum) 233 

24. AFTER JAN VAN EYCK. CANON VAN DE PAELE. Hampton 

Court Palace. Copyright of H.M. The King 232 

25-. BRUGES MASTER OF 1499. MADONNA WITH DONORS. 

Paris, Louvre 244 

26. LUCAS CRANACH. PORTRAIT OF PHILIP OF POMERANIA. 

Rheims, Museum 244- 

27. WORKSHOP OF LUCAS CRANACH. PORTRAIT OF PHILIP 

OF POMERANIA 25-2 

28. HANS CRANACH. PORTRAIT OF A MAN. Lugano, Castle 

Rohoncz Collection 2^3 

29. STYLE OF THE ANTWERP MANNERISTS. WINGS OF AN 

ALTARPIECE. London Art Market 2 $6 

30. HANS HOLBEIN. PORTRAIT OF A MAN (Drawing). Royal 

Library, Windsor. Copyright of H. M. The King 2 57 

31. FORGERY. PORTRAIT OF A MAN 25-8 

32. HANS HOLBEIN. PORTRAIT OF (?) ANTOINE, DUKE OF 

LORRAINE. Berlin Picture Gallery 259 

33. HANS HOLBEIN. PORTRAIT OF A MAN. Vienna Picture 

Gallery 260 

34. FORGERY. PORTRAIT OF A MAN 261 
35-. FORGERY. PORTRAIT OF A MAN 262 

36. FORGERY. PORTRAIT OF A MAN 263 

37. FORGERY. NARRATIVE SUBJECT 268 

38. BRUGES MASTER c. 1480. MADONNA WITH TWO 

FEMALE SAINTS. Formerly London Art Market 269 

39. FORGERY. MADONNA WITH Two FEMALE SAINTS 272 
*4<>. REMBRANDT. MOSES SHOWING THE TABLES OF THE LAW 

TO THE PEOPLE. Berlin Picture Gallery 273 



INTRODUCTION 

AMONG art historians of to-day there is hardly any- 
one who enjoys a position comparable to that of 
Dr. Max J. Friedlander. He is universally recognized 
as being probably the greatest living expert, notably, 
of course, on the early Netherlandish and German 
masters; and in normal times not a day passed on 
which pictures were not submitted to him for opin- 
ion from all parts of the world. But he is much more 
than the mere, if accomplished, expert, worried 
without respite by people eager for his verdict on 
their possessions: the list of his writings all of them 
revealing the outlook of the born historian makes a 
truly imposing series, culminating in his monumental 
History of Early Netherlandish Painting issued from 1924 
onwards in fourteen substantial volumes. And for a 
long time the whole of this ceaseless activity had for 
its background Dr. Friedlander's connection with the 
Berlin Picture Gallery and Print Room: their mar- 
vellous growth during the period in question owes in 
fact an enormous debt to the distinguished scholar, 
whose career as an official came to an end in 1933, 
when Dr. Friedlander relinquished the post as Head 



INTRODUCTION 

of the great Picture Gallery, to which he had been 
appointed as Wilhelm von Bode's successor. It is, in- 
deed, the very aroma of that institution in its best days 
which pervades the whole activity of one of the great- 
est of those who stand to it in the relation of at once 
alumnus and creator. 

The opinions on art and connoisseurship, which re- 
present the ultimate wisdom and considered judg- 
ment of a man whose performance has here been sum- 
marily outlined, must inevitably be of the most pro- 
found interest; and it is, indeed, a matter of con- 
gratulation that Dr. Friedlander should have made 
them accessible to a much larger audience than that 
of those friends in many lands who have been admitted 
to the privilege of his conversation. The views ex- 
pressed in the present volume obviously derive a 
peculiar significance from the author's first-hand con- 
tact with the problems concerned, as well as from his 
power of independent thinking. In the Preface he 
characteristically stresses his lack of acquaintance 
with the existing literature on the theory of art, and 
readily, if over-modestly, admits the possibility that 
opinions similar to his 'may already have been ex- 
pressed by others, perhaps even on the basis of better 
reasoning'. The students of aesthetics will, indeed, 
know how to value the judgments of the author pre- 
cisely because they confirm the results independently 
arrived at by others. Thus to give an example 
among many when he speaks (p. 87) of a disinclina- 



10 



INTRODUCTION 

tion to use the expression 'beauty*, it is interesting to 
recall the statement once made by Roger Fry: 'The 
word "beauty" I try very hard to avoid/ 1 

Very wisely, Dr. Friedlander has attempted noth- 
ing in the nature of a cut-and-dried system. In this 
connection it is worth while noting that his manu- 
script of the present volume is headed by the follow- 
ing quotation from Grillparzer: 'In these remarks I 
set out, regardless of any system, to write down, on 
each subject, that which seems to me to spring from 
its own nature. The resultant contradictions will even- 
tually dispose of themselves automatically ; or, inas- 
much as they cannot be got rid of, are going to prove 
to me the impossibility of a system/ 

The present volume is translated from the author's 
original complete manuscript in German. 2 The task 
of translation, if a thoroughly enjoyable one, has 
nevertheless presented considerable difficulties. The 
German literary vehicle, especially if handled by an 
accomplished stylist like Dr. Friedlander, tends to- 
wards a combination of characteristics uniting bold- 
ness of exuberant construction to expressiveness 
which is reminiscent of the Baroque ; whilst the natural 
trend of English is towards the method of dissolving 
phrases into the simplest component parts, a method 
of which Basic English represents the most consistent 
application. I can only hope to have done some jus- 

1 See The Burlington Magazine, vol. xxxv (August 1919), p. 85-. 

2 When last heard of, Dr. Friedlander was an 6migr in Holland. 



II 



INTRODUCTION 

tice to the mastery of Dr. Friedlander's style. I have 
felt encouraged in my effort by the approval which he 
has been good enough to express of such occasional 
translations as I have made in the past of articles by 
him; and I particularly want to acknowledge the help 
which throughout my work I have received from Mr. 
Herbert Read, whose contribution towards the crea- 
tion of an aesthetic terminology in English notably 
in relation to German has been of such importance. 
One or two brief passages in the manuscript, point- 
less in any other tongue but German e.g. when 
relating to etymologies have necessitated the very- 
slightest editing. 

In his Preface Dr. Friedlander has referred to the 
difficulty offered by the problem of illustrating the 
book. His own selection of illustrations has, of 
course, been incorporated with the present volume ; 
but it has occurred to the translator that the author's 
meaning in certain cases might be made clearer by the 
inclusion of a few reproductions beyond those chosen 
by him. These additional illustrations are marked in 
the List of Illustrations with an asterisk. 



TANCRED BORENIUS 



University of London, 
University College, 
^ 1941 



PREFACE 

THE views set out in this volume are the outcome 
of personal experiences gathered during the life- 
time of one man. It will help towards an understand- 
ing of the text, and further a friendly reception, if 
a few clues are given about the author, particularly 
about the manner in which he arrived at his general 
outlook. 

Born in 1 867 in Berlin, I grew up in a house which 
was barely two hundred yards from the Altes Mu- 
seum. I studied art history in Munich, Leipzig and 
Florence, my natural inclination being from the start 
towards the attitude of the 'connoisseur' rather than 
that of the university lecturer. For the practice of the 
science of pictures I was fortunate in coming across 
three distinguished men as my teachers: during the 
time of my stay in Munich, Adolph Bayersdorfer ; then 
for the period of a year when I worked as attache to 
the Cologne Museum Ludwig Scheibler ; and finally 
when for decades I found myself a member of the 
staff of the Berlin Museums, the Picture Gallery and 
the Print Room Wilhelm Bode. 

Bayersdorfer, Keeper of the Alte Pinakothek, has 
written but little; without aiming at a far-reaching 



PREFACE 

influence he, primarily by word of mouth, unselfishly 
shared his wide experience with others. His brochure 
Der Holbein-Streit (1872) and his posthumously pub- 
lished writings (Munich, 1902) give some idea, 
though by no means an adequate one, of the many- 
sidedness of his interests, the depth of his under- 
standing of art, and that blending, which was char- 
acteristic of him, of acuteness, humour and the atti- 
tude of an eccentric, contemplative amateur. 

Ludwig Scheibler shared with Bayersdorfer a lack 
of ambition, and like him has left a literary estate of 
but modest extent* Working untiringly, he grew into 
the first expert on the painting of the Cologne 
School, and on the early Netherlandish School. When 
in 1 894 1 was privileged to be taught by him at Bonn, 
his period of research was already a closed chapter in 
his life. At that time he was providing Carl Alden- 
hoven with facts, thereby making it possible for this 
litterateur with the schooling of a humanist to write 
his history of the Cologne School of Painting. Him- 
self, he had by then turned to the history of keyboard 
music. The universal, tragic fate of the expert has 
been experienced by Scheibler with unwonted harsh- 
ness. The many true things, which he had been the 
first to recognize and had expressed with pertinent 
brevity, became even in his lifetime common pro- 
perty ; but his own name was mentioned almost only 
when it was a question of contradicting this or that 
'attribution' of his. 



PREFACE 

Wilhelm Bode is survived by such fame as an ex- 
pert, collector and organizer, and his importance for 
the blossoming forth of the Berlin Museums is still so 
dazzlingly present in everybody's mind that I need not 
devote many words to what I owe to him ; to the stim- 
ulus and inspiration which his incomparable energy 
communicated to his assistant, who had the good for- 
tune to collaborate with him during the decades in 
which the Berlin Museums were enriched in so truly 
happy a manner. Bode's fanatical eagerness for work, 
his universal connoisseurship and his authority, 
created a close network of connections with collec- 
tors and dealers all over the world, with the result 
that in his study works of art were offered for sale, 
placed on show and came up for judgment each and 
every day. 

If I have failed to become an expert, the fault is, 
decidedly, to be laid at my door; it is impossible to 
ascribe it to unfavourable circumstances. 

Inclination and official duty have led me to prac- 
tical contact with concrete problems. When to-day, 
not without satisfaction and not without regrets, I look 
back upon the way in which, dissipating my energies, 
I have day by day with greater and lesser certainty 
given attributions out to the world, I feel the need of 
collecting myself, of explaining and justifying question- 
able activities. 

In these essays I endeavour to reach an under- 
standing in principle of the nature of art in general 



PREFACE 

and painting in particular, I aim at a greater definite- 
ness of terminology, and I build up for myself ideas as 
to the relation of scholarship to art. The first sections 
sound theoretical, with unwarranted intrusions into 
the domain of philosophy ; those which follow deal 
with the practice of picture criticism. History of Art 
is touched upon in order to furnish proofs in support 
of my views, the instances being naturally chosen 
above all from the domain, familiar to me, of Nether- 
landish painting of the i^th and i6th centuries. 

I am of the opinion that every true observation 
concerning any individual work of art may contribute 
to the better understanding of visual art as a whole, 
indeed of art activities in general. 

Out of indolence, perhaps also from a sound in- 
stinct, I have hardly read any literature on the theory 
of art. It may be that most of that which has struck 
me, or occurred to me, has already been expressed by 
others, perhaps even on the basis of better reasoning. 
I venture however to speak from the conviction that 
knowledge, gained directly from one's own consider- 
ation of the work of art, as honest evidence possesses 
some educational value, and may claim some notice. 

The illustration of this volume has been a trouble- 
some matter to me. A reproduction is only justified 
if it supplements what is said in the letterpress, and if 
it makes understanding easier. I have been forced to 
recognize that over a wide extent of my studies the 
small monochrome reproductions are of no use. The 



16 



PREFACE 

illustration has turned out meagre and unequal, only 
occasionally coming to the assistance of the written 
word with graphic force. 

Certain things, here formulated, I have already be- 
fore tried to express with different words, namely in 
the brochures Der Kunstkenner (1920) and Echt und 
Unecht (1929). I may perhaps hope that these obser- 
vations and definitions, now worked into a wider con- 
text and submitted with better pleading, have gained 
in effectiveness. 

To Dr. Crete Ring I owe a debt of gratitude for 
various helpful suggestions. 

MAX J. FRIEDLANDER 



I 

SEEING, PERCEIVING, PLEASURABLE 
CONTEMPLATION 

THE eye is, to the anatomist and physiologist, 
something like a camera obscura, a working ar- 
rangement of mirrors. What it means to see is, how- 
ever, not explained, as long as we think of it as a 
passive attitude, as the mere reception of irritations 
by light. Seeing is not suffering something: it signifies 
doing something, it connotes spiritually emotional 
action. The word 'perceive' indicates that we grasp 
something with the pincers of the sense of sight and 
take something in. 

At all events, here is an object from which light 
signals hit the eye. I know full well that philosophers 
deny the object, do not want to know anything about 
it; I side however with the empiricists and realists. If 
several painters at the same time, and under the same 
conditions of light, portray the same thing, pictures 
result which are different from one another; this 
has often been noticed and is a truism. Up to a point, 
however, the pictures resemble each other, and to 
the extent that they resemble each other, and inas- 



SEEING, PERCEIVING 

much as they resemble each other, they afford evi- 
dence regarding the 'thing-in-itself'. The spiritual eye 
works, setting things into order, supplementing, dis- 
carding, selecting not creating, not inventing. No 
philosopher can forbid us to hold as true that which 
we observe. If we do not believe in the object, then 
we cannot explain how an understanding between 
artist and spectator becomes possible. 

I have a bunch of roses before me at a distance 
of about two yards. The eye at rest receives the mir- 
rored image of one of the roses, by no means of 
the whole bunch. In order to take in the whole the 
eye moves, and, in concentrating upon one part of 
the whole, it cannot see any other, so that even in 
this case of allegedly direct taking in of nature, a 
visual action directed by the mind takes place, con- 
sisting of a linking together of recollected images, a 
gathering and assimilating of many impressions. Every 
movement has its origin in the enquiring spirit. The 
eye also moves while it accommodates itself to strata 
of depth. Finally we see stereometrically with two 
eyes. Seeing is hence not the taking in of a flat image, 
but the combining of two images which give a two- 
fold account of three-dimensional space. 

We must differentiate as follows : 

(A) The object, the fragment of nature, whence 

issue irritations by light. 

(B) The image on the retina, on which, bit by bit, 

the three-dimensional object projects itself. 



THE ARTIST'S VISION 

(C) The recollected image which one might call 

vision created by spiritually emotional 
action. 

(D) The writing down, the realization of the 

vision by the means employed by the 
draughtsman or painter. 

Already at the stage indicated by (B), more cer- 
tainly when (C) has been reached, and not only when 
we arrive at (D), there co-operate habits which have 
formed themselves, possibilities of manual skill and 
intentions of reproduction to which must be added 
knowledge of the object, which springs from previous 
visual experiences, and reports of other senses. The 
draughtsman's vision differs from that of the painter. 
Every artist arranges his inner sight not only in accor- 
dance with his own individual disposition, but also 
with reference to means of realization which are fami- 
liar to, and mastered by, him means which, it is 
true, in their turn are dependent upon his indivi- 
dual disposition. The artist, says Nietzsche, paints 
'ultimately solely that which pleases him; and that 
pleases him which he is capable of painting 5 . The 
draughtsman sees in nature drawings, the painter 
pictures. 

It is not a question of that which is visible, but 
rather of that which we have perceived ; and surpris- 
ingly little of that which is visible gets perceived, that 
is, absorbed into our memory of forms. Here is a 
crude instance. I have read a thick volume, and in 



71 



SEEING, PERCEIVING 

doing so I have seen many thousand letters ; if some- 
one asked me in what type the book is printed, I note 
with amazement that I cannot answer this question. 

The dividing line between working from nature 
and creating from imagination is generally too rig- 
idly drawn. Strictly speaking, neither one thing nor 
the other exists. It is solely a question of differences 
of degree. When we draw or paint, we turn our 
glance away from nature, make an effort to realize 
a recollected image; and on the other hand the 
imagination in its free flights lives on the recollection 
of visual experience it cannot give without having 
received. Keenness and vividness of the vision are 
much less dependent on the shorter or longer interval 
between the impression received from nature and the 
writing down, than on the intensity of the visual im- 
pression and the retentive strength of memory. 

Hence you can well understand that Max Lieber- 
mann, a master who is labelled the consistent Natura- 
list, from his experience of creative work, extended 
the concept of imagination to his own productions ; 
indeed, from his point of view he has with every 
right claimed that his imaginative activity is the only 
legitimate and permissible one. As a matter of fact 
every artistic activity which, in order to depict ideal 
forms, consciously frees itself from the memory of 
impressions from nature, runs the risk of falling into 
mannerism. Only an exceptional memory of forms 
enables the artist to fly away from earth. Vestigia ter- 



22 



CONTEMPLATION AND BEAUTY 
rent. In the igth century more than one artist whose 
visual imagination was not strong enough has got 
miserably stuck in thought, literature, non-sensuous- 
ness. 

To the extent that seeing is a spiritual activity, the 
inchoate mass of colour dots gets sifted according to 
concepts. We interpret the dots which are communi- 
cated to us by the eye, and recognize a tree, a house, a 
mountain, an animal, the human body. To see is to 
recognize the outer world which is familiar to our 
consciousness. 

Perceiving, we select that which attracts us, which 
pleases us, which is 'beautiful', as long as we contem- 
plate pleasurably. We see differently, and different 
things, when we take notice with an end in view say 
scent a danger and our will gets worked up. 

Schopenhauer knows no other happiness than the 
negative one of freedom from pain. By extending this 
thought to contemplation we can argue that the vis- 
ible world is beautiful, is enjoyed, as long as it does 
not threaten us, is nothing but an image ; as long as 
we may remain spectators before it. Not only the 
artist but everyone, more particularly the lover of 
art, stands happily gazing before nature, even if the 
non-artist only through art gets educated to such a 
manner of vision. The gap between enjoyment of 
nature and enjoyment of art gets closed, or at least 
narrowed down. 

On the tombstone of each artist might fittingly be 



SEEING, PERCEIVING 

written the words of Lynceus in the second part of 
Goethe's Faust: 

happy eyes, never 
Unblest; for whatever 
Ye have looked on, whenever 
It metjre, was fair. 

Thus speaks the artist born to see. Lynceus is, how- 
ever, also the official guardian of the tower, and the 
leaf is turned over, as he suddenly is reminded of his 
duty: 

Not for my enjoyment merely 

Am I stationed here so high, 

From the dark what horror drearly 

Breaks with menace on my eye. 

The glow of fire, the red flames are no longer 'fair' 
to the guardian of the tower, because he is filled with 
compassion for the old people, who 'will perish in the 
smother' because his 'will' is summoned to con- 
sciousness of the danger, to action, to rescue; because 
he is rudely awakened from the idleness of contem- 
plation. 

We are all of us both artists and tower-guardians, 
to a greater or less degree, according as our soul re- 
acts to the visible outer world. Life in a picture and 
practically the same thing real life as a picture is 
in a sense no business of ours ; we have placed our- 
selves at a distance from it; its harshness no longer 




2. MATTHIAS GRUNEWALD, THE CRUCIFIXION 

Colmar, Museum 
Painted about 1510 




3 . PAUL CEZANNE, AUVERS-SUR-OISE 
about 1885 



GRUNEWALD'S 'CRUCIFIED CHRIST' 
hurts us ; we have become neutral, unconcerned. We 
are grateful to the artist that he has carried us away 
from the evil world. Acting in life or suffering, we 
are creatures, chained to one another and ruled over; 
contemplating, we feel ourselves as lords and masters. 

It is impossible, to be sure, wholly to discard ele- 
mental feelings, such as sympathy with sorrow or joy, 
care and fear; but communicated to us through a 
work of art they alter their constituent qualities. The 
aestheticians generally declare that this alteration is 
due to a re-shaping, stylizing purification and that 
the secret of artistic effect is herein contained. This is 
an explanation which, whether true or false, has had 
a deplorable effect on the creative artists who, thus 
enlightened, have wilfully taken to emasculating life, 
with questionable results. 

Let us think of Griinewald's Crucified Christ, which 
displays the maximum of bodily suffering crudely, 
closely, over-distinctly. Why is the aspect of this 
bearable ? Why does it not release a torrent of hor- 
ror, which sweeps away all pleasurable contempla- 
tion? Because, in spite of the utmost closeness to 
nature, not the tortured body but the picture of it 
rises before us ; because the master communicates to 
us his vision and, thereby, his religious fervour in such 
purity and so decisively, that our imagination, re- 
moved far away from disturbing, unrelenting actual- 
ity, experiences the distant, sublime myth; and the 
fearsomeness becomes deeply affecting drama. In the 



SEEING, PERCEIVING 

picture Christ dies not once, not here: on the con- 
trary, everywhere and always; hence never and no- 
where. 

In the extreme case which the boldness of Griine- 
wald's genius offers us, much is demanded from our 
willingness, our readiness to meet half-way ; and time 
passed before Griinewald had educated lovers of art, 
had made them ripe for his vision and he has perhaps 
not even yet succeeded in the case of everyone. 

In every instance the fragment of nature which 
appears in the work of art has, not without loss or 
gain, been filtered through the nature of one man, 
existing individually and for once. We perceive what 
the artist has seen as far as we are able to do so. 

Enjoyment of art and of nature are mingled; and an 
attempt at analysis produces complicated results. 

I contemplate, say, a landscape picture by Cezanne, 
and can understand it because I have perceived 
similar motifs in nature. Nature is lasting, eternal; 
the changing styles are ephemeral: thanks to our ex- 
periences of reality we can more or less reach an un- 
derstanding with artists of all periods. I go into the 
open air, after having looked at a picture by Cezanne, 
and perceive in nature paintings by Cezanne. I have 
learnt to see from the master. This one finds often 
formulated thus or similarly. Now I cannot, however, 
see more than my disposition permits, and scarcely 
what Cezanne has perceived. Moreover the work of 
art is a fragment of nature seen through a tempera- 



26 



'DISINTERESTED PLEASURABLENESS' 

ment, but I see the picture by Cezanne through my 
temperament. 

All things considered it is impossible to deduce 
more than this : the lover of art perceives nature as 
well as art with his own eyes, the same eyes, only 
that artists have given direction to the way of seeing. 
The lover of art learns from nature to understand 
works of art, from works of art to enjoy nature. 

The * disinterested pleasurableness' of which the 
aestheticians are so fond of speaking, is not to be 
taken too literally. If I sit in the auditorium I am, it is 
true, not taking part in the events on the stage, but 
I am not unconcerned. My curiosity and my thirst for 
knowledge get stimulated. 'Disinterested' can here 
only mean that the events in question do not belong 
to the reality to which I am harnessed ; I am able to 
look at them as it were with the blissful eyes of the 
deceased. A genre picture reminds me of domestic 
happiness, of homely cosiness, or of gay parties of 
conditions and experiences of my own reality. Land- 
scape pictures call up memories of travels and excur- 
sions, of parts in which I have loved to stay or else 
have experienced something tragic. But everything is 
transfigured and lightened, bereft of its sting, as it 
were at a distance of time and space. Voluntarily, 
without being constrained, I turn to the portrayal 
this is a decisive point and in so doing gain the supe- 
rior restfulness, the happiness of pure contemplation. 
Art creates a second world, in which I am not an actor 



SEEING, PERCEIVING 
but a spectator, and that world resembles Paradise. 

Art performs the function of a servant in that it 
adorns, reports, tells a story, teaches, embodies ideals, 
awakens devotion. Under the protection of the 
Church, art has expanded brilliantly. Artists shook 
the barriers of pure visual art and could with im- 
punity take up with myth, poetry, satire and anec- 
dote, as long as they communicated intellectual or 
spiritual values exclusively by means of form and 
colour. 

Roughly and generally speaking, a development 
in the direction of absolute visual art may be traced 
in history. The motto Tart pour Van 'art for art's 
sake' which gained currency in the igth century, 
proclaimed a desired end. One arrived at emanci- 
pation, as once from the Church, so now from poetry, 
mythology, history ; and threw oneself into the arms 
of visible nature. 

The suppression of human sympathy has now and 
then been evinced most emphatically by the French 
Impressionists. Thus Monet once said to Clemenceau: 
*I am standing by the bed of a dead person, a woman 
whom well, I had loved very much indeed . . . and 
still loved very much. I looked at her eyelids. I said to 
myself: "There is a kind of purple . . . what kind of 
blue is contained in it ? And red ? And yellow T ' ' The 
absolute visual art was preparing to become inhuman. 

The Impressionists have made it a plank in their 
programme to eliminate everything that stands for 



THE IMPRESSIONISTS 

spiritual orderliness and interpretation in favour of 
appearance to the senses : with them it becomes true 
in real earnest 'Whate'er ye have looked on, when- 
ever it met ye, was fair'. Subjectively speaking, in 
their emotional attitude towards the visible world 
they succeeded in carrying out this part of their pro- 
gramme; but not objectively speaking, so far as the 
result is concerned. Their eye is an organ in the spiri- 
tually emotional whole, whose inclinations and in- 
terests decide the choice of the standpoint, the direc- 
tion of the glance, the object. Hence their works are 
no less stamped with personality than those of the 
intentionally idealizing painters. 

A last consequence has been drawn by what is 
known as abstract art. Even now it seemed that out 
of nature too much spiritualness, too much thought, 
streamed into the picture. Following up the endea- 
vour after visual art, one turned one's back on Nature. 

The irony of this last change of front or is it the 
last but one ? lies in this, that those who were bold 
and radical ended up in the primitive category, orna- 
ment an ornament that fulfils no serving function 
but hovers in empty space, free as a bird. 

That which prevents us from speaking of pure visual 
art in front of a picture or a sculpture, lapses com- 
pletely in front of ornament, which is mere decora- 
tion. Here there mingles into the play of form and 
colour values nothing of associated ideas or any 
reminiscence of our reality. If we try roughly to 



SEEING, PERCEIVING 

differentiate from one another the concepts of Emo- 
tion and Feeling by calling the stirring and irritation 
of the soul emotion, and the stirring of the senses 
feeling, then we may say that ornament appeals more 
to feeling than to emotion. The play of mastered 
forces satisfies the need of entertaining the sense of 
sight, and through symmetry, and the turn and return 
of the identical feature, symbolizes order and the rule 
of laws : it arouses general spiritual moods such as rap- 
ture, tension, gaiety, lightness, balance or restlessness. 

It is half praise, half blame when you call a work of 
art decorative. A Persian carpet, a piece of brocade 
cannot, will not, and ought not to be anything more 
than decorative. If however I call a picture ' decora- 
tive', then this verdict contains a derogatory note, 
since by recognizing a satisfying stimulation of our 
senses a lack of more profound effect is admitted. 
Now since every more profound effect touches upon 
matters of spirit and thought, upon human destiny, it 
follows that Art free and noble is less pure visual art 
than the art which serves and adorns ; than industrial art . 

Architecture has, subtly and somewhat wrong- 
headedly, been called frozen music. For architecture, 
tied to a purpose and serving needs, stands in contrast 
to music. More properly you might call ornament 
visible music, and music audible ornament, only that 
music affects the life of our soul more profoundly than 
does ornament. 

I have spoken of 'seeing', but in so doing I have 



GENESIS OF A WORK OF ART 
come to the chapter of creation quite naturally, since 
the reader no doubt will have noticed that the activity 
of the formative artist is essentially contained already 
in the action of seeing, not only in execution, in 
concretizing that which has been seen. 

To make, to shape, to carry out, to produce, to 
execute, to draw, to create: all these are words which 
are used about the genesis of a work of art. The 
French have a particularly pregnant word, realiser, 
which means 'to transform vision into something 
which we can apprehend with our senses' . 'To make* 
is a colourless, neutral expression, indicating an 
action whose result stands before us in the work of 
art. 'To shape 5 indicates an action conscious of its 
aim. 'To carry out', 'to produce' hint at an obscure 
region, where the artistic form remains hidden until 
the artist has brought It out into the light. 'To draw', 
thus to get something out of water, pre-supposes an 
existing matter, a chaos out of which the work of art 
is taken. You 'execute' a copy or a replica; and the 
sober word indicates that the work of art was pre- 
existent to the performance. The nearer we get to 
the concept of genius, the more appropriate are ver- 
bal images of such mystical sublimeness as 'create', 
though strictly speaking there is no originating out of 
nothing. The expression 'to invent' from the Latin 
invenire, to find really contains a legitimate doubt as 
to independence of creation, since a 'find' obviously 
must have had a previous existence. 



II 

EXISTENCE, APPEARANCE, OBJECTIVE 
INTEREST IN THINGS 

THAT which exists is given to the eye as appear- 
ance. The spirit interprets appearance, and de- 
duces from it something that exists, builds up its 
vision and thereby the work of art; in so doing it 
not only supplements, fills in and emphasizes, but also 
exercises tolerance, forbearance and selection. 

The relationship of the artist to the appearance, 
existing here and now, will be modified in accordance 
with his conception of his task, of that which he has 
to create and wishes to create. The countless degrees 
may roughly be classified in three categories, chrono- 
logically following upon one another. 

As long as the master had to depict divinities or 
saints, to retell legends or myths, he took as his start- 
ing point spiritual conceptions and emotions of the 
soul if not a pictorial tradition and used impres- 
sions from nature in order to invest his creations 
with the illusion of being alive and having the pos- 
sibility of existence. 

In order to cope with his task, he did not have the 



THE ARTIST SINGLES OUT 

least occasion to observe a feature of nature in its 
accidental setting and context, or even to regard it as 
picture-worthy. He 'took', he singled out and picked 
out, that which he required for his purposes. Gott- 
fried Keller speaks somewhere of 'the sneaking 
thefts of the artist'. How little did a Greek vase- 
painter, or a medieval painter of altarpieces, need! 
Little, not indeed from incompetence, from lack of 
accurate vision the little is not infrequently aston- 
ishingly true to nature but, on the contrary, be- 
cause the immensely much and notably earthly space 
and individual character not only did not serve his 
intentions, but even threatened to degrade, to con- 
fuse and to defile his vision and thereby his work. He 
had to show something which, as a whole, was not to 
be seen in the open air, and took as his starting point 
a pictorial idea, not a visual experience. 

Every period asks for naturalness in a different 
degree. That which centuries ago seemed natural im- 
presses us now as stylized. 

The period of myth, faith and superstition was 
succeeded by a period full of curiosity, the period of 
discoveries. Interest turned from the invisible Creator 
to visible creation. With the ith century the artist 
becomes something like a devotee of natural science. 
His observation gains in neutrality, tolerance, and 
many-sidedness. More especially, increasing attention 
was given to the organic connection of things such as 
that of man and space and light. That which exists is 



33 



OBJECTIVE INTEREST IN THINGS 
no longer rendered according to preconceived ideas, 
but in conformity with appearance: and appearance 
was trusted to the extent that it promised to give 
reliable information about the gay and confused 
world, which now had become picture- worthy. Any- 
way, visual experiences were combined and arranged 
with the intention of displaying reality in a lucid 
view. 

An objective interest in things born of thirst for 
knowledge intervened in this tendency. We expect 
to find such an objective interest, in its purest form 
and its highest degree, in a dry-as-dust botanist, who 
examines the leaf of a tree. To start with, the botanist 
knows more about the leaf than the artist does, and 
hence he sees more. Upon his observation there is, 
however, imposed a limit, because the leaf interests 
him not as an individual but as a specimen of its 
species. That every leaf of a tree differs from every 
other leaf of the same tree Is something which can 
only disturb or confuse the botanist. It is also a source 
of trouble to him, that the object of his scholarly 
attention, in its location in space, in its distance from 
the eye, in the given conditions of light, appears dis- 
figured, bent, foreshortened and discoloured. What 
concerns him is the inherent shape and colour, freed 
from everything accidental, of the leaf. 

A purely objective interest in things is, indeed, a 
stranger to art, but in conjunction with formative 
power it is capable of fecundating artistic production. 



34 



DUTCH 'REALISM' 

Thus Dutch painting in the i jth century has profited 
freely enough from the thirst for knowledge in re- 
lation to that which exists a fact which becomes 
particularly patent in the work of artists of more 
modest talent and lacking in imagination. If the 
vaunted Dutch 'Realism' failed to grasp individuality 
in the degree that might have heen expected, this is 
to be explained as follows : A report on existing mat- 
ters, accurate and dependable as a record, was asked 
for and produced; but one was not permitted to con- 
tent oneself with appearance this had to be supple- 
mented from knowledge, which brought about an 
approach to type. An exception, of course, is supplied 
by portraiture, in the case of which the objective in- 
terest in things and response to individuality overlap. 

The Dutch painters were specialized experts on 
real things. Potter knew cattle like a farmer, Saen- 
redam buildings like an architect, Willem van de 
Velde knew all about shipbuilding. 

Just as the objective interest in things, so did the 
tendency towards narration invade painting. Jan 
Steen was not impunely a witty judge of men and an 
inventive writer of comedies. As regards native talent 
not inferior to any contemporary or fellow-country- 
man, he did not as a painter achieve, or at least did 
not retain, the uniform mastery of a painter like 
Gerard Terborch, to whom discretion taught the 
wisdom of restricting himself. All too loud and all too 
pointed when he tells a story, makes merry or enter- 



OBJECTIVE INTEREST IN THINGS 
tains, Steen often sacrificed the conscientiousness of 
execution. Intellectually active painters ran the risk 
of crossing the boundary where the visible world 
comes to an end and the imagined one begins. The 
harm which the inclination to be 'poetic' did to art 
in the first half of the i^th century is obvious. 

In the second half of the i9th century painters 
turned their backs on poetry, history and anecdote. 
As the objective interest in things waned, and the 
tendency towards narration was suppressed, art be- 
came independent and autonomous. The relationship 
of the painter to appearance altered once more. The 
philosophers have thrown suspicion on 'the thing-in- 
itself , and have declared appearance to be a creation 
of the human spirit. From sound instinct if not in 
defence of legitimate interests the artist is bound 
to oppose this view. Nevertheless, in the igth cen- 
tury there arose everywhere as a positive deduction 
from that negative doctrine an enthusiastic regard 
for appearance. If the philosopher said pessimistically 
'Reality is nothing but appearance', the artist replied 
optimistically 'Appearance is reality'. From fear of 
destroying the organic connection, the painters came 
to look upon composing, stylizing, supplementing, 
in brief upon every active intervention, as bungling. 
Impressionism directs the artist to a standpoint from 
which he, without misgivings, must portray that 
which enters his field of vision. Confidence in the 
unique visual experience entails -heightened illusion; 



JAN VAN EYCK AND MANET 
accidental singling out of the scene; broad, quick 
handling of the brush ; indifference to inherent form 
and inherent colour: since all things appear as acci- 
dentally conditioned by their location in three- 
dimensional space, and given circumstances of light. 
Since one no longer takes concepts as one's starting 
point, type gives way to individual form. The painter 
lets the picture report, excite, tell a story 'lets' in 
the sense of laisser y not offaire. He is reluctant to be- 
come an interlocutor. 

Jan van Eyck, when he painted a brocade mantle, 
subordinated himself, with an objective interest in 
things, to an existing object: and he created some- 
thing that produces the same impression as a real 
brocade mantle, whereas a Manet contents himself 
with the appearance. It is not to be objected, that this 
difference only consists in a subjective notion: it is 
patent enough in the result. Jan van Eyck's work is 
productive of illusion if we stand at a distance of one 
foot, three feet or two yards, whereas the work of 
Manet is tied in its effect to a definite point of vision 
the very one from which the painter has given his 
rendering of the object. 

Van Eyck's eye moved in front of a world at rest; 
Manet's eye rested in front of a world in motion. 

I want to avoid the impression of my labouring 
under the delusion of being able to assign to each 
master a room on one of the floors of a mansion. 
Every painter -entirely apart from the period to 



37 



OBJECTIVE INTEREST IN THINGS 

which he belongs according to his individual dis- 
position takes up a different attitude towards appear- 
ance from that of every other. Especially from the 
i ^th century onwards boundaries have been displaced. 
Masters of genius such as Titian or Rembrandt 
towards the end of their careers surmount the bar- 
riers which I have set up. Richness and complexity 
of production refuse to be compressed within a 
formula. 

Anyway there remains recognizable one essential, 
main tendency: the transition from active, selective 
fashioning to receptiveness, to unreserved and rever- 
ent devotion to the many-coloured reflection of life, 
and unquestioning acceptance of that which is given 
us precisely here and now in a connected fashion. 



Ill 

ART AND SYMBOL 

COMMON to all artistic activity directed towards 
imagery is the task of making something that has 
been seen by the spiritual eye accessible, through this 
or that manual procedure, to the physical eye. That 
which has been seen is a complex of that which has 
been taken in by the senses and put into order by the 
spirit. The relation of these two factors to one an- 
other determines the countless manners of art, differ- 
ing from each other, which have emerged in the course 
of historic evolution. 

The aes dieticians interpret the secret of art, taking 
Plato as their starting point, by considering that the 
thing to be reproduced is not the appearance, offered 
here and once, but rather the idea thus an image 
which exists in imagination as the deposit of many vi- 
sual experiences, which is perfect, beautiful, cleansed 
of everything accidental. However dangerous this 
theory may become to the creative artist, and ques- 
tionable as a norm for the judge of art but fortun- 
ately the relations between aesthetics and artistic 
judgment are extremely slight it is possible by subtle 



39 



ART AND SYMBOL 

interpretation to give it something like universal vali- 
dity. 

If we remember that even a consistently naturalistic 
painter, who portrays a rose which stands before him, 
has seen many roses, recognizes the rose qua flower, 
that he perceives of the motif only what pleases him 
and what he expects to see, then we approach the 
theory of Plato and are able to say with some justifica- 
tion that the 'idea 5 of the rose is the thing to be por- 
trayed. This theory, however, only holds good if, in 
relation to a given stage of art history, it is restricted 
or expanded, since the individual visual experience 
and hence the individual appearance in the course of 
historical evolution has determined artistic form with 
increasing strength. From a non-philosophical stand- 
point we prefer instead of an 'idea 5 to speak of a 
'clarified recollected picture 5 which differs from the 
individual item through its value as a symbol. 

The symbol is a sign which, through convention, 
habit or immediately through its form and colour 
awakens notions which it is incapable of conveying 
explicitly. In the symbol the visible represents some- 
thing invisible, as the letter represents the sound. 
The banner signifies home, country and patriotism. 
All art is symbolic, since the artist through image, 
word or note communicates to the spectator or 
listener by the material the transcendental, by the 
sign the thought or emotion, by the particular the 
general, by the example the category. 



40 



SYMBOLIZING FUNCTION OF ART 

In front of a Greek statue, no one can mistake the 
symbolizing function of art. The beautiful youth re- 
presents the God. Art, inspired by religion, is ob- 
viously symbolic, presenting as it does things human 
and of the earth as superhuman. The halo is a palpable 
symbol. And even if, in the course of artistic evolu- 
tion, the symbolic effect seems to conceal itself, it 
never disappears. By the capacity to overcome the 
narrowness of the individual case, the formative 
power may be recognized; and by the ease with 
which that which is visible or audible points beyond 
itself, the artistic value may be measured. We must, 
however, pre-suppose that artist and spectator under- 
stand each other, that the artist is strong enough to 
raise the spectator to his own level, and that the spec- 
tator is ready to allow himself to be so raised. 

Architecture and music awaken general impulses or 
vibrations of the soul ; sculpture and painting rather 
more particular ones, and such as have been modified 
by thought. 

Symbol is not to be confused with allegory. The 
former addresses itself partly to the senses, partly to 
the emotions ; the latter as a riddle or a charade 
to the intellect, and is, from its essence, a question- 
able vehicle of art. 

All art is bilingual, speaks as well as sings. In 
showing something factual, concrete, it communi- 
cates at the same time spiritual feelings. Only, 
the listener or the spectator must be capable of re- 



ART AND SYMBOL 

ception. Otherwise it might happen, that someone 
might read a line like Goethe's Ueber alien Gipfeln 
ist Ruh as if it were a weather report. The formative 
artist offers signs which on the one hand enable us to 
recognize something of the familiar world around 
us, on the other convey to the spectator the artist's 
conception of it, his judgment on it and the pleasure 
he derives from it. 

Often and not without reason have art and play 
been brought into connection with one another. 
Especially when the earliest expressions of artistic 
activity are analysed, and the ever burning question 
of the origins of art is being ventilated, do we come 
upon results akin to those which we arrive at when 
investigating the play of children. The child plays, and 
so does the domestic animal the child who has not 
yet been claimed by the struggle for life, the animal 
which is no longer claimed by it. To play is nothing 
but the imitative substitution of a pleasurable, super- 
fluous and voluntary action for a serious, necessary, 
imperative and difficult one. At the cradle of play as 
well as of artistic activity there stood leisure, tedium 
entailed by increased spiritual mobility, a horror vacui, 
the need of letting forms no longer imprisoned move 
freely, of filling empty time with sequences of notes, 
empty space with sequences of form. 



IV 

FORM, COLOUR, TONALITY, LIGHT, GOLD 
ORM and colour are tied to one another. There 



F 



only exists coloured form, and which is the same 
thing colour that has been formed. Everything 
visible consists of parts, which are forms, colours or 
notes, according to the quality which one takes under 
consideration. The extent, quantity and boundary of 
the part is called form; its content and quality is 
called colour; its degree of light, tonality. The limit 
of form lies where one colour ends and another be- 
gins. The human mind has divided up the unified 
appearance. It is true that the selecting, isolating, 
even partially blind visual action is able to disregard 
colour or at any rate to neutralize it, but even in an 
extreme case such as that of the outline drawing, 
black and white which after all are also colours 
remain indispensable for the sake of visibility. Con- 
versely, the boundary of the part called colour may 
have become indistinct or obliterated, in which case 
report and information as regards that which exists 
lose in clearness. Form addresses itself more to 
understanding, colour rather to feeling. Colour pro- 



43 



FORM, COLOUR 

duces an immediate effect as a symbol. White sug- 
gests that which is empty, immaculate, innocent. 
Form and colour stand in the same relation to each 
other as word and note in a song. On the effect of the 
various colours on the mind, on the mood, Goethe 
has spoken in great detail in his Theory of Colour. It is 
strange, by the way, how his sense of colour, so 
amazingly sensitive in front of nature, is of so little 
avail in his judgment of art. 

We come upon the contrast between pictorial 
method of vision, and the draughtsman's method of 
vision. After I have seen a red circle I can retain in 
my memory the circular outline or the red colour, 
according as to whether I am rationally or sensually 
minded or disposed. 

God did not first create the world, and then pro- 
ceed to paint it. The dividing up of appearance was 
furthered by the educational curriculum, the method 
of work and academic teaching. One dealt with that 
which is visible in conformity with the advice: 
Divide et impera. One drew and one painted. As 
students attended to these activities, isolated from one 
another, their visual memory lost the organic con- 
nection between form and colour, and they often 
added to form an unsuitable colour. Their attitude 
towards form was earnest, conscientious and rever- 
ent; towards colour arbitrary and playful. 

The i^th century painter will, say, introduce into 
his composition the figure of a saint which, four or 



44 



THE WORKSHOP OF RUBENS 

five inches high upon the picture surface, seems to 
stand at a considerable distance from the eye of the, 
spectator; his red robe shows however an intensity 
of tonality which corresponds to a far smaller dis- 
tance. Form and colour are here not seized with the 
same visual grasp, nor from the same standpoint. 

To draw is to measure, to lay down proportions of 
size. Since the coloured surface contains implicitly 
the measures and proportions of size, which the 
draughtsman produces explicitly, it follows that 
painting is a complete portrayal, drawing a partial 
one. 

Let us try to realize the genesis of a large picture, 
an elaborate composition in the workshop of Rubens. 
The master drew 'from nature* figures in movement, 
on a small scale, and disregarding colour. He also 
4 out of his head' produced small sketches for the en- 
tire composition, adopting a colouring which was 
appropriate to the view from a distance. With the 
assistance of the drawings the composition, con- 
ceived on a small scale, was transposed by the master 
or the assistants into life size, and the colouring 
strengthened, adjusting it to the view from close by. 

The standpoint is thus continuously shifted: visual 
impressions, received at varying distances, are com- 
bined, and with regard to colour there prevail con- 
vention, habit and routine. A picture such as the one 
now postulated how utterly unspontaneously it 
originated, how heterogeneous it is, how definitely 



45* 



TONALITY 

it did not spring out of one single visual experience ! 
The masters drew 'from nature * and painted 'out of 
the head' . Recollection of that which had been seen 
was keener, better trained with regard to form than 
colour. One may say that for a long time painting was 
nothing but coloured drawing. 

Painters of the xyth and i8th centuries overcame 
the conflict here alluded to up to a point in a higher 
or lower degree according to their individual dis- 
position and the tasks which they set themselves; 
Aelbert Cuyp more effectively than Ruisdael, Chardin 
more effectively than Watteau. Often the intellec- 
tually modest had, as painters, the advantage and led 
the way. It was left, however, to the unacademic Im- 
pressionists of the i gth century to settle the conflict 
completely, to overcome it deliberately. 

Just as appearance in the practice of the school is 
divided up into form and colour, so is it possible again 
to differentiate between colour and tonality. It is true 
that we never perceive a coloured surface without 
a tonal value, which depends on the light. If we 
therefore speak of 'inherent colour' local colour 
we cannot by this understand anything else but colour 
placed under normal and uniform conditions of light. 

Light may flare up in one place, fail in another, 
transform inherent colour and alter its nature. The 
green leaf on the tree looks white when touched by 
the light. In one place colour begins to glow, shows 
the maximum of its force; in another it is extin- 



46 



'CLAIR-OBSCUR' PAINTING 

guished by the flood of shadows. In what we call 
grisaille-painting the inherent colour is completely 
negatived, and appearance reduced to contrasts of 
tonality. In clair-obscur painting, which triumphed in 
the i yth century, the artists one-sidedly and deliber- 
ately paid heed to lighting and, out of wilfully em- 
phasized contrasts of light and shade, gained moving, 
dramatic and mystic effects. In doing so they ven- 
tured upon a subjective intervention, a violation 
which allows us to deduce that the painter's con- 
sciousness of his own worth had risen high. Splendour 
and charm of colour need not as a result be sacrificed: 
the colours may even gain in intensity through the 
contrast with the neutral masses of shade, and by 
being as it were set like jewels. The late works of 
Rembrandt offer a case in point. 

If an objective interest in things prevails, then clear 
and complete information about the inherent colour 
and inherent form of things is asked for, and light 
with its capricious changes is felt to be a mischief- 
maker. An instructive instance is offered by Saenre- 
dam's pictures, if you compare them, say, with those of 
Emanuel de Witte. In the former the light is neutra- 
lized for the love of the factual data of architecture. 

The function of light is a complex one. It makes the 
inherent colours 'light up', but in an extreme degree 
be it strength or weakness it destroys form as 
well as colour; it also yields colour gold and silver. 
Bodies appear three-dimensional by not letting 



47 



LIGHT 

through the light. In this fashion light emphasizes the 
cubic illusion, and so gives greater depth to space. 
Finally light composes, since there exists a solidarity 
between beings and things in this, that they are made 
visible by the same source of light, that they, in a 
sense, owe their existence to it. The parts of a pic- 
ture form a family from being the children of light. 
A consistent observation of lighting a late conquest 
came to replace the rhythmic disposition achieved 
by symmetry and equipoise. 

A body which is placed against the light appears 
dark on one side this is the shadow of the body 
itself. A shadow which lies outside the boundaries of 
the body is called a cast shadow. It displays at times 
a mirrored image of the body which intercepts the 
light, and is more or less distorted. It serves as a 
bridge between the body and the world around it. 
The shadow of the body itself satisfies the objective 
interest in things; the cast shadow, on the other 
hand a mimicking addition, tail or train tells you 
nothing about inherent form, but it does tell you 
about spatial connection and the source of light. Only 
at a late period was the cast shadow made to serve 
the ends of the effect of a picture. Because sunlight by 
day, in the open air, spreads a diffused luminosity and 
does not let the cast shadows appear very distinctly 
by contrast to artificial light in dark rooms did it 
come about that Northerners, especially Dutchmen, 
took the lead in observing cast shadows. 



48 




4. MASTER OF ALKMAAR 
PANEL FROM THE SERIES OF THE WORKS OF MERCY 

Amsterdam, Kijksmuseum 
Early example of strongly developed cast shadow, l$Oj 



WARM AND COLD COLOURS 
Through rays which have been thrown back re- 
flections the shadow of a body is partly made lighter. 
This phenomenon remained unnoticed for a long 
time, more particularly since it seemed to confuse 
rather than to clarify the information about cubic 
shape. The rendering of the reflections as a symptom 
of close observation may be followed in the Low 
Countries. Jan van Eyck, Dirk Bouts, Hugo van der 
Goes, and in the i6th century more particularly Jan 
Gossaert, turn their attention to this lighting at second 
hand. Stephan Lochner, the willing pupil of the 
Netherlandish artists, uses the light streak in the 
shaded portion to set off the body against the dark 
background. 

We speak of warm and cold colours, as in music of 
the minor and major key. Where colours are con- 
cerned it is a question not of absolute differentiation 
but of something more or less. Red stands at one end 
of the scale, blue at the other. Among the red colours 
there are some which are warmer and others which 
are colder. The red of strawberries, scarlet and crim- 
son are warmer than red lake and vermilion. Ice is 
white and blue, fire glows red. The sense of feel has 
given the sense of sight the terms. We say 'cold 3 of 
the appearance of the sky, of infinity, of distance, of 
everything bald and torpid; warmth is suggested by 
that which is near, which grows organically, which is 
filled with sap, which is alive. 

Colours, according as to whether they belong to the 



49 



COLOUR 

cold or warm category, act as symbols, and indeed 
direct upon emotion, not according to convention. 
The cool colours express remoteness, distance, trans- 
figuration also reserved dignity; the warm colours 
express nearness, seclusion, intimacy, earthly narrow- 
ness. That which is seen in the distance contains 
cooler colours than that which is seen in the vicinity. 
The Impressionists, who carried out their observa- 
tions in the open air, favour the cool colours. In the 
choice of this or that key, race and individuality re- 
veal themselves. 

Here, as an instance, is a comparative table of 
painters who face definitely one way or the other, 
often as a reaction against accustomed tendencies : 

Cold Warm 

Piero della Francesca Dirk Bouts 

Hugo van der Goes Titian 

Greco Pieter de Hooch 

Vermeer van Delft Kalf 

Snyders Adriaen van Ostade 
Teniers 

Gold, regarded as a colour value, is a colour value 
of a special kind, fulfilling in a picture a different 
function from all others. It does not belong to the 
means by which the illusion of reality is conjured up ; 
on the contrary, it is one of the means which remove 
the work of art from the sphere of earthly existence 



GOLD IN THE HISTORY OF PAINTING 
and counteract illusion. The gilt background nega- 
tives space. The precious metal striking the highest 
note of decorative splendour, mysteriously glistening 
in dark churches resembles solemn, wordless music. 
The priceless matter becomes in the work of art a 
symbol of that which is spiritual and lacks body. 

The painted antependium and the altarpiece origi- 
nated as substitutes for metalwork, and for this 
reason long adhered to the high splendour and glitter- 
ing magnificence of plastic decoration and enamel. In 
North Italian painting of the i th century, the in- 
fluence which proceeded from Greek Icons may be 
followed into the art of Carlo Crivelli by reason of the 
accumulation of gilt decoration. The quantity of gold, 
occurring in a devotional picture, betrays the degree 
of conservative, hieratic spirit. 

Step by step we may trace how gold is eliminated, 
from a desire for the ordinary things of this world as 
also for depth of space. Jan van Eyck and some Floren- 
tines in the i th century thus the most progressive 
forces rejected the precious material as irrecon- 
cilable with their method of vision. The Germans 
and the Venetians clung yet awhile to the traditional 
vehicle of glorification and adornment. In the haloes 
whether shown as disks or rays and in the orna- 
ments on the borders of the draperies, gold still re- 
mained in considerable use in the i jth century. Ger- 
man painters about 1470 have, naively and illogically, 
combined the landscape setting with a gilt sky space. 



GOLD 

Altogether, every material In itself 'beautiful' be- 
came more and more neglected in serious art in 
favour of the claims to achieve illusion and given 
over to decoration and industrial art ; as for instance 
silk, ivory, porcelain, gold, and silver. 

At last, ejected from the surface of the picture, ex- 
iled as it were into the ante-room, gold encloses and 
frames the fragment of nature and separates art, born 
of spiritual conception, from our reality. 



THE CONCEPT OF 'PICTORIAL' 

WE have come upon the concept of 'pictorial' in 
the antithesis 'pictorial method of vision and 
the draughtsman's method of vision'. In English there 
exist, alongside of the term 'pictorial', certain ex- 
pressions of cognate significance notahly 'paintable' 
and 'picturesque' ; and these two are of value in com- 
ing to the rescue of 'pictorial', while at times 'paint- 
able' and 'picturesque' may be used almost indiscrimi- 
nately. Thus certain effects observable in the sky 
with its play of mists, clouds and coloured spaces 
fading away imperceptibly suggest both 'paintable' 
and 'picturesque' . 

The concept changes, according as we link it anti- 
thetically with concepts connected with draughts- 
manship, plastic art or mathematics. The contour, 
the outline can produce an effect which, relatively 
speaking, is picturesque. A jagged rock in the moun- 
tains is more picturesque than the Pyramids. Even in 
the domain of geometry is it possible to observe a 
more or less picturesque effect. The oval is more 
picturesque than the circle, the rectangle more pic- 



THE CONCEPT OF 'PICTORIAL' 
turesque than the square. Between two points there 
are innumerable distances, of which the shortest is 
the least picturesque. The unexpected, unforeseen 
form appears picturesque, it occupies the eye, spurs 
it to activity, rescues it from lethargy. As a witty re- 
mark affects the spirit, so does picturesque form its 
zigzag silhouette, its unexpected turns and twists, its 
slanting, displaced boundaries, its interrupted flow 
affect the eye. The eye, in being hunted to and fro, 
reminds the spirit of movement in the outer world. 
A form which shows the traces and results of active 
forces is felt to be a picturesque one. That which has 
come into being organically, has grown organically, 
is more picturesque than that which has been made by 
man: an apple, a tree are more picturesque than a 
billiard ball or a column. Genuine jewels are more 
picturesque than the flawless and uniformly coloured 
ones which are produced synthetically. The desire for 
the picturesque means thirst for nature, flight from 
the prison of common sense, delight in untamed wil- 
fulness, in adventure, in licence, in the amusingness 
of the incalculable. 

Nature adorns and decorates her creatures in the 
animal and vegetal domain. Growth shows itself in 
the yearly circles in a tree, in nodes, layers, or rows 
of dots. On the model of Nature's ornaments man 
has adorned his body, his garments, his tools, his 
bull dings. Succession and repetition, as brought about 
by nature, do not show the dead and stiff orderliness 



VISION AND VEHICLE 

which is achieved by the calculations and measurings 
of the human spirit, but a looser regularity, which 
appears picturesque alongside the one produced 
with ruler and compasses. And attempts to rival the 
art forms of nature in richness of phantasy never 
wholly succeed. You get an idea of this richness of 
phantasy when you reflect that no two people have 
the same fingerprint that is, the same patterning of 
the skin. The individual is more picturesque than the 
typical : the human spirit, with its fondness for order, 
seeks refuge with rules and regulations. 

The painter uses the brush, the draughtsman the 
pencil or the pen. The brush produces spaces and 
dots, the pencil lines. Method of vision and vehicle 
affect each other reciprocally. 

Up to a point, the brush can draw and the pencil 
paint; namely, indicate spaces of a certain tonality, or 
suggest them by hatching. The natural function of the 
one vehicle is and remains, however, to part, to cut 
up, to divide; and of the other to unite and to collect. 

The method of vision is linked up with the interests 
of the contemplative spirit. Drawing appeals to the 
one who turns his attention to things and organ- 
isms in the outer world, classified according to cate- 
gories; who seeks to account for tilings, who aims 
at grasping in appearance that which is permanent, 
solid and constructive at understanding its essence. 
Drawing means to gauge, to fix proportions, to ab- 
stract, to pass over, and to eliminate the confusing 



THE CONCEPT OF 'PICTORIAL' 

play of light and colour. Masters of an actively mascu- 
line disposition conscientious, bent upon imparting 
instruction, severe, thoughtful, concerned with con- 
veying information, telling a story and achieving ex- 
pression have been great draughtsmen. Naive and 
sensual natures, of feminine receptiveness, may be 
found among the great painters. 

The student begins by drawing, and so did the 
human race. Roughly speaking art has developed from 
drawing in the direction of painting: though many 
swervings from the main road may be noted. The 
method of vision becomes modified in accordance 
with ethnical character, climate and individual dis- 
position. It is often said that the Germans are prim- 
arily draughtsmen. But even the more generalized 
claim that the Teutons are draughtsmen, say by con- 
trast with the Latins, would be entirely mistaken. In 
the Low Countries the Teutonic Dutchmen display 
more sense of the pictorial than the half-French mas- 
ters in the Southern Provinces. The great English 
painters of the 1 8th century were indifferent draughts- 
men, which is all the more strange since as collectors 
they showed a profound understanding of the draw- 
ings by the Old Masters. The influence of climate, of 
the condition of the air, seems to be stronger than 
that of race. Amsterdam and Venice the cities of the 
painters lie by the water's edge. The East, where 
sensual, non-spiritual ornament ruled, appears as a 
fountain head of a pictorial method of vision. Venice 



FALLACY OF INHERITED DISPOSITION' 
lay open to impulses from the East; Greco came from 
there, via Venice, to Toledo, the half-Moorish city. 

The favourite notion, that the artistic activity of 
nations and races from inherited disposition always 
follows the same direction, is often contradicted by 
observation. The French evince in the i2th century- 
superior gifts for monumental sculpture. Then follow 
rather barren periods. In the 1 8th century the French 
lead the way as constructive draughtsmen ; in the igth 
they reach their zenith through their sense of nature 
and pictorial method of vision. The Germans, pro- 
minent as draughtsmen and engravers, are suddenly 
for a short time, through the work of Cranach, Alt- 
dorfer and Griinewald between 1503 and ii5", 
painters in the narrower sense of the word, and in a 
higher degree than the contemporary Netherlandish 
Masters. 



57 



VI 

SIZE AND SCALE, DISTANT VIEW 
AND NEAR VIEW 

THE size of a painting, provided that the artist's 
choice has been spontaneous, tells us about the 
intentions, and hence about the individuality, of the 
painter. In many cases it is not a question of choice ; 
the commission automatically settled the point in 
question. A space of wall of given dimensions had to 
be filled, an altarpiece of such and such a spatial ex- 
tent was ordered. Latterly the painter has had relative 
freedom in his choice of size. 

A reciprocal relationship exists between the spatial 
extent of the picture and its conception ; and again 
between the latter and the manner of painting. Emo- 
tional and spiritual greatness widens the picture-space, 
narrowness of mind favours modest size and small 
scale. The large picture directs the spectator to a 
standpoint at some distance from the picture-surface, 
and forces the painter to change his standpoint, to 
step back in order to take in the whole and judge of 
the effect ; it also makes it necessary for him to adopt 
a broad, quick, summary manner of painting. The 



CAUSES AFFECTING SIZE AND SCALE 
little picture is, in itself, dainty; the large one monu- 
mental, whatever the subject. Every painter may, in 
accordance with his inclination, find a certain size 
convenient, welcome and pleasant, only in past cen- 
turies he was often obliged by action from the outside 
to avail himself of a size which was uncongenial to 
him. 

The small size of the pictures by Dou appears 
natural, like the large size of the paintings by Tin- 
toretto. To each size there corresponds a definite 
quantity of content of form. Hence each size demands 
a definite measure of knowledge of form. 

Period, local conditions and tradition determine 
size and scale. The i6th century strove after monu- 
mentality with increased pretension and ambition, in 
comparison with the i^th century. To the Italians, 
schooled in wall painting, the large scale came more 
naturally than to the Netherlandish artists, whose 
panel pictures in the i^th century do not disown a 
descent from manuscript illumination. As late as the 
i yth century the Dutch felt most happy when painting 
a picture of small or medium size. 

To each picture theme within movable limits a 
definite size is appropriate. Ostade's boors, on a life- 
size scale, would be unbearable. Potter's Bull, large 
as life, makes a grotesque impression. A painter of 
tact and self-discipline, like Gerard Terborch, clings 
with determination to the size which fits his talent, 
his method of painting, and his subjects like a glove. 



SIZE AND SCALE 

Apart from the compulsion springing from the 
order, ambition not infrequently entails discord, lack 
of harmony between size on the one hand, expressive 
power and method of painting on the other. Indis- 
solubly connected with large dimensions is rhetorical 
pathos, which sounds hollow and disappoints the 
spectator, except when conception and expression 
are attuned to the size. 

In the individual development of a painter one can 
notice the direction from small to great, presuppos- 
ing that a strong talent is striving for release and that 
forces from the outside do not intervene with paralyz- 
ing effect. A genius struggles forward along this path 
even against the current of the time, as for instance 
Rembrandt in a period which chooses to move in the 
opposite direction. As they get more mature and 
older, painters gain an ever wider view, a sense of con- 
nection ; they sacrifice details and become psychically 
far-sighted. Attention is turned from the leaf to the 
branch, from the branch to the tree, from the tree to 
the forest. 

Rembrandt has at all times, in all the phases of his 
development, painted large and small pictures ; in the 
early period conception and expression were such as 
befitted the small size, in the late period such as be- 
fitted the large size. In the early period a view from 
close on prevails, in the late period a view from a 
distance. 

A human figure in the distance looks small. Since 



60 



JAN VAN EYCK AND TINTORETTO 
we know its real size the one which we have noticed 
reveals to us its place in the depth of space. The figure 
in the distance, seen across dulling strata of air, loses 
some of its volume, its colour and its distinctness. 
Erroneously it might now be deduced that a figure, 
painted on a small scale, must show the character- 
istics linked up with the distant view. This is by no 
means the case. On the contrary the small picture 
attracts the spectator and pleases him by reason of its 
smooth technique, richness of detail and definiteness. 
Jan van Eyck painted on a small scale with a near 
view a fact which we already noted when we re- 
marked on the discrepancy between form and colour ; 
Tintoretto on a large scale with a distant view. Jan 
van Eyck gives us a reduced near view, Tintoretto an 
enlarged distant view. A life-size near view such as 
Leibl sometimes gives us runs the risk of producing 
the effect of wax- works. Every painter gets accus- 
tomed in conformity with his period and his in- 
dividual bent to a vision derived from a near view 
or a distant view, and he carries this out more or less 
independently of size and proportion. 

Roughly speaking the evolution of painting moves 
from a near view towards a distant view. The primi- 
tive delight in splendid and positive local colours, the 
objective interest in things, which forbade the sacri- 
fice of the facts of inherent form and inherent colour, 
were for a long time more authoritative than the logic 
of sight as conditioned by atmosphere and perspec- 

61 



DISTANT VIEW AND NEAR VIEW 

tive. Up to a point the evolution of the individual 
corresponds to that of mankind. 

The open, broad, bold method of painting, in con- 
trast to the one which is firm and enamel-like, deve- 
loped parallel with the transition to a distant view. 
One may compare for instance the stone walls in the 
pictures of Jan van der Hey den and Jan van Goyen. 
In one case the stones are indicated emphatically and 
accurately, they can be counted ; in the other they are 
rendered by irregular, 'pictorially' lively brushwork, 
in their effect from afar. Jan van der Heyden of the 
near view was younger than Jan van Goyen. There 
was also such a thing as retrograde movement in the 
history of painting. 

Atmospheric perspective cannot be calculated, or 
construed, like linear perspective. According to 
the distance from the eye and the condition of the 
air, colour is subject to alteration, becomes in the 
direction of depth, lighter, paler and more neutral: 
hence it has something to tell us about the locality of 
an object. The painter observes this fact, with the 
intention of arousing and emphasizing the illusion of 
depth of space. While, so far as linear perspective is 
concerned, he is tied by rules, he can treat atmo- 
spheric perspective with comparative freedom. In the 
1 6th century, as depth of space was greeted with 
enthusiasm like a new discovery, the effect of the 
local position on the alteration of colour was often 
exaggerated, intentionally emphasized, schematized 



62 



ATMOSPHERIC PERSPECTIVE 
and distributed in degrees. Three zones were abruptly 
differentiated from one another. The first zone, in the 
foreground, was to be of warm brown colour; the 
second predominantly green; the third light, cool, 
blue. This tripartite disposition carried out for in- 
stance in the landscapes of Jodocus Momper gradu- 
ally gave way to a more natural, imperceptible transi- 
tion of the different grounds into one another. 

Atmospheric perspective, as a compositional de- 
vice, has, notably in the igth century, been accen- 
tuated beyond a relation to observation, for instance 
by Corot, whose late manner is essentially based upon 
gradation of tone. A reaction against it has not failed 
to materialize. 

Within the picture the zones of depth were fairly 
early and particularly eagerly at the beginning of the 
1 6th century in Holland gradated in such a fashion 
that the larger figures in the foreground were done in 
conformity with a close view, and the sma^tr figures in 
the distance in conformity with a distant view. And 
the contrast strikes one at times as very abrupt, for 
instance in Pieter Aertsen's pictures. 



vn 

ON LINEAR PERSPECTIVE 

PERSPECTIVE is, it seems to me, regarded one- 
sidedly in literature as a method of subjective 
expression, as a conquest of spirit striving for order 
and too little as a phenomenon having an objective 
existence. A philosopher may perhaps object that 'a 
phenomenon having an objective existence' is in itself 
a contradiction, but I hope the reader will understand 
what I mean. The boundary lines get displaced in 
accordance with the location of the object in the 
depth of space. The horizontal lines of the side wall 
of a cube of which we know that they run parallel 
converge towards one another. This strikes the eye 
independently of a knowledge of the laws of per- 
spective, a knowledge which as a matter of fact, 
historically speaking, is a late discovery. The pheno- 
menon was always visible, only it was not perceived. 
A painter who, wholly unfamiliar with the laws of 
perspective, shows us the saint large, the house at 
some distance behind him comparatively small, has 
already begun to see in conformity with perspective. 
It is the subjective point of view of the spectator, 

64 



PRIMITIVE ART 

which decides whether ,and how consistently , the modi- 
fication of inherent form through displacement, fore- 
shortening and overlapping is observed and realized 
in a picture. As long as a cube, independently of its 
accidental position, is the primary object of interest, 
its appearance in perspective can be rejected as a dis- 
tortion, as an optical delusion, and wilfully corrected. 
Against the evidence of appearance, without taking 
any notice of it, one was capable of drawing the hori- 
zontal lines of the lateral space of the cube as parallels, 
as long as one gave more credence to knowledge than 
to seeing. This was the case when one aimed at show- 
ing in their inherent forms, and at reproducing, the 
individual object, the thing, the human figure, the 
beings classified according to concepts without the 
least interest in spatial connection or the relation of 
the bodies to one another. 

In Primitive Art, far into the Middle Ages, the pro- 
portions of size were determined less through obser- 
vation than through a spiritual table of precedence. 
Godhead was honoured by size, the sovereign distin- 
guished by size, the donor in prayer, the slave in his 
servitude depicted on a relatively small scale. Look- 
ing in amazement at such compositions people have 
come upon the questionable concept of the 'inverted 
perspective' which has led to absurd deductions. As 
people began to ponder over the world around them 
and over structural surroundings, over the spatial 
relations that existed between the individual things. 



ON LINEAR PERSPECTIVE 

some attention was given to perspective. Step by step 
it may be followed in examples belonging to the 
Middle Ages, how the draughtsmen rendered such or 
such a displacement due to perspective, with approxi- 
mate correctness, at first purely on the evidence of 
their eyes. Early attempts at construing partial and 
inconsistent have, indeed, also been noticed. The 
thirst for space, which in the ijth century grew 
powerfully, increased the capacity for seeing in con- 
formity with perspective. The need was present ear- 
lier than the understanding. Just as the wish to make 
books accessible to the many even to the poor re- 
leased the thought of printing with movable letters, 
so did the thirst for space drive people on to the dis- 
covery of the laws of perspective. The decisive revol- 
ution, furthered by masters of genius like Jan van 
Eyck, preceded the successful geometrical construing. 
After the rules had been discovered, and as they were 
being learnt, everybody was capable of conjuring up 
the illusion of depth on the surface of the picture, and 
the trick was being practised with passion during the 
1 6th century. 

Art which decorates surfaces I have in mind such 
categories as wallpainting, stained glass, vase painting, 
tapestry observes, more or less at all stages of de- 
velopment, a discreet reserve as regards the pheno- 
menon of perspective from disinclination to pierce 
or destroy the wall or the vessel through arousing a 
strong spatial illusion. 



66 



THE MIDDLE AGES AND PERSPECTIVE 

The Medieval master who painted an altarpiece had 
to adorn a piece of church furniture, and to show the 
congregation saints who do not breathe like human 
beings in such and such a room. Neither form nor 
spiritual content, therefore, directed him to aim at 
spatial illusion. Less as a result of immediate observa- 
tion, than of visual knowledge, he took from natural 
appearance exactly as much as he needed for his pur- 
poses of decoration and the concretizing demanded by 
the cult he served. The phenomenon of perspective he 
did not require. 

The painted panel originated in the Middle Ages as 
a substitute for precious panels in relief; and, in con- 
formity with this, it remained for a long time tied to 
the laws of style which govern sculpture. Now to the 
sculptor it is something alien to see in accordance 
with the rules of perspective. 

Painting in Asia, even in its finest performances, 
does not know that thirst for space, which in Europe 
led to the realization of the laws of perspective. Paint- 
ing in Asia cultivates the flat surface and decoration. 
It has well been said that this kind of painting is essen- 
tially writing. If one considers artistic activity all over 
the world, one discovers that intensive interest in 
producing the illusion of depth of space is restricted 
to a relatively small field, both in time and space. 

The development of European painting from the 
ith century onwards may be regarded as a fight 
against the picture surface, and a glorious victory in 

67 



ON LINEAR PERSPECTIVE 

the history of this campaign was the realization of 
the mathematical rules. The thirst for space belongs 
to the period of discoveries, a period during which 
the spirits of men longed to be different and else- 
where ; a period of a generation energetic in its 
endeavour, eager for conquest and relentlessly push- 
ing forward. 

Labouring under a misapprehension, one has said of 
some great masters of the igth century that they had 
renounced the method of vision which takes account 
of perspective. Notably Cezanne has been praised as 
a breaker of mathematical tables of the law. Such a 
view has this much truth, that depth of space, con- 
quered and secured, no longer calls forth enthusiasm 
as a newly discovered land of wonders, and that the 
passion for a complete harmony of the picture-surface 
keeps artists from emphasizing the lines which create 
space, from elaborating depth of space wilfully. Ap- 
pearance, idolized in the igth century, is after all a 
matter of surface, and the objective interest in that 
which exists three-dimensionally has waned. 



68 



VIII 
MOVEMENT 

THE formative artist cannot represent movement, 
although he experiences it in his contemplation: 
he can only interpret it at second hand. The striving 
after movement is intense for more than one reason. 
Movement is life ; it is the symptom of being alive, and 
it is the illusion of life with which art has been and is 
concerned. To give information about events and hap- 
penings entails a change of locality in the course of 
time, which confronts the artist with an insoluble 
problem. It is, indeed, the fate of the artist to find 
himself faced with insoluble problems. 

Space and movement further each other mutually. 
The stronger the illusion of space, the more readily do 
the bodies seem to move in it. On the other hand the 
body in movement creates for itself space as the stage 
of its change of place. Even an object at rest, seen in 
conformity with perspective, creates in our imagina- 
tion the space of which each object occupies a part. 
A body in movement widens space. A figure, turning 
round in dancing movement, emits space around it- 
self. Walking, running, flying, the body suggests the 

69 



MOVEMENT 

space that it has left, and the space which it will reach 
even beyond the frame ; in this way it contributes 
not a little towards making picture-space appear as 
part of boundless space and thereby towards making 
the picture appear as something cut out of nature. 

With movement, time, which is alien to the char- 
acter of formative art, is so to speak introduced by 
stealth : and the door is thrown open to narration, to 
epic, to drama. 

The pendulum of a clock hangs in one phase of its 
oscillations in a perpendicular position, in all others 
in an oblique position. Whoever tries, as best he can, 
to render the swing of the pendulum in a picture, fixes 
on the canvas any position of the pendulum except 
the strictly vertical one. Why ? Because the pendu- 
lum might be at rest in this but in no other posi- 
tion. Whoever will conjure up the appearance of 
movement choses positions in which the body is in- 
capable of remaining permanently. From the report 
'this cannot last', we receive the information 'this is 
in a state of transformation' . 

Pictorial rigidity is to be conquered by cunning and 
discretion. The goal is reached by a circuitous route. 
If in reality I observe a man running or a bird flying, 
and, with an intention of reproduction, try to impress 
on my memory of forms some of the outlines which 
show themselves in a continuous flow cross-sections, 
as it were, of extension in time then distance favours 
me more than vicinity, because the body moves or 



70 



MASTER INTERPRETERS OF MOVEMENT 
rather seems to move more slowly the greater its 
distance from my eye. 

A comparative lack of detail in the form enclosed by 
outline, predominance of the silhouette, neutralized 
local colour these are characteristics linked up with 
a distant view. The more distant, paler, smaller, and 
more unreal a body in itself appears to be, the more 
easily does the impossible become possible namely 
that it moves in the picture. If a master paints a run- 
ning horse in each of two spatial zones of a picture 
and each time with the content of form which 
strikes his eye at the distance concerned then the 
small horse in the distance will produce a stronger il- 
lusion of movement than the large one in the fore- 
ground. And this is the reason why the masters, who 
most eagerly and successfully have gone in for move- 
ment, have shown a preference for small size and a 
sketchy, rapid handling or even were draughtsmen 
or engravers, who were content with black and white, 
the convention which is a stranger to nature. I think, 
say, of Bruegel, Toulouse-Lautrec, or Slevogt. A 
check from the opposite angle is easy. In a wax- works 
a wax figure in real clothes, life size and then in an 
attitude of running would surely produce a ridiculous 
effect. 

Generally speaking this law is valid: the more truth 
to nature is offered, the greater are the claims for even 
more of such truth. Illusion resembles the god that 
consumes his own children. The wax figure just re- 



MOVEMENT 

ferred to, so true to nature, produces an unnatural 
effect because it does not move; thus does not do 
what it ought to do in conformity with the intentions 
of whoever made it. 

Let us compare the Bull by Potter the life-size one in 
the Mauri tshuis with a bull by Rubens. The former 
has been intended for a near view, is true to nature in 
all its details, and bulky; the latter is seen from a 
greater distance, is poorer in detail, flatter in effect 
and partakes of the character of a vision. The former 
is a portrait; the latter more typical, with the char- 
acteristic qualities of this animal species. The latter 
moves, or at any rate appears capable of movement: 
the former, on the other hand, in spite of all natural- 
ness of structure and texture, is without life and pain- 
fully stiff. Potter's bull seems to belong to the real 
bulls, of which we ask that they shall really move. 
Potter, the leading expert on the animal body, fails 
when he tries to depict animals in motion, more par- 
ticularly on a larger scale. 

There exist many devices for increasing the illusion 
of movement, and for strengthening the illusion of 
being alive ; and all converge on the necessity of les- 
sening the reality of a body in movement. We are un- 
able to observe a moving body as closely as a body at 
rest. We lack the necessary time. Sympathetic appre- 
hension of movement forces pencil and brush to quick 
action. 

The opposite pole to the work of art is the 'living 



THE INSTANTANEOUS PHOTOGRAPH 
picture 5 . While the work of art aims at illusion of 
reality, the * living picture' is reality which conducts 
itself as a work of art. And the effect is painful, for 
reasons similar to those which cause us to be 
irritated by a wax figure. 

The instantaneous photograph, the cinematograph, 
the slow-motion picture enable us to check the efforts 
of artists. The camera, thanks to the instantaneous 
photograph, seizes each and every phase, a thing which 
the human eye, experiencing the flow of movement, is 
unable to do. Slow motion shows this flow delayed, 
so that the eye is enabled to grasp picture phases with 
better success. 

By means of the instantaneous photograph however 
instructive and informative it may be the longed- 
for illusion is not to be achieved. It gives less than 
is needed, because the eye cannot be so quick at the 
uptake as the camera; and more, because as com- 
parative checking proves no instantaneous photo- 
graph achieves what the artist aims at. We see in a 
photograph too natural a horse, which leaps without 
being able to move from its place. In order to give 
the appearance of this 'moving from its place' the 
draughtsman employs devices which fall beyond the 
camera's range: he chooses a phase which communi- 
cates to the imagination of the looker-on the before 
and after and thereby the flow of movement. No 
instantaneous photograph shows precisely such a 
phase. Even the quickest visual grasp is offered many 



73 



MOVEMENT 

forms of the body in a rush, but, strictly speaking, 
none is offered it. From his recollection of visual ex- 
perience, and sympathetically apprehending action, 
the artist creates the form which is contained in no 
instantaneous photograph, yet conjures up the flow of 
movement more effectively than any such photo- 
graph. 

Since it is the master who most inclines towards 
the dynamic that is to say, towards what is essenti- 
ally pictorial, the art of the brush and who there- 
fore aims above all at expressing movement; and 
since on the other hand movement is captured rather 
with pencil than with brush, we come up against 
an apparent contradiction which, however, settles 
itself when we consider how 'pictorially* the fana- 
tics of movement say Toulouse-Lautrec draw, and 
conversely how closely akin to drawings are the 
paintings by BruegeL 



74 



IX 

TRUTH TO NATURE, ARTISTIC VALUE 
AND STYLE 

THE philistine finds it easy indeed to give reasons 
for his artistic judgment, and is accustomed to do 
it cheerfully, full of his own importance and without 
hesitation. He compares the artist's expression with 
nature or rather with that which he has noticed or 
does notice in nature and then proceeds quite un- 
perturbed to decide on the value or otherwise of the 
rendering. A painter is blamed because he has pro- 
duced a figure unnaturally long, a nose incorrectly 
drawn. The lover of art, who wishes to stress his dis- 
tance from the misguided dry-as-dust spectator, is ut- 
terly disinclined to discuss such a thing as * correct- 
ness 5 . 

Yet it is impossible wholly to eliminate truth to 
nature in passing aesthetic judgment. Art, noble and 
free, has never, and at no stage of its development, 
been able to do without observation of appearances 
available to vision. Let us take the subjective as our 
starting point, namely, the wish and aim of the artist. 
He perceives what pleases him, and he wishes to seize 



TRUTH TO NATURE 

it because it pleases him. Without question he desires 
to render it correctly. 

An understanding between artist and spectator is 
only possible because both speak the same language 
that is, in their vision have experienced the same from 
nature. However bold his imagination, the formative 
artist is thrown back upon nature. If we take a master 
like Greco, who from abnormal disposition did vio- 
lence to natural forms, we can establish that there 
are limits even to so definite and furious an interven- 
tion. Bodies and hands, space and landscape, remain 
in any case within the domain of the recognizable. 

Truth to nature is by no means everything, as the 
philistine fondly imagines : but it is something and it is 
indeed the conditio sine qua non for effect. Perhaps 
graphology may help to clear up the complicated re- 
lationship between accuracy and the artistic value 
which exists in formative art. A particular handwrit- 
ing is not beautiful because the letters are shaped so 
as to be legible; but without such a condition it is 
impossible to admire the writing as beautiful, or full 
of character, or personal in such a case it is nothing 
but meaningless scrawls, meaningless playing about. 
Similar considerations apply to formative art, which 
speaks to us by accustomed valid signs, by natural 
forms visible to all: just as a poem by Goethe consists 
of words which everyone uses. Michelangelo's Moses is 
monumental, powerful, mighty in its effect. If this 
work, however, no longer reminded us of the normal 



PAINTERS AND PHOTOGRAPHY 

human body, which the artist has taken as his starting 
point, then we would be unable by unconscious 
comparison to feel the mightiness, the greatness. 
The superhuman is not to be expressed without the 
human. 

Subjectively every formative artist is addicted to 
Naturalism; objectively he is less so, according as he 
produces visions which differ from common sights. 

With painters, as is well known, you should not 
talk of photography. They fear a confusion and dis- 
claim, often with suspicious violence, any interest in 
photography. It is not an accident that photography 
was born very nearly at the same time as the pictorial 
courage of the Naturalistic painters and the Impres- 
sionists. The camera produces a lasting reproduction 
which has not passed through a human being's head 
and brain: it is hence something new nature's auto- 
graph. 

A photograph is, however, objective only in a 
relative sense namely, in relation to a work of art. 
The photographer too follows the dictates of indivi- 
dual taste, and is a child of his time. In the choice of 
subject, the standpoint, the particular detail singled 
out on the field of vision, the conditions of light, also 
in the focussing of the negative or the greater or lesser 
sharpness, there remains entirely apart from re- 
touching considerable latitude for subjective inter- 
vention. Hence photographs reflect the style of pic- 
tures belonging to the same period. 



77 



TRUTH TO NATURE 

Nevertheless, the photograph gives a standard, a 
reliable report, from which the artist can check his 
vision; and the photograph is not infrequently used, 
and misused, as the basis of artistic activity. 

Now and then a painter on looking at a photograph 
from nature may exclaim in despair: 'If this is the re- 
sult of all striving after truth then all our effort has 
been misdirected!' And perhaps the painter then 
turns to 'abstract art". Such an effect has at times been 
promoted by photography, and it may be assumed that 
in future it will make itself felt even more strongly. 

The optimist consoles himself with the reflection 
that since photography has taken over all the func- 
tions of the pictorial report, the complete satisfying 
of the objective interest in things, and grows more 
competent every day to discharge these functions, art 
can now retire to, and concentrate upon, the inner 
and real field of its activity. 

If we confront the work of art with the photograph 
say in order to prove errors which may be laid to 
the artist's charge and assuming that we are con- 
cerned with a true and authentic work of art, then we 
find ourselves face to face with the personality of the 
author, his style, and, incidentally, perhaps also with 
the secret of artistic activity. 

We say that this church is built in the Gothic style, 
this picture is painted in the style of the Dutch School 
of the 1 7th century, that picture in the style of Mur- 
illo. Work done by the hand of man reveals itself to 



STYLE AND TRUTH TO NATURE 
the senses and to the spirit through its style, as having 
been produced at a certain time and a certain place, 
or by a certain master. 

We use the expression with the intention of praise 
in cases where a personality, or the artistic endeavour 
of a period or a people, or a technical method, evinces 
itself in full purity and clearness, and moreover per- 
vades the whole. Now since the word 'style 5 denotes 
something in the work of art that does not originate 
in existing reality, or at any rate indicates a visual ex- 
perience given not to everyone but precisely to this 
one master, we are inclined to look upon style and 
truth to nature as opposed to one another. And tibis 
anti thesis looms disastrously in writings on aesthetics. 
We must, however, sharply differentiate the subjec- 
tive attitude of the artist from the result. Style is bom 
at all times and everywhere, on individual formative 
power being exercised even in cases where the 
artist gives himself up unreservedly to nature. Style is 
born, where wilful stylizing is by no means intended. 
Nay, perhaps we really ought to speak of style only 
when a work has come into shape, as it were, of 
itself on passing through the brain and heart of a 
human being. A conscious endeavour to purify, ele- 
vate and beautify nature leads to 'mannerism' . In any 
case style connotes producing and stylizing, that is to 
stamp form consciously in this or that fashion: it is 
of a definitely dual nature. 

Drawing, with its emphasis of boundaries and its 



79 



TRUTH TO NATURE 

renunciation of colour, can produce just as profound 
effects as painting. This alone indicates that closeness 
to reality does not exclusively supply the standard for 
judging the degree of merit of a work of art. 

Engraving, woodcut and lithography stand in a 
closer and more convenient relationship to poetry, 
history, satire, and humour than painting. The cir- 
cumstance that truth to nature is limited and les- 
sened through technical vehicles and absence of col- 
our allows of boldness and concretization of thought. 
There opens to the art of black-and-white a room for 
action locked off from reality in which imagina- 
tion can indulge in free flights, without being ham- 
pered by demands for illusion and without being 
dragged down to earth. 

The sculptor, in giving shape to the human body, 
favours materials which as regards colour and texture 
are far removed from the quality of human flesh, such 
as marble or bronze. This is done with instinctive, 
hence infallible, sense of style. Seldom and only ex- 
ceptionally does he choose wax, a material which is 
relatively similar to flesh. 

We admire the interpretation of soft flesh in the ex- 
quisite figure of Judith by Konrad Meit in the National 
Museum at Munich. The figure, carved in marble or 
alabaster, stands no higher than about six inches. If it 
were larger and, say, coloured its naturalness would 
produce a disagreeable effect. 

The painter is not as anxiously afraid of confusion 



FUNCTION OF 'CLAIR-OBSCUR' 
with reality as the sculptor, if only because frame and 
flatness have the effect of barriers. But even he tries 
to evade nature with innocent cunning. Thus clair-ob- 
scur, which was developed by the 'Naturalistic' pain- 
ters of the i jth century with such fanaticism, was a 
counterpoise to the pitiless sharpness of the observa- 
tion of nature, to the brutal closeness to the individual 
model in the figures. The ascetics of Ribera have need 
of the mercifully concealing action of night, which 
with its pathos offers a substitute for that idealism 
which is wanting in the figures, with their prosai- 
cally portrait-like character. Rembrandt's clair-obscur 
signifies an escape from the commonplace luminosity 
of everyday life. 

Two forces produce dynamic tension in the work of 
the formative artist: consciously, a never satisfied im- 
pulse to get to close grips with nature ; and, subcon- 
sciously, a fear to come too close to her. When that 
fear penetrates from subconsciousness into conscious- 
ness, an unfortunate situation arises. Goethe, in say- 
ing * There is no surer way of evading the world than 
through art and no surer way of linking up with the 
world than through art' , may not have had exactly in 
mind the antinomy here formulated ; but I am unable 
to refrain from quoting phrases which provide an ap- 
parent confirmation. 

A gifted but inexperienced draughtsman who, with 
a naive sense of his own worth and with praiseworthy 
intentions, sets about to reproduce nature, becomes 



81 



TRUTH TO NATURE 

desperate as a hailstorm of impressions descends upon 
him. The chapter in Gottfried Keller's novel Dergrune 
Heinrich, describing what happened to the boy as he 
tried to draw a tree, is more instructive than aesthetic 
treatises. The educational curriculum of the artist in 
the past, and up to a point even now, brings it about 
that the student seldom experiences that wholesome 
failure which Keller has described so graphically. This 
is because by copying and imitating he has ab- 
sorbed into his visual memory, and carries with him, 
those comprehensible formulae of art which corre- 
spond to his powers of expression: into these formu- 
lae, as into a narrow river bed, he canalizes the savage 
torrent of his impressions of nature. He often steps 
from the school of art into life precocious and fool- 
hardy. 

There used to exist recipes and rules, say, for the 
painting of foliage. And while we smile at such a ped- 
antic degeneration of academic instruction we must 
yet, on unprejudiced examination of free, indepen- 
dent, 'Naturalistic' production, and not without 
amazement, pronounce it to be the case that every 
master has sought safety from the rising flood by 
withdrawing into a narrow home of his own. Weak and 
dependent artists evade disturbing nature by finding 
. refuge in houses ready-made, self-contained and built 
by predecessors. 

To emphasize one's personal style means to make a 
virtue of necessity. 



82 



THE ARTIST AND NATURE 
The artist is in love with nature, not, like the dilet- 
tante or the virtuoso, with art. The more passionately 
he presses his suit, the more vigorously does he push 
individually differentiating forces sprung from his 
own bent out into the light, and stylizes against his 
will. Only the virtuoso and the dilettante work to 
their own satisfaction ; the artist never loses the feel- 
ing of standing before an insoluble task. 'Him I love, 
who asks for the impossible.' It is love of this kind 
that the artist may claim. 

If he imagines that he has reached his goal, he has 
really come to the end and stands at that boundary 
where mannerism begins. 



X 
INDIVIDUALITY AND TYPE 

MAN, the most significant item among those with 
which formative art concerns itself, occurs in 
real life as an individual ; that is, physically and psycho- 
logically as a creature with unique qualities. There do 
not exist people who are identical. Individual char- 
acter is the hallmark and symptom of the more highly 
developed creature. As formative art strove to pro- 
duce an illusion of life, it made the eye more keenly 
observant of that which is individual. It was attracted 
by the individual shape, but also repelled by it. That 
which was individual appeared common, accidental, 
not corresponding with any concept or ideal. In order 
to give shape to gods and demi-gods, the artist was 
called upon to give life to ideals, through observation, 
without lowering them or distorting them. Thought 
conduces to type, as observation conduces to indi- 
viduality. 

Later, as painting was emboldened to mirror life on 
this earth, the relationship to individual appearance 
became modified. The qualities of inadequacy and 
faultiness, which attach to the individual being, gave 

84 



ATTITUDE OF THE DUTCH 

no offence. But what was it all about, say, in the genre 
painting of the ijth century? The peasant, the cava- 
lier, the housewife were to appear ; customs and man- 
ners of a community, a society, a class were to be 
shown. 

Interest was not taken in the peasant named Willem 
Muller or it was taken only to the extent that he 
was a characteristic specimen. This intention con- 
duced to category, and thereby once again to type. 
The smaller the scale, the easier was it to achieve 
satisfactory lifelikeness, without strong individualiza- 
tion. A small figure in the distance shows that which 
is typical ; a large figure in a near view that which is 
individual. Ostade's boors all resemble one another. 
Jan Steen individualizes in a higher degree, more par- 
ticularly when painting on a large scale. 

Portraiture is definitely conducive to observation of 
the individual case. 

Roughly speaking, the trend of development runs 
from the typical in the direction of individualization: 
but the goal is never reached. Something is always 
offering resistance. Thought making for order and 
judgment lead on to category. 

The deeper we descend into the realms of creation, 
the less developed do we find individuality. In an ani- 
mal, in a plant, we only notice that which is charac- 
teristic of the species. Man alone, among all creatures, 
shows himself a unique individual being. One may re- 
cognize in it an advantage and the 'crowning happi- 



INDIVIDUALITY AND TYPE 

ness of earth's children' (Goethe). There is, however, 
a standpoint from which individuality is recognized 
in the aesthetic sense as something which is ugly ethi- 
cally, as a curse in the Christian sense, as a conse- 
quence of the Fall of Man, as something which is un- 
redeemed. And in order to understand the extent to 
which artistic activity is dominated by religion, one 
ought to approach this standpoint. 



86 



XI 
ON BEAUTY 

THE concept of beauty suffers from an ominous 
generalness and painful vacuity. We call beautiful 
that which pleases the eyes, which we look at with 
pleasure. Pleasure, however, offered by nature, is dif- 
ferent in kind from the sense of pleasurableness en- 
gendered by art. That which pleases us, as embodied 
by the artist, can displease in nature, nay prove un- 
bearable. Since beauty in nature and that which has 
value in art are divergent, we feel inclined to avoid 
the expression 'beauty' in judging of art. It would 
however be a mistake simply to exclude from the do- 
main of art beauty in nature that is healthy fairness 
of form, gracefulness of movement, regular features, 
charm. But the relations between beauty in nature and 
artistic value are complicated. 

To begin with, our judgment on beauty in nature is 
dependent on our experiences of art. After all, it is 
from artists that we have learnt to appreciate beauty 
in nature. 

That which in life gives us a sense of happiness, 
through form, colour and movement, is something 

87 



ON BEAUTY 

which we do not by any means seek to find again in 
pictures. Beauty in nature is no indispensable condi- 
tion for artistic production; but it does supply the 
artist with a device. In a picture it symbolizes a noble, 
lofty mind, purity of soul, innocence, saintliness. As 
a sensual attraction it also operates in the context of 
mythological or bucolic subjects. 

Physical beauty in a picture is something typical 
and not individual. This is confirmed by the examina- 
tion of works which we owe to masters thirsting for 
beauty, such as Raphael, Correggio or Watteau, and 
also to the Greek sculptors. 

The more pronouncedly individual a human being 
appears, the more imperfect does he seem. One can 
look upon every individual creature as an attempt that 
failed. The formative artist is faced with the alterna- 
tive of sacrificing either fairness of form or truth to 
nature ; and as long as he had to provide pictures of 
gods and saints, his choice was prescribed to him. 
Later, the cleavage between beauty and truth became 
increasingly wider. 

People have been fond of romancing about great 
masters who derive their ideal of beauty from the face 
of a beloved woman. That this should have happened 
is perfectly possible, but the painter concerned must 
have modified the existing forms, consciously or un- 
consciously. If the Sistine Madonna resembled a 
woman who lived in Rome about i2o, then she 
would not be a Madonna ; and if the Roman woman re- 



88 



THE ARTIST AND BEAUTY 

sembled the Sistine Madonna, then she would lack full 
power of life, nay the possibility of existence. Beauty 
in nature is provided for us in individual form and we 
demand it, because we demand life; but in a picture 
it produces a sense of tedium, because it is here de- 
void of breath and movement. The artist who avails 
himself of well-proportioned limbs and fairness of face 
for purposes of glorification, transfiguration and sanc- 
tification, has no use for that which is individual, 
since he must fear that it would bring his creations 
down to the imperfection and transitoriness of the 
earthly sphere. Everything individual calls to mind 
change and decay ; that which is typical assures us of 
permanence and inviolability. No artist faces reality 
with such fastidiousness as the artist who strives for 
beauty. No artist strives less for beauty than the one 
who is in love with nature. Indifferent artists try to 
derive an advantage from the beauty of their subject. 

It was left to some igth century painters, lacking in 
instinct and ill-advised, to paint Madonnas with lovely 
women as their models and with deplorable results. 

The portrait of a lovely woman, ceteris parjbus, is 
not superior as a work of art to the portrait of an ugly 
old man. To expect a maximum of artistic value if a 
great master paints a prize beauty, is nonsense, as 
everybody sees. 

However trite the expression ' beautiful' may have 
become, even the nicely differentiating lover of art 
will always be driven by necessity in the decisive 



89 



ON BEAUTY 

moment to go back to a word which, in spite of 
its emptiness qua thought, is highly saturated with 
feeling. 

That which is unbeautiful, and repels in reality, is 
by no means excluded from formative art. Its func- 
tions are many. It increases by contrast the loveliness 
and charm of that which is well proportioned. It is 
used as a symbol for what is low, ignoble, evil. The 
desire for change is conducive to irregular formations 
since there exists but one norm, but the deviations 
from it are many. Ugliness is manifold, and hence 
'picturesque'. Fate, or mental suffering, disfigure and 
distort fairness of form ; increasing age consumes it. 
The wish to create, to scourge, and to deride char- 
acters; to indulge in pictorial humour or pictorial 
witticism: many such intentions impel the painter in 
the direction of the unbeautiful and make it picture- 
worthy. 

Masters who keenly observed that which is in- 
dividual, but were unable to free themselves from that 
which is conceptually typical, pushed forward to cari- 
cature, to the ideal of that which is mis-shapen 
Leonardo or Quentin Massys, for instance. 



90 



XII 
ON COMPOSITION 

TO the eye there is displayed a confused and inar- 
ticulate juxtaposition of things ; and to put this in- 
to order is the task of the human spirit. The painter 
told a story, reported, reproduced and strung to- 
gether the picture parts, aiming at achieving a lucid, 
intelligible, limited whole. That which is visible is 
part of infinity: through being taken out by the artist 
it becomes something finite and entire. 

The picture-space contains fields which, differing 
from each other in value, occupy as it were more and 
less honourable places. The middle appears as the dis- 
tinguished position, and towards the sides the impor- 
tance of the locality grows less. The severe symmetry 
of the primitive devotional image corresponded with 
the natural inclination to put the chief accent on the 
centre, and to strengthen the middle axis through the 
equipoise of the side portions. In the Northern wing- 
altarpiece this system of form corresponds with pic- 
torial content. Symmetry more and more loosened, it 
is true rules throughout the design at all times. The 
need of it became paralysed through the striving after 



ON COMPOSITION 

naturalness, after picturesque wealth of movement, 
after variety. Symmetry was felt to be something stiff 
and monotonous. Whoever traces the development 
of composition through the centuries, witnesses a 
struggle against geometrical regularity, and the dura- 
tion of this struggle betrays that the scheme to be 
overcome put up a stout resistance. In a thousand var- 
iations of the lifeless arrangement variations which 
give the appearance of an accidental juxtaposition 
symmetry still remains operative, even when it is 
evaded. The accomplished compositions of Raphael 
convey the impression that the actors spontaneously, 
from a sense of noble and pious decorum, from a feel- 
ing of precedence, have arranged themselves in well- 
balanced groups. The happy marrying of articulating 
and organizing compulsion to inner freedom may be 
the basis of the unique perfection which is felt to 
be classic, and which is also characteristic of Greek 
reliefs. 

Ever since the ijth century, depth was reckoned 
with in an increasing degree. The less the picture- 
surface was regarded as a space to be decorated, the 
more did the central axis lose in importance. While 
in the composition, based on the flat surface, vertical 
and horizontal lines predominated, the painters of the 
1 6th and ijth centuries, by means of diagonal, of 
oblique disposition of the masses, aimed at variety 
and in particular at movement against the direction of 
the picture plane. 




if. DIRK BOUTS, THE LAST JUDGMENT 

Detail. Lille y Museum 
Early example of a principal figure seen from the back, about 1460 




6. ADOLPH VON MENZEL 
SCENE FROM THE LIFE OF FREDERICK THE GREAT 

Woodcut, about 1840 



TREATMENT OF HUMAN FIGURE 
As a symptom of the change in the method of vision 
and in connection therewith in the composition, 
we must regard the human figure which turns its back 
on us. In Primitive art, and far beyond it, everything 
was concentrated upon man: to show him in the com- 
pleteness of his physical structure was eagerly de- 
manded, and in this connection only the full face and 
profile aspects were taken into account as decent, 
clearly informative, and dignified. Just as a tree or a 
house was not displayed in a bird's-eye view, so would 
a human being not be shown seen from the back. (By 
the way, Goethe strictly forbade his actors to turn 
their backs on the spectator.) A tremendous revolu- 
tion was needed before a painter dared to present 
nothing but the back of a human figure. Early ex- 
amples of such a bold enterprise may be found in 
works of Ouwater and Dirk Bouts, those two masters 
who came from Haarlem and worked about 1460. 
The figures which are turned away from us create the 
depth of space towards which they are turned, and 
increase the impression of accidental naturalness in an 
event or a situation which exists for itself and has 
not been got up for the spectator. 

In the i jth century the illusion of spatial depth 
was vehemently demanded. Composition had to be 
subordinated to this demand. Heavy masses were dis- 
posed along the lower edge of the picture, or emerged 
from the foreground at the sides, acting as a repoussoir 
and causing the distance a relative one to appear 



93 



ON COMPOSITION 

airy and luminous by contrast. Enclosed by wings at 
the sides, the 'empty' centre gained a new import- 
ance, because it opened itself like a door towards in- 
finity. 

It was the endeavour of the painters subtly to 
modify and to conceal a method of disposition which 
was based on the laws of mechanics. They strove to 
get away from architecture, to whose severe close- 
ness of construction they had formerly been subordi- 
nated. Not infrequently the art of composing con- 
sisted in concealing the action of composing. A con- 
struction which places that which is light, luminous, 
loose on that which is heavy, dark and solid; which 
places that which is borne on that which bears ; that is 
what reason demands. But the delight in 'picturesque' 
freedom offered resistance to the basic principles of 
the building spirit. Picture composition in its evolu- 
tion allows us to witness a struggle, in which the pre- 
ference for natural caprice gradually became more and 
more victorious over mathematical regularity. 

The consistent Impressionists no longer link up the 
parts of a picture into a whole, so that their interven- 
tion seems to be limited to the selection of the stand- 
point. 

The function of the frame consists in this, that it 
assures us of the wholeness of the composition and en- 
closes the work of art, isolating it. Pictures have at 
all times been framed, even if the manner in which 
this was done has varied. The wall painting, like any 



94 



PICTURE AND FRAME 

picture adorning a flat surface also the colourless 
drawing requires only an enclosing rim of slight 
strength, if any ending at all. The stronger the illusion 
of reality conveyed by a picture, the more insistently 
does the latter demand a frame not one into which it 
may merge as a tapestry merges into its border; but 
a frame which, in colour as well as plastically, will 
stand out against the picture surface in clear contrast. 

The harmony between picture and frame has in 
recent times been completely eliminated. It seems as 
if the frame proclaims, with exaggerated loudness, 
that the thing isolated within its limits is not reality, 
but a picture. In the i8th century, the frame was 
scarcely yet part of the picture : it was rather a bridge, 
which connected it with furniture, wall-decoration 
and architecture, and sprang from the same concep- 
tion of taste. 

The indifference with which the painters in the 1 9th 
century allowed their pictures to be framed produces 
a sense of distrustfulness. An amateur from the Far 
East is likely to find the styleless, haphazard, purely 
decorative gold fillets in which pictures are shown 
with us, simply barbarous. And it is hard to rebut 
such a taunt. As a matter of fact nothing is so compro- 
mising a revelation of the dissolution of expressive 
power throughout the whole domain of formative art 
than the inability nay, the unwillingness to pro- 
duce a type of frame which would be appropriate to, 
and worthy of, the great painting of the igth century. 



ON COMPOSITION 

SchinkeFs enterprise, to put all the pictures of the 
Berlin Gallery into a uniform frame, allows us still 
to guess at a wish to bring the artistic contents into 
some connection with the museum building. It is not 
without a sense of shame that we look upon the wav- 
ering between anarchy and uniformity and turn admir- 
ingly to the past, in which picture and frame formed a 
whole, an organic unity, and the moulded fillet was 
subtly adapted to the picture space, heightening or 
supplementing the effect differently in each case; 
now treating the frame simply, now with great com- 
plexity ; making it a continuation and a barrier at the 
same time. 

Efforts in recent times to invent frames 'correct 
in style* have been rather unsuccessful and often em- 
barrassing. The modern picture must for the most part 
appear in old-fashioned get-up in order to be wel- 
comed as something valuable and precious. 



96 



XIII 
ON THE PICTURE CATEGORIES 

IN conformity with a division according to the con- 
tents, the subject, the picture categories came into 
being in the course of historical evolution. They are 
rooted more or less deeply in the soil of ecclesiastical 
art and allow us at the beginning of their growth to 
recognize their origin from a stiff dignity and serious- 
ness. 

We may make the division as follows : 

I. Altarpiece and Devotional Picture. 

(This includes the rendering of divine or holy 
beings existing and claiming veneration 
either singly or in rows or groups; also the 
narration of legends and the illustration of the 
Scriptures.) 

II. Pictures whose subjects are drawn from Myth, Fairy 

tales, Poetry, History. 

(Imaginative invention adds allegory and apothe- 
osis to myth; and it may be noted that events 
from historical periods have mainly been de- 
picted with an informative intention only from 
the i ^th century onwards.) 



97 



THE PICTURE CATEGORIES 

III. The Genre Picture. 

(This includes the rendering of everyday occur- 
rences originating in customs, work, family life, 
festive occasions.) 

IV. The Landscape Picture. 

(This includes the rendering of animals in the 

open.) 
V. The Architectural Picture. 

(This includes the city veduta and the interior of 

a building.) 

VL The Still Life Picture. 
VII. Portraiture. 

(This includes the single portrait and the portrait 

group ; the rendering of individuals without any 

intention of producing a portrait; and heads 

done as studies.) 

That which is visible, in its complexity combined 
with homogeneousness, resists being divided up ac- 
cording to concept; hence the boundaries between 
the picture categories are in a state of flux. 

A Manet gives us details cut out of life, with no 
other intention than that of realizing a unique visual 
experience: with the result that his works apart 
from his portraits do not seem to belong to any cate- 
gory. To us such a mode of procedure appears so much 
a matter of course indeed, the only duty of painting 
that we take into consideration far too little the 
circumstances which in past centuries directed the 



MANET'S STANDPOINT 

will of the painter. A tendency, issuing from the 
'What' of the rendering, determined, differently in 
each picture category, the 'How', with a vigour of 
compulsion which has become alien to us. Manet 
takes up his standpoint with regard to the whole of 
nature and life as the portraitist has done at all times 
with regard to his sitter. Everything that Manet has 
produced is, in a wider sense, a portrait, to the extent 
that one can paint portraits of flowers, fruit, houses 
and trees. And in this way we come to a provisional 
end and result of the evolutionary process. 

In sketching the rise of the picture categories in the 
chapters which follow, I produce the evidence for my 
statements, since the considerations of principles 
which I have just set out will be, at any rate to some 
extent, confirmed, and indeed supplemented. 



99 



XIV 

RELIGIOUS AND SECULAR HISTORY IN 
PAINTING 

THAT which was considered and up to a point is still 
considered as 'noble' art concerns itself with the 
'subjects of great significance to mankind' ; and for 
long periods of time these are supplied by religion and 
myth. Now, for instance, the Death of Wallenstein is an 
unique and significant event, as is the Procession to Cal- 
vary. The advantage lies, however, with the historical 
picture devoted to a religious subject, for more than 
one reason. To start with, it is at once comprehen- 
sible ; it impresses devout Christians immediately as a 
symbol. Moreover, tied to a long chain of tradition, 
the artist draws from a source of strength which does 
not exist when he paints the Death oj Wallenstein , in 
which case mere poetry and bookish learning provide 
a poor substitute for pictorial tradition. As a matter 
of fact the most admired and permanently valid works 
of art are modifications and paraphrases rather than 
inventions Goethe's Faust, for instance. The greatest 
artists do not reach the summit by flight: on the con- 
trary, they move upwards on trodden paths until they 
cover the very last distance on untrodden 



roo 



DIFFICULTIES OF HISTORICAL SUBJECTS 

History, "which is not sanctified and spiritualized by 
faith, offers subjects which are not easily mastered. 
Historical knowledge and patriotic enthusiasm lag be- 
hind when it is a question of depth, width and dura- 
tion of effect. The picture in itself is silent. It is as an 
illustration where the text gives information about 
that which is visible that it can penetrate most suc- 
cessfully into the domain of poetry. Its real function 
is description of that which is known, not narration of 
that which is unknown. 

Painting of subjects of secular history in the iyth 
century, as handled by Rubens, is not objectively in- 
formative or carried out in the spirit of the chroni- 
clers. Even a Rubens, boldly and unhesitatingly ming- 
ling history, myth and allegory, can have achieved full 
comprehension and consequently immediate effect 
only with his patrons, say the princely house which 
he glorified. To our eyes, it is true, the decorative 
splendour remains sufficiently admirable in itself. In 
the domain of the decorative, for that matter, enig- 
matic quality and obscureness of subject are at home: 
indeed, they may at times have provided a stimulus of 
attractiveness, as for instance in the historiated tapes- 
tries of the 1 6th century, which scholars were wont 
to explain, subtly and entertainingly, to the princely 
patrons. 

There is much which can yet be brought up against 
the historical picture. The mere fact that it is only 
the painters of the i9th century who have taken to 



101 



RELIGIOUS AND SECULAR HISTORY 
this field, pretentiously and with an ambition directed 
towards the monumental, may be regarded as a sus- 
picious symptom. 

If I see a lady of fashion at her toilette, in a picture 
by Terborch, I am capable of finding delight in the 
carpet rendered in a masterly fashion. But whoever 
paints the Death of Wallenstein, and in this connection 
claims my attention for the design of a carpet, is not 
the man from whom I want to hear something about 
the fate of Wallenstein. Even theatrical producers 
have gradually come to see this. 

'Historical accuracy', as to which some painters, 
like Piloty and Alma Tadema, indulged in so many 
illusions, is at all times only capable of convincing the 
artist's contemporaries, who know equally much or 
equally little. But entirely apart from the fact that 
accuracy of portraiture and costume is not to be 
achieved, it is not even desirable: for it lessens the 
sense of sublimeness through which alone the render- 
ing of historically important happenings is justified. 
In the case of an event which took place centuries ago, 
it is only its effect in its historical context, and, say, 
the tragic destiny of a great personality, which can 
claim still to be alive in our imagination certainly 
not costumes and again definitely something spiri- 
tual, which the formative artist cannot immediately 
express. De-realization, removal from its chronolo- 
gical surroundings, apotheosis that is what the ima- 
ginative artist will aim at in order to overcome the 



102 



THE CASE OF MENZEL 

unyielding qualities of the historical picture. What 
was the cause of the failure of the wretched historical 
painters of the igth century ? They aimed at greatness 
and dramatic effect, but simultaneously at reality in 
the highest possible degree. They painted John Huss 
at the stake, the executioners and the deeply moved 
spectators from studio models. Pathos killed natur- 
alness, and naturalness pathos. 

But Menzel? This is a special case. Menzel was 
something like an investigator, operating with a drill 
bore. Also the King whom he depicted was, both in 
space and, relatively speaking, also in time, close to 
him. Finally: he was partly and that part was not the 
worst one an illustrator. 



103 




XV 

THE NUDE 

nude cannot strictly be regarded as a picture 
category, but asserts itself from the i jth century 
onwards, especially in the South of Europe, so insis- 
tently that the subject of a picture often appears to be 
merely an excuse to paint, or an opportunity of paint- 
ing, disrobed people. Nudity is timeless, because 
every cloaking of it assigns to the figure its place in 
time; nudity is primary and perfect. God made the 
body, man the garment. This is the view which pre- 
vails ever since the Renaissance and dominates the im- 
agination of formative artists. Nowadays, at a period 
when physical exercises are eagerly practised, we re- 
gard nudity as that which is natural: formerly it 
denoted the sublime and, earlier still, that which is 
devilish. Morality, as dictated by religion, could not 
be separated from aesthetics, and a Christian could not 
regard the body, burdened with hereditary sin, as 
'beautiful'. More particularly in the North of Europe 
nudity remained connected with fear of witchcraft. 

When artists in the i$th century began to notice 
that only knowledge of the nude made the appearance 



104 



9 ? 



. : 'r:;-M--vV?^ ; v4i^ 

! ; e ^> : : f; :>^;f,:,|^;p| 
V 'Mi JsR'&J -;:'"> > <: .'L 5 :. V ! 




j m ALBRECHT DURER, A NUDE WOMAN 

Drawing. Bayonne, Museum 
A transcript direct Jrom nature, 1493 




8. ALBRECHT DURER, ADAM AND EVE 

Drawing. New York, Morgan Library 

The proportions construed; study for the engraving 

Adam and Eve, 1504 



EARLY STUDY OF THE NUDE 
of the clothed figure comprehensible, keen attention 
was devoted to the organism at work beneath the 
clothing, and also to the skeleton at work beneath the 
flesh. Simultaneously, and in reciprocity with this 
endeavour, pagan sculptures were unearthed and at- 
tracted admiration as authoritative models. In this way 
prejudices crept in, and also something of the stylistic 
character of sculpture, inasmuch as the nude body en- 
gaged imagination less in flesh and colour than in hard, 
cold, colourless stone. Contrapposto attitudes of clas- 
sical statues, and illusions as to measure, law and rule 
guided by which the revered ancients were sup- 
posed to have shaped the beautiful and normal body 
confused the spirits and drove them away from 
observation of nature. 

Diirer never again rendered the nude body with 
such lack of prejudice as in his youth. The honest, 
'unbeautiful' sketch of a nude woman which he drew 
in 1493 may be compared with late results of his un- 
tiring endeavour to gain the mastery of the unclothed 
human body. The ideal was conveyed along curious 
paths. The Greeks had been misunderstood by the 
Romans, as the Romans were by the Italians, and 
again the Italians by the Northerners. 

In the 1 7th and i8th centuries study of the nude 
model alongside of drawing from plaster casts was 
introduced into the academic curriculum, in which 
connection admiration of classical sculpture made its 
influence felt more or less decisively. The nude body 



THE NUDE 

did not appear as innocent in the North as in the South. 
Lewdness with bad conscience crept into artistic ex- 
pression, and instead of natural nudity the stripped 
body was shown. Study of the nude caused a shudder, 
more particularly in puritan countries, and contributed 
to detach the artists from ordinary society, to give 
them a sense of freedom which now and then degener- 
ated into debauchery. 

The study of the human body was carried out in 
conformity with the maxim 'divide and learn' : it was 
studied first in its nudity, then draped in garments. 
The duality thus engendered is regrettably present all 
through artistic activity, to the extent that it was un- 
able to free itself from the scholasticism of the aca- 
demies. 

It is with approval that we read the phrases of 
condemnation which Diderot addressed to the aca- 
demic methods of instruction notably as regards the 
drawing from the nude model and the study of ana- 
tomy. Goethe, who considered the French writer's 
Essai sur la peinture so valuable that he translated it and 
published it with critical annotations, contradicts the 
passionate 'Naturalist' with a sense of superiority, en- 
deavouring to fix the boundary between art and nature 
and defending rule, law and theory. He remains, how- 
ever, at some disadvantage, since Diderot takes as his 
starting point the consciousness of 'good painting' , a 
consciousness which the German lacked. And the 
emotional life of painters, at least, was more familiar 



1 06 



DIDEROT AND CHARDIN 

to the enthusiastic lover of art than to Goethe, who 
looked upon paintings with the coolness of a judge. 
Diderot knew and loved at any rate one real painter, 
namely Chardin. 



107 



XVI 
6 GENRE' PAINTING 

GENRE painting which grew up late as an indepen- 
dent category of art is, in the first instance, deter- 
mined by the standpoint which the master has taken 
up with regard to life ; and only in a secondary degree, 
by the situation which is shown in the picture, or the 
event which is narrated. Up to a point every occur- 
rence can be apprehended from the point of view of 
genre. Caesar crossing the Rubicon is, to be sure, the sub- 
ject for a historical picture, but only to the extent 
that the painter, who develops this episode in his ima- 
gination, is acquainted with the greatness of Caesar 
and the important consequences of his action, and, 
from such acquaintance, interprets the event in an 
heroic key. An innocent observer would perceive 
Roman soldiers who are getting across a river thus 
a scene of genre character. A peasant sowing is un- 
doubtedly the subject for a genre picture. Jean-Fran- 
9ois Millet, from his idea of the sanctity of work on 
the land, raises it, however, into the sphere of the re- 
ligious. In itself, nothing possesses the character of 

108 



EARLY * GENRE 3 PAINTING 

genre or history: it is thought which produces that 
character. 

Historical painting concerns itself with that which 
has happened once and at a given place ; genre painting 
with that which every day happens here. 

Genre-like motifs are to be found in abundance in the 
ecclesiastical art of the Middle Ages. Notably in the 
pictures of the months in the books of hours oppor- 
tunities were provided for pictures of manners and 
customs. In the i6th century the genre picture sepa- 
rated itself, as an autonomous category, from the de- 
votional picture, and did so at first shyly. Motifs from 
everyday life, which had crept into the religious pic- 
tures, push themselves audaciously into the fore- 
ground for instance in the works of Pieter Aertsen 
or legitimize themselves through a moralizing ten- 
dency. Genre, which had arisen in the devotional pic- 
ture in contrast to that which is noble, holy or fair in 
shape, retained for a long time an inclination towards 
distortion and deformation. The^enre picture suffered 
from the caricature as from an infantile disease. 

In the i yth century, when genre painting blossomed 
forth and split into branches superabundantly, notably 
in the Low Countries, an optimistic view of the world, 
a youthful satisfaction with our present life, provide 
the foundation for an observation which penetrates 
into every nook and corner of existence. That which 
had been considered as an earthly vale of tears pre- 
sented here and now an exhilarating aspect. Pictures 



109 



'GENRE' PAINTING 

which show us merry company and gay drinking 
bouts praise life as a festival, the earth as a pleasant 
dwelling-place. Especially in the Northern provinces, 
the Dutch Free State, did one contemplate with a 
sense of happiness the conditions of security obtain- 
ing in the narrow homeland, so victoriously defended 
and now enjoying the benefits of peace. 

Political and economic causes account for the in- 
credible extent and complexity both of content and 
form of the artistic production. On the sociological 
conditions under which the Dutch easel picture blos- 
somed forth, a volume by Harms Floerke may be con- 
sulted, 1 

Not princes, not the Church, not a princely Court, 
not individuals, distinguished and rich patrons no, 
a people, a large number of well-to-do burghers, de- 
termined with their inclinations, wishes and taste the 
artistic activity. A completely secular character was 
here more easy of attainment than in the Catholic 
South, since, as a result of the puritan iconoclasm pro- 
claimed by religion, the tradition of form had been 
interrupted. Moderately priced pictures in tremen- 
dous number, mostly of small or medium size, found 
their way to modest dwellings. Here for the first time 
did painting assume that bourgeois character which 
later became distinctive of modern art generally 
speaking. It is true that it is impossible, on political, 

1 Hanns Hoerke, Studien zur mederldndischen Kunst- und Kultur- 
geschichte (Munich, 



no 



1 7 th CENTURY HOLLAND 

sociological or philosophical grounds, to account for 
the abundance of talents which satisfy the require- 
ments of a new stratum of society. 

The real theme of genre painting is condition, not 
event. It is true that sometimes a descent is made upon 
the domain of the anecdote or short story with par- 
ticular gusto, for instance, by Jan Steen. Vermeer, 
however, and Gerard Terborch are, as narrators, 
reserved and still-voiced. That Goethe could inter- 
pret the so-called * Fatherly Admonition' by Terborch 
so utterly wrongly allows us to realize how undemon- 
stratively the painter told his story. 

Modestly and contentedly comfortable, one only 
wished to see oneself, one's family, one's doings, one's 
possessions, one's house, one's garden, one's country 
in the mirror of 'good painting' . And since accuracy 
and knowledge of facts belonged to good painting, 
purpose namely, the provision of satisfaction of the 
objective interest in things and means namely, 
good painting coincided happily. 

It is mostly a contemplative restfulness which pre- 
vails in the genre picture, as in Adriaen van Ostade's 
mature works and in those of Pieter de Hooch. The 
atmosphere of the day of rest is favoured. The peas- 
ants are not shown working on the land, but smoking 
and drinking in the tavern. To the extent that people 
act, they do what they do regularly and continuously, 
they do what they are accustomed to do. Only actions 
of that nature could count on immediate comprehen- 



iii 



'GENRE' PAINTING 

sion. The housewife looks after her child, the lady is 
at her toilette or makes music or writes a letter, the 
peasants drink, smoke or come to blows, the soldier 
plunders, the child is at play. It is always that which 
is typical, characteristic of the particular stratum of 
society, that is perceived thus, that which is human 
in a general sense, and not indeed that which is unique. 
People appear not as individuals but as representatives 
of their rank, profession, sex, age. Terborch, who as 
regards stylistic instinct is superior to all his rivals, in- 
dividualizes hardly at all in his genre pictures, and that 
in spite of his being an excellent portrait painter. 
The genre picture, within its securely drawn boun- 
daries, existed strictly speaking only in the iyth cen- 
tury and in the Low Countries. Later and elsewhere 
it changes like the chameleon, becomes bucolic, ero- 
tic, decorative, illustrative, sentimental, satirical, 
merges with the historical picture in brief, it loses 
its self-contained, self-supporting character. The emo- 
tional colour which befits the genre picture is gay and 
in humorous agreement with that which is custo- 
mary and belongs to every day. Humour is missing 
in the Italian and Spanish genre pieces, which, often in 
a sullen tone, speak of poverty and need, and, more- 
over, frequently losing harmony with the subject 
resort to the emphasis implied by the large size and 
the large scale of the figures. 



112 




9- JAN VAN EYCK, THE ROLLIN MADONNA 

Detail. Paris-, Louvre 
Approximately half the size of the original 





M 



Q 

CL 





w 
X 



g 6 -. g 

< 73 a. 

o 4 

o| & 

2 ^ b! 

< ^ 



-g 

p 

< 



u 



c 
o 

"42 

I" 

8 



I- 



XVII 
LANDSCAPE 

OF the nature of Primitive art one can learn a 
great deal by speaking to children, observing 
their playful activities, and making experiments with 
them. If I ask a child to draw a landscape, it will be 
embarrassed and fail me ; if on the other hand I sug- 
gest that it should draw a tree, it will eagerly set to 
work. What a child perceives, recognizes, absorbs in- 
to his pictorial memory are things. By * things' I would 
like here to be understood a whole which exists in 
significant contours and may be singled out. In the 
first instance man is a thing in this sense, and, as 
Primitive art almost exclusively had to do with 
creatures moving about freely, it recognized nothing 
as picture-worthy except the self-contained bodies, 
and, weighted down by millennial traditions, it gained 
no access to landscape. 

A mountain, a river are not 'things' : at least they 
cannot be perceived as such and conveyed from their 
natural context into the pictorial context. The moun- 
tain is a tumefaction on the surface of the earth, the 
river a trench, filled with water, in the surface of the 



113 



LANDSCAPE 

earth. In Cennini, who instructively hands on the view 
of the Italian Trecento, we read: 'If you wish to ac- 
quire a good manner of depicting mountains, and 
make them look natural, get some large stones, which 
should be rough and not cleaned, and portray them 
from nature, applying the lights and darks according 
as reason permits you/ 

In order to represent mountains, a stone which is 
a 'thing' is portrayed: and thus that which is essen- 
tial in the formation of a landscape, the local condi- 
tions, the situation in space, is completely left out of 
account. I look at a medieval illumination and per- 
ceive a man and next to him a tree. The subject de- 
mands the tree, since the man experiences something 
in the open air. The height of the picture-surface is 
about 4 inches, that of the man 3 J inches, that of the 
tree 2 J inches. The 'faulty * relation of size between 
the human figure and the tree is thus to be explained: 
the man is, to the draughtsman, that which is primary 
and significant, the tree, as representative of the land- 
scape, is of secondary importance. In order to retain 
the * correct' proportion of size the man would have 
had to be depicted small and insignificant, since the 
picture-space existed once and for all. 

Originally every part of the picture was an indepen- 
dent entity, as large as the picture-space allowed, and 
the human being had predominance ; the rest had to 
content themselves with such space as was left over. 
The real proportions of size were not considered. 



114 



THE MEDIEVAL POINT OF VIEW 
Such a habit of vision placed the greatest difficulties in 
the way of landscape painting. The * thing * is finite, but 
to landscape belongs infinity. It is true that one could 
juxtapose trees, houses, rocks in a jumble, but the 
relation of things to the ground in the depth of space 
and to the source of light in a word, to that which is 
essential remained unattainable, was not considered 
as something to be striven for, as long as the view of 
the world was anthropocentric. And, moreover, in 
the Middle Ages too much interest was not taken in 
the earthly vale of tears. 

In Netherlandish painting of the i th and 1 6th cen- 
turies it is instructive to follow, step by step, how the 
landscape picture develops out of love for nature, 
thirst for space, and knowledge of the laws of perspec- 
tive. The juxtaposition of things, with its logic of 
space, was first grasped by means of a gradation of the 
measures of size. The zones of depth are marked off 
from one another. The horizon is placed high, in 
order to make superimposition in the plane produce 
the effect of succession in space. Very slowly the abo- 
riginal preference for single and complete ' things' gets 
overcome. Roger van der Weyden introduces trees, 
standing far from one another, for the vegetation of 
the background; Memling round bushes, which over- 
lap in rows, yet in such a fashion that through dots of 
light they are distinctly set off against one another. 
Gerard David, who, as regards landscape, is an auctor 
imperil, sets up walls of foliage, and that in the middle 



LANDSCAPE 

distance. And finally the thicket is represented, and 
you no longer see the trees for the wood. 

Since a tree in relation to the picture-space is ex- 
cessively large, it was realized at the time when one 
began to notice proportions of size that everything to 
do with landscape was to be rendered from a con- 
siderable distance. The strong inclination to make the 
land clearly visible compelled the artists, however, to 
render the background from a near view. Giorgione's 
superiority to his Northern contemporaries is not 
least to be seen in this, that he rendered his distances 
from a distant view. Geographical and topographical 
information was demanded from the Netherlandish 
landscape painter. An ideal was to show the earth as 
a whole, with everything that is to be seen on it 
an absurd task. And yet this is what Jan van Eyck, ac- 
cording to Facius, produced for his sovereign: Mundi 
comprehensio orbicular! forma . . . quo nullum consumma- 
tius opus nostra aetatefactum putatur, in quo non solum 
loca, situsque regionum, sed etiam locorum distantiam meti- 
endo agnoscas. Thus something that was half map of the 
world, half landscape picture, was produced. And 
from such landscape backgrounds as we possess by Jan 
van Eyck, we can form a vague idea of this missing 
work; we can understand also why Philip of Bur- 
gundy could entrust his painter with, and expect from 
him, such a performance. 

While an architectural interior can be construed 
in conformity with the rules of perspective, it is im- 



116 




ii. LUCAS CRANACH, REST ON THE FLIGHT INTO EGYPT 

Berlin f Picture Gallery 

1504 



^fefefc. ;,"' 

*j^*:-:^%:^ 



^'-''v^V- : "-;'; ; '': : ^ ; ;v; ; 'V':'- r l 

^" : . '''i-:-^;-. .:..':" s> :\'' 

,'v*-' 1 '-.'';. % * -, ' i '" : : " , ; . < ' , ' -" ' 




EARLY NETHERLANDISH MASTERS 

possible to apply rational means to landscape. The ten- 
dency of the early Netherlandish masters, and especi- 
ally Roger van der Weyden, to introduce buildings in- 
to the background, and to stimulate the illusion of 
space, say, with streets which lead into depth this 
tendency is connected with the embarrassment which 
these masters experienced in developing pure land- 
scape from the point of view of perspective. Jan van 
Eyck, the pioneer, in whose art the old and the new 
clash violently, finds a way out, whenever the subject 
permitted it, by drawing though not actually con- 
struing an interior in the foreground in perspective : 
by concealing the middle distance : and by letting the 
elaborate landscape distance appear as a vista be- 
tween columns, or enclosed by a window. He has 
done this for instance in the Rollin Madonna in the 
Louvre. 

The custom of spreading on a surface figures, and 
whatever else was to be portrayed, had a long lease of 
life. When one had got so far as to see parts of the pic- 
ture at this or that distance from the eye, one pro- 
ceeded to place several picture-planes, layers, zones 
behind each other. In consequence the parts of land- 
scape offer themselves as towering up, frontal and 
parallel to the picture-plane, just as, on the stage, land- 
scape is created by wings, which are set up at inter- 
vals in the direction of depth on the right and the left, 
frontally and parallel to the dropscene. It is thus a 
question of a multitude of degrees, and of a stemming 



117 



LANDSCAPE 

of the spatial tide. Only at a late stage for the first 
time occasionally in Bniegel's pictures was the pri- 
mitive flatness of effect done away with. 

Landscape painting became in the 1 6th century an 
independent branch of the profession. Landscape in a 
devotional picture was a stage, a wall in front of which 
the saints are standing, a vista, a background, a dis- 
tance. It was still thus conceived, thus seen even after 
landscape painting had emancipated itself. Thirst for 
knowledge and respect for creation gave the youthful 
picture-category meaning, justification and impor- 
tance. One wanted to take in the whole of the ele- 
ments of landscape mountains, trees, buildings, 
roads, fields, meadows, rivers and to show anything 
that was to be seen on earth in its spatial connection. 
The principal characteristic of the world was its wide- 
ness, and in order to make it visible one ascended 
mountains, hills and towers. More particularly the 
mountain-chain, seen with the eyes of the carto- 
grapher, and striking the dweller in the lowlands as 
something sensational, became picture- worthy. The 
land had to make considerable efforts tower up 
heroically, assume excitingly picturesque aspects 
in order to be accepted in the first half of the i6th 
century as a sufficient and satisfactory content for a 
picture. Often a biblical or mythological motif pro- 
vided an opportunity or excuse to show a glimpse of 
landscape. That which was primary from the point 
of view of content thus became secondary from the 



118 



GIORGIONE AND OTHERS 

point of view of form: it was embedded as a staffage 
and in many cases is scarcely to be discerned. 

The landscape painter, from a pantheistic sense of 
the world, proclaimed and praised at the beginning 
the extent and richness of Creation; later on, its 
sublime and solemn quiet in contrast to petty and 
restless human nature. 

The geographer with his love of description, the 
architect given to elaborate projects, the lyrical poet, 
all entered into conversation with the observer. 

That, already at the beginning of the i6th century, 
one was capable of realizing unique visual experiences, 
notably when travelling, is proved by Diirer's water- 
colours, which however and this is the decisive point 
were not regarded as pictures in the full sense of the 
word. 

If we leave out of account whether we have before 
us a landscape picture, strictly speaking, or a figure 
scene in which the landscape formations not only 
occupy much room, but are the bearers of the effect, 
then no one can deny Giorgione, the Venetian mas- 
ter, the distinction of being the first great landscape 
painter. More important in the historical context than 
the rapid and intermittent advances of a master of 
genius are the gradual advances which may be traced 
in the Low Countries, in the works of Jan van Eyck, 
Dirk Bouts, Geertgen tot Sint Jans, Jerome Bosch, 
Gerard David and Joachim Patinir. 

Spontaneously there stirs in Southern Germany 



119 



LANDSCAPE 

especially between Ratisbon and Vienna, down the 
Danube independent observation of landscape. In a 
short space of time between 15-00 and 15-10, impres- 
sions of astounding directness are produced here. The 
schemes and conventions referred to above and de- 
rived from tradition whose rule we can observe in 
the Low Countries, appear to be powerless in South- 
ern Germany. What we find here is more visual ex- 
perience than composition, and frequently a low hori- 
zon, a forest in the middle distance, the human beings 
in the landscape, not in front of it, a love of ramifica- 
tion and thickets. It is not a question of a prospect, a 
vista, a panorama, but rather of moving and straying 
in nature, of mingling with nature. 

Three examples may suffice to illustrate this mir- 
acle, sudden as the appearance of a meteor. Lucas 
Cranach's Rest on the Flight into Egypt in the Berlin 
Gallery ( 1 5-04.) , the four panels by Rueland Frueauf the 
Younger in the monastery of Klosterneuburg (i 5-07 ?), 
and the drawing by Wolf Huber in the Germanisches 
Museum at Nuremberg showing the Mondsee (15-10). 
Martin Weinberger speaks, when treating of this draw- 
ing, 'of the straightforward notation of a given locality 
.... mysterious in its unaffected truthfulness'. 1 The 
pictures at Klosterneuburg also contain accurate tran- 
scripts of the locality. The subject the foundation of 
the monastery caused the artist to portray hills, 
fields and forest-land simply and with complete lack of 
1 Martin Weinberger, Wolf Huber, (1930), p. 30. 



I2o 




13. RUELAND FRUEAUF THE YOUNGER 
PANEL FROM THE ALTARPIECE OF ST. LEOPOLD 

Monastery of Klosterneuburg 

1507? 




LUCAS CRANACH, PORTRAIT OF JOHANN CUSPINIAN 
Winterthur, Collection of Dr. O. Reinhart 
1503 



SOUTH GERMAN EXAMPLES 

prejudice: and, in so doing, he unrolls the smiling, 
springlike scene before us in a manner so independent 
of time that the name of Moritz von Schwind has 
come to the lips of many spectators. Incidentally it is 
said that Schwind knew these pictures well. 

Cranach in his picture of 1504 provides for the 
Holy Family a safe resting-place in a German forest, 
and links up the figures indissolubly with landscape. 
How powerfully his imagination was haunted about 
this time by things growing and the life of nature is 
shown by his portraits notably the marvellous pair 
in the Reinhart Collection at Winterthur, in which 
the trees with their knotty branches have penetrated 
into the spatial zone of the portrait-heads in the fore- 
ground. 

Wilhelm Fraenger's compelling descriptions in his 
volume on Matthias Griinewald (1936) direct atten- 
tion to the harmony between man, forest and forest 
retainers, a harmony which in the picture of the Her- 
mits, the panel of the Isenheim Altarpiece, is due to 
the fact that the old men are approximated to the wild 
vegetation of the primeval forest by a kind of mimicry . 

It is extraordinary how quickly this blossom was 
blighted. Cranach' s feeling for landscape gave out 
completely. Altdorfer and Huber remained rather 
more true to themselves, but went somewhat astray 
Huber into calligraphy, Altdorfer into a manner 
that is daintily miniature-like and favours garden-like 
sophistication. The venturesome advance in South- 



LANDSCAPE 

Eastern Germany remains an episode. The main evo- 
lution is centred in the Low Countries. 

Art reached an astoundingly high water mark in the 
work of Jacob Ruisdael and Hobbema. Dazzled by such 
superior mastery, one easily overlooks, however, how 
little these masters, in any particular case, took visual 
experiences as their starting-point; how restricted 
was the view taken of nature as regards standpoint, 
lighting, time of the year and day. Ruisdael never 
wearied of rendering the melancholy peace of the 
country in high summer, late in the afternoon under a 
clouded sky. 

An attitude of greater freedom towards nature, less 
narrowed down by professional routine, was that of 
the masters who occasionally painted landscapes 
above all Pieter Bruegel. He conceives landscape as a 
stage set for dramatic, epic or anecdotic episodes and 
human action; his follower in the iyth century is 
Rubens he, too, scarcely one of the professional 
landscape painters. 

A symptom of the fully matured vision of landscape 
lies in the extent of the sky and its importance for the 
effect of the picture. If there is something which can- 
not be regarded as a finite 'thing' it is aerial space, 
which in consequence, to the primitive method of 
vision, was empty, null and void, simply non-existent. 
To the great landscape painters of the i jth century, to 
a Claude, an Aelbert Cuyp, a Jacob Ruisdael, the sky 
meant a great deal, and beginnings of an observation 



122 



HAARLEM AND VENICE 

of the clouds and effects of light in the works of Jan 
van Eyck, Dirk Bouts and Altdorfer must be regarded 
as tentative advances. 

Finally, it strikes one that the path followed by 
landscape painting runs parallel to the road on which 
painting, drawing away from sculpture, became that 
which it now is. It was in Haarlem and Venice, not in 
Florence, that landscape art blossomed forth. And the 
English artists, in the i8th and igth centuries, evince 
much talent for landscape painting, and little indeed 
for sculpture. 



123 



xvni 

PORTRAITURE 

THE portrait occupies a special position. Aesthe- 
tical purists have wanted to exclude it from the 
domain of artistic activity. Such intolerance on the 
part of the guardians of the temple, however, re- 
quires justification. The desire to invest the transient 
single case with the immortality of the picture has led 
to portraiture. Information as to some given facts of 
form is asked for. The task demands reproduction of 
the individual form, prosaic information, and seems 
to leave little scope for the will to Expression. Karel 
van Mander tells of the Dutch academic artist Corne- 
lls van Haarlem, that he painted wonderfully fine por- 
traits, but unwillingly, since his spirit felt hemmed in 
thereby. 

Within certain limits, the tendency to typify affects 
even portraiture. Whoever has to paint the portrait of 
a military leader, carries with him ideas as to the char- 
acteristics of the profession concerned, and shows us 
in this particular captain the ' cap tain as such', all 
the more definitely since it is desired to pay honour 
and confer distinction. The caricature carries that 



124 




i. LUCAS CRANACH 

PORTRAIT OF THE WIFE OF JOHANN CUSPINIAN 
Winterthur, Collection of Dr. 0. Reinhart 
1503 




1 6. LUCAS VAN LEYDEN, PORTRAIT OF THE ARTIST 

Detail. Brunswick, Museum 

about 1S1O 



EARLY STAGES 

which is individual in another direction, but it also 
eventually reaches a type. 

According to a charmingly ingenious myth, portrait 
was born as a silhouette, as the outline of a shadow. 

At an early stage portrait took the form of a coin 
image. The small scale permitted emphasis of the typi- 
cal aspect and the suppression of individual features ; 
the profile made it possible to fix the essentials of the 
individual form by means of lines, as a draughtsman 
would. The donor, in an altarpiece or a devotional 
picture, also appears in a side aspect, since he turns 
In veneration towards a divine or holy character seen 
alongside of him in the picture-space. The aspect in 
profile secures for the personality represented, in re- 
lation to the spectator, a proud isolation, an existence 
in another world. The side aspect, particularly wel- 
come to the draughtsman and the maker of reliefs, en- 
joyed favour for a long time. 

The full face aspect is embarrassing to the draughts- 
man, inasmuch as in this position the boundary lines 
do not convey as much and as distinct information as 
the spaces of colour and tone. But even at a primitive 
stage a decision was taken in favour of the full-face 
aspect, from the naive sense of need to retain every- 
thing and not to allow anything to be taken from one. 
We have two eyes, and a child will not reconcile it- 
self to only one being visible. 

Painting seeks to combine the advantages of both 
these early types by choosing the side aspect halved 



PORTRAITURE 

the three-quarter aspect. This turn makes it possible 
to show the profile of the nose, the cheek and the fore- 
head: the linear plan of the head, as it were. Both eyes 
become visible. And the relation of the person repre- 
sented to the spectator can be varied, since he can 
turn his glance towards us or away from us. At the 
same time the illusion of movement and hence of 
life is aroused. Vision in conformity with perspec- 
tive made it possible to develop this aspect, and to 
favour it at the expense of the primitive form of pure 
profile and pure frontality. With a special intention 
at times an archaistic one the medal-like profile 
or the decided full-face aspect have, in more or less 
recent times, been chosen perhaps from a wish to 
remove the person represented to a distance in the 
one case, or to force his imperious presence upon the 
spectator in the other. Holbein has thus immortalized 
Henry VIII very effectively when seen full-face, and 
Prince Edward when seen in profile once, for that 
matter, also seen full-face, an aspect which he will- 
ingly chose on other occasions as well. 

Individual character expresses itself in the figure 
and the way of moving, not only in the face. But it 
took a long time before attention was paid to the 
human organism as a whole, except in the case of the 
portraits of donors. The face, as it were the window 
of the human body, the part which is not screened off, 
seemed to suffice as an image of the personality. In the 
i jth and i6th centuries one may follow, step by step, 



126 



CATEGORIES OF SITTERS 

how the hands, half length, three-quarter length, full 
length, become picture-worthy and how the char- 
acter of the person represented, his sphere of life and 
even momentary mood, were more and more ex- 
pressed in his gesture and attitude. The full-length, 
life-size portrait was exceptionally developed already 
by Bernhard Strigel (about 1510), and by Holbein; 
definitely later by Titian, Giambattista Moroni, An- 
tonio Moro and Nicholas Neufchatel. Jan van Eyck's 
double portrait of Arnolfini and his wife is in its 
period something unique. 

In order to understand the development of por- 
traiture one must realize that, in the distant past, it 
was pre-eminently sovereigns that were depicted, and 
that this was done in the spirit of setting up a monu- 
ment to them. A reverent gaze upwards fixed the 
features of the great for the benefit of posterity. In 
the Middle Ages, the figures of donors were treated as 
portraits. A personality was thus in those days por- 
trayed in passive or active connection with venera- 
tion. That which is individual asserted itself slowly and 
gradually against ritual or courtly convention. First 
man, then woman, finally the child were realized in 
their individual and concrete selves. Since the painter 
was a man, and the social position of woman was not a 
free one, the feminine face was more or less decisively 
moulded upon an ideal of beauty. In the double por- 
trait by Jan van Eyck, Arnolfini has much more in- 
dividuality than his wife. 



127 



PORTRAITURE 

Among Memling's portraits there are many halves 
of diptychs, with the sitter's hands joined in prayer 
and turned towards the Madonna. The single portraits 
by the master, which did not come into being as parts 
of devotional pictures, hardly differ in conception and 
expression from the portraits of donors, so that in 
this respect the influence of an ecclesiastical spirit upon 
portraiture still makes itself felt towards the end of 
the i ^th century. 

The self-portrait provides the psychologist with an 
opportunity for stimulating speculation. Externally it 
may be recognized through the glance directed de- 
cisively at the spectator since the painter looked at 
himself in a mirror, and the attention, seemingly ad- 
dressed to us, was devoted to his own appearance. 
This entails a self-revelation, an emergence from the 
picture to a degree which usually is not characteristic 
of portraits. Man does not take up a neutral or objec- 
tive attitude towards his own appearance; his parti- 
cipation is coloured more by his 'will' than by his 
'idea'. Self-portraits do not confirm the view that we 
know ourselves better than others. They are not in a 
particularly high degree 'good likenesses'. Observa- 
tion is interfered with by vanity, by ambition. The 
painter wants to cut a figure ; he takes himself over- 
seriously, portrays himself in a definite situation, 
namely, as gazing, with open eyes, in tension and 
action. The ordinary sitter on the other hand, the 
person whose portrait is being painted, gets tired and 



128 



IDEAL PORTRAIT 

bored. For this reason self-portraits are aggressive and 
dramatic, and not infrequently theatrical. They con- 
vey to us less what the painter looked like than what 
he wanted to look like. One may speak of a rhetoric 
of the self-portrait. 

The ideal portrait that is, the likeness of a per- 
sonality never seen by the artist was a task with 
which the painter often used to be confronted. Every 
saint was an individual being. The medieval painter 
hardly felt the contradiction implied by his task, all 
the more so as the perishable earthly body did not 
claim much attention. A vague pictorial tradition, a 
longer or shorter beard, and mainly the emblems suf- 
ficed to make the saints recognizable. Only when 
heightened illusion of life and, consequently, a por- 
trait-like appearance was demanded, did the artists 
find themselves faced with the difficult and compli- 
cated duty of establishing a harmony between their 
knowledge of a personality and a chosen model and 
at times, moreover, a more or less binding pictorial 
tradition. An important illustration of the manner in 
which a Netherlandish painter of the ith century 
dealt with the problem is provided by the series of 
famous intellectuals in the Ducal Palace at Urbino, 
painted by Justus of Ghent. The grandest example, 
however, the monumental solution of the problem, is 
provided in Diirer's picture known as the Four Apostles. 
With deep earnestness the German artist strove fo 
didactic effect beyond the feeble pictorial tradition 



129 



PORTRAITURE 

and that which was available in the models: he per- 
ceived the four witnesses of faith as representatives of 
the temperaments, and forced thereby that which was 
chance and accident into a system. What he achieved 
by high tension of imagination is appearance full of 
vitality, but also types ; a self-contained whole, since 
according to the ideas of the time there only existed 
four temperaments; moreover, powerful contrasts, 
and, finally, the impression that these men, however 
different from one another, are united in one exhor- 
tation, one warning, one message, one faith. Portrait 
and Idea are here married to one another ; the ideal 
portrait in the highest sense has been achieved. 



130 




1 7. HANS MEMLING, STILL LIFE 

Lugano, Castle Rohoncz Collection 
on the reverse of the Portrait, reproduced Plate 1 




i8. MARINUS VAN REYMERSWAELE, ST. JEROME IN HIS STUDY 

Madrid, Prado 
Origin of the Vanitas Still Life 



XIX 
STILL LIFE 

STILL life, too, had its germ in the devotional pic- 
ture. Notably the Northern masters did not resist 
the inclination to turn their attention to 'dead' ob- 
jects, long before this picture category had emanci- 
pated itself. In the lyth and i8th centuries fruit, 
flowers and objects of every kind were depicted every- 
where, often, however, in a decorative context and 
by painters whose talent was held in relatively low 
esteem. Almost exclusively in the Low Countries did 
one look upon a vase of flowers, a table laden with 
food, as a wholly valid subject for a self-contained 
easel picture. 

In Holland the i yth century was the period of the 
rise of the middle class. The youthful delight in 'good 
things' in wine, fruit, fish, lobsters was stimulated 
in front of pictures. A prince, who is accustomed to 
have delicious food placed before him, does not at- 
tach so much importance to these products as the 
burgher who has got on in the world, or may hope to 
get on. The quantity, the superb and magnificent 
ripeness, of nourishing things glorify the bliss of the 



STILL LIFE 

earth. Kalf proclaims luxury tempered with taste. In 
Antwerp an overwhelming mass of welcome goods 
is accumulated Baroque-fashion say by Snyders and 
Jan Fyt while in Holland exquisite things are ar- 
ranged, as by Kalf and van Beyeren. In the former case 
it is the gourmand, in the latter the gourmet who is 
catered for. In the healthy delight in life, which irra- 
diates from the Netherlandish still life, the Italian and 
Spanish painters of such subjects have scarcely any 
share. 

In the devotional pictures of the i th century the 
details of furniture especially in the cosy habitation 
of the Virgin were carried out most lovingly, also, 
as a symbolic addition, such things as the vessel con- 
taining a lily in renderings of the Annunciation. One 
of the most beautiful portraits by Memling, which 
from Scottish ownership has found its way to the Roh- 
oncz collection at Lugano, shows on the reverse, as a 
decorative, perhaps also symbolic, afterthought, a 
table with a Persian rug on which stands an Italian 
vessel filled with flowers. 

The meat-counters and market-stalls, which Pieter 
Aertsen unhesitatingly about i o placed in the fore- 
ground of Biblical scenes, represent early stages of the 
Flemish fruit pieces, as Snyders and Fyt painted them 
in the i7th century. The Vanitas still life, with books, 
originated in the pictures of St. Jerome, in which we 
see the Saint meditating in his study. Presumably a 
picture by Jan van Eyck in the possession of the Me- 



132 



EARLY AND LATER STAGES 
dici and now lost was the source from which Petrus 
Christus drew, in his picture at Detroit; and it even 
influenced Ghirlandaio in his fresco in the church of 
the Ognissanti at Florence. Moreover, in the altar- 
piece by Jan van Eyck which belonged to Alfonso, 
King of Naples, St. Jerome was represented in his 
study, and the Italians marvelled here also at the still 
life of books. In the i6th century Marinus van Rey- 
merswaele often painted the saintly scholar and in so 
doing always alluded to the transitoriness of the things 
of this world by means of a mass of dry and desiccated 
things, such as written-on parchments and a skull. 
Still life has, as a symbol, proclaimed growing and 
flowering just as much as passing and dying. 

No people and no period have represented every- 
thing visible with so uniform and loving a sympathy 
as the Dutch of the iyth century; with so objective, 
almost scientific, an accuracy of perception ; without 
asking about the spiritual significance and valuing the 
things accordingly. They could as it were copy from 
nature grapes, or the skin of a dead hare, without in- 
curring any risk, seeing that a single apple perfectly 
represents the category, or 'idea 3 , of the apple. 

Woe to the master who looks at the human face 
with the eye of the still-life painter! 



133 



XX 
THE ARTIST: GENIUS AND TALENT 

FORMERLY pictures and sculptures were produced 
in the same spirit as furniture ; that is to say, the 
professional attitude, the relation of the producer to 
his patron or client, and his social position were those 
of the craftsman. Art separated in recent times from 
craftsmanship, or rather, craftsmanship and art parted 
company to the disadvantage not only of craftsman- 
ship. Punctuality of delivery, fulfilment of the agreed 
conditions, solidity of execution were in past days 
demanded from painters and sculptors, and remunera- 
tion adjusted to the time spent. Even Diirer, on asking 
for a higher honorarium from a Frankfurt patron, still 
refers not to his name or the superiority of his artistic 
performance, but to the unexpectedly heavy claim 
upon his time, and to the high cost of the colours 
employed. The sons of painters became painters: in 
choosing a trade one did not wait for special gifts to 
announce themselves. 

We link with the concept 'artist' ideas of special 
qualifications of selectedness, of rare gifts, of an ability 
which is not gained by industry and practice. The 
activity of the artist appears uncontrollable and does 
not fit into the general order of useful, necessary and 

134 



THE ARTIST AND FAME 

profitable labour. The craftsman is more deeply in- 
debted to society, to the community, than the artist, 
and in a different way. The admiration which comes 
the way of the artist has an admixture of suspicion. 
His position depends on fame, and fame on the uncer- 
tain, changeable artistic judgment. Youth lives on the 
hope for recognition, old age on the prestige gained by 
performance or else on defiant contempt of public 
opinion and expectation of posthumous fame. The 
duty towards himself, to which the artist appeals so 
proudly, is strictly speaking nothing less than a duty. 

The 'great name', which compensates the artist for 
the loss of social security, becomes only with the i th 
century once again accessible to painters and sculp- 
tors. The thought that a master of classical Greece 
was able to keep his fame unobscured all through the 
dark ages probably fired the spirits of the i th cen- 
tury, heightened the general esteem of the artist's pro- 
fession, and stimulated ambition to great deeds. 

Already about 1400 the Limburg Chronicle speaks of 
a painter named Wilhelm 'who, according to the 
judgment of the Masters, had been the best in the Ger- 
man lands' . Here we thus already find an order of pre- 
cedence, the bold superlative of appreciation. It still 
took, however, a long time before social position was 
determined through the recognition of uncommon 
gits. To begin with, in the i th century some mas- 
ters, who had freed themselves from the restrictions 
of the Guild organization, managed to get absorbed 



THE ARTIST: GENIUS AND TALENT 
among the crowd of Court retainers the jesters, the 
mistresses, the adulatory poets. 

In the 1 6th century painters and sculptors, striving 
to rank with the intellectual workers, associated 
themselves with the scholars and thinkers Leonardo 
and Dtirer are cases in point. In the practice of crafts- 
manship, conception and execution were indissolubly 
linked together. The dualistic idea, which in the 1 6th 
century crops up in the inscriptions 'invenit et fecit* , 
points to a pride which separates spiritual ownership 
from manual labour and claims the former for itself. 

The i jth century witnessed the painter-prince, a 
type which Rubens embodies most perfectly. His 
financial success, his social rise, were at any rate 
partly due to his artistic powers. But it must be re- 
cognized that qualities of his character, intellectual 
gifts, the manner of a man of the world, diplomatic 
ability, all contributed to the result. A reflection of 
this light fell on the English portrait painters of the 
1 8th century and still upon some painters of the igth 
century like Makart or Lenbach; painters who en- 
joyed their posthumous fame while still alive, and 
consumed it. 

In the i ^th century the concept 'artist' became 
sharply determined. To quote Goethe: 

The song that rises from the throat 
Repays the minstrel well 

that is a typical Romantic idea. 
136 



BOHEMIANISM 

Outside, and at times above, bourgeois society, run- 
ning the risk of a financial decline, free and beyond the 
law, the artist despised the c philistine' and created 
for himself a social class, with special renunciations 
and special pretensions. Bohemianism blossomed 
forth when Romanticism turned into the taverns and 
cafds of the great cities. 

A posthumous fame has above all come the way of 
those painters of the i^th century who belonged 
neither to the type of the painter-prince nor to that of 
the Bohemian, but who, on the contrary, in austere 
and untiring industry, led a middle-class life, and even 
felt some yearning for the solidity of handicraft. 

The process of transformation here sketched 
memorable in the history of civilization modified 
the nature of production. The intensified striving after 
fame carried nervous tension, jealousy, and a desire to 
attract attention into the workshops. One was accus- 
tomed to go to market in order to buy things which 
were good value and were useful; one goes to ex- 
hibitions in order to experience thrills and discover 
talent. A shrill note, a wilful emphasis and over- 
accentuation of individuality, the extravagant instead 
of the extraordinary these become doubtful features 
on the surface of up-to-date production. Only au- 
thentic gifts, firmly self-contained, could hold out in 
this mad confusion. 

The style of a period changes in reciprocal contact 
with fashion, which moves on a lower plane. While 



137 



THE ARTIST: GENIUS AND TALENT 
fashion, prescribing dress and coiffure, rushes for- 
ward fleet of foot, that taste which determines artis- 
tic production advances rather with circumspection, 
conditioned by a necessity which is deeply rooted. 
Fashion in recent times, baited by economic interests, 
has assumed a whirlwind tempo; and it has infected 
artistic production. 

Refined taste alongside of slight gifts a combina- 
tion which nowadays is not exactly rare must entail 
that the painter feels that the things which he sees are 
commonplace, and thereupon tries passionately to in- 
vent something that he has not seen. Thus are born 
manners of art, not very differently from fashions of 
dress. Since it has been successfully conveyed to the 
public that it must praise what displeases it, the public 
agrees to everything. 

The craftsman became an artist, but he also became 
a scholar or a manufacturer, a contractor or a virtuoso . 
The relation of virtuoso to artist is that of manner to 
style. The expression, originally meant as praise of 
exceptional ability, contains in our terminology a con- 
notation which is not devoid of reservation, doubt and 
caution. A consciously developed skill, an exhibition 
of one's ability in an endeavour to please, that is what 
we call virtuosity, in contrast to naively original 
power of creation. Most frequently the expression is 
used, without a derogatory implication, of repro- 
ductive musicians. 

That the artist tells us his name and assumes re- 



138 



ARTISTS' SIGNATURES 

sponsibility for his performance through his signature, 
is nowadays the general custom. This developed gra- 
dually in connection with the awakening and intensi- 
fication of the consciousness of one's own worth. It is 
no accident that great masters like Giotto, Simone 
Martini, and Jan van Eyck should call out their names 
in times when this was by no means the general cus- 
tom. This habit spread curiously and unevenly. As 
late as the i yth or 1 8th centuries many painters either 
did not sign at all, or did so only exceptionally Ru- 
bens, Van Dyck and Watteau, for instance. Here it 
was customary to sign, there not. Rules cannot be for- 
mulated, psychological explanations are unavailable* 
As a trade mark, as a protection against copying, the 
signature was fairly regularly employed in engraving 
and woodcut. Precisely those masters who also pro- 
duced woodcuts and engravings such as Diirer, Lucas 
van Leyden and Jacob Cornelisz. have signed their 
paintings too; if not always, yet frequently. 

A place entirely apart is occupied by the celebrated 
lines on the Ghent altarpiece. Prolix, laudatory, rhe- 
torical, like the inscription on a monument, they are 
hardly conceivable as the utterance of a painter in the 
first half of the i th century, and are therefore sus- 
pect, apart from other arguments which have been 
produced against them. 

The artist with his inner struggles and tragic con- 
flicts, not understood by the dull crowd, has in the 
1 9th century become the subject and the hero of high- 



139 



THE ARTIST: GENIUS AND TALENT 
flying poetical treatment. The conception was height- 
ened into something mystical. The painters are hard 
put to it to correspond to the expectations which an 
imagination, fed on novels, entertains with regard to 
them. In consequence there followed inevitably dis- 
appointment on one side, the pose of "genius' on the 
other. The artist who, in the popular view, only has 
the choice between being a genius or nothing, looks 
like a 'sick eagle'. The phrase was coined by Hugo 
von Hofmannsthal. 

To draw the boundary line between genius and tal- 
ent has never been done quite successfully, though 
attempts are continuously being made. If one thinks of 
a difference of degree, then it would surely be diffi- 
cult to indicate a point on the scale where genius be- 
gins. A generally valid indication of the contrast is 
probably not to be found. 

Michelangelo, Griinewald, Rembrandt are unhesi- 
tatingly described as masters of genius: indeed, to 
apply the concept of 'talent' to these creative person- 
alities would be inappropriate and almost sound like 
blasphemy. A performance which exceeds expectation, 
prevision and estimate ; sharp-edged individuality ; de- 
fiant opposition to compromise; an attitude of 'this 
cannot be otherwise' ; a spiritual obsession, which is 
akin to madness these are somewhat obvious char- 
acteristics of the genius, which operates outside the 
conventions of taste of its period, in tragic isolation. 
We speak of a melancholia ingenii and think of mental 



140 



GRADATIONS OF RANK 

struggles a g a inst inner and outer resistance. Genius 
seems to evince itself in will rather than in achieve- 
ment, in conquest rather than in rule. 

We extend to talented artists our appreciation and 
sympathy, feeling that in their praiseworthy activity 
they nevertheless remain on our level. Their vision is 
familiar to us, it contains indeed more and better 
things, but not anything fundamentally different from 
our own vision. They occupy a height which can be 
scaled. Of Rembrandt and Frans Hals it has been said, 
subtly and appositely, that in front of works by the 
latter one is seized by a wish to paint, and before 
works by the former one loses any such wish. Thereby 
a difference, which I have tried to define, is perhaps 
not badly indicated, and the incomprehensibleness 
and unapproachableness, which are peculiar to a crea- 
tion of genius, fixed within their limits. 

Among the greatest masters to whom the title of 
genius is commonly not denied, there are happy and 
harmonious natures to which my definition does not 
seem to apply ; who stand before us not as fighters but 
as victors. On the other hand we find artists who 
do not incontestably belong to the category of the 
greatest and nevertheless, through their boldness of 
imagination and strongly marked individuality, claim 
the title of genius Greco, Van Gogh or Bocklin, for 
instance. Now and then we feel indeed an inclination 
to say: a genius, but not a talent. In this way a differ- 
ence of kind between genius and talent outlines itself 



141 



THE ARTIST: GENIUS AND TALENT 
perhaps most clearly. The concept of genius indicates 
a closer alliance to the spiritual than to the visual. 
Grillparzer has said: 'Only from the union of char- 
acter with talent issues that which is called genius/ 
However, one can probably find elsewhere other, 
more convincing definitions. But the union of specific 
gifts to greatness of soul and strength of spirit may be 
peculiar to genius. 



142 



XXI 
ART AND ERUDITION 

ERUDITION has turned to art late, has perhaps 
hesitated to do so from an expectation that this 
union would not be productive of many blessings. 
The scholar, who concerns himself with art, generally 
clasps emptiness. He woos the capricious beauty, now 
in this fashion, now in that, and as a result succeeds in 
seizing the garment rather than the body. He experi- 
ments as historian, as philologist or as a scientist. He 
is a philologist inasmuch as he examines literary ac- 
counts critically, an epigraphist when he investigates 
inscriptions, a scientist when he examines works of 
art with a view to deriving chemical and physical in- 
formation from them. Naturally he is compelled, by 
such many-sided activity, to seek help from others and 
to read a tremendous lot. The excessive black-and- 
white diet is not wholesome to his eyes. Since we live 
in the age of the scientific disciplines, whose methods 
enjoy an almost superstitious veneration, he is parti- 
cularly fond of keeping company with the scientists. 
He lives in a continuous fear of losing dignity through 
his contact with something as slight as art, and of not 



143 



ART AND ERUDITION 

being admitted to the circle of the serious and re- 
spected University dons. 

History, whatever its subject, is in a difficult posi- 
tion among the scholarly disciplines. Frequently 
enough people have refused to acknowledge its true 
scientific character. Art history has apparently an ad- 
vantage over other branches of history; it claims 
superiority with some justification, especially over 
political history. The works of art are there before our 
eyes, as it were, as a preserved 'spirit of the times'. 
Only, let us not forget the words of Goethe: 

What Spirit of the Times jou call 
Good Sirs, is but jour spirit after all 
In which the times are seen reflected. 

Nevertheless, something survives that mirrors itself, 
while the historian, so far as politics are concerned, 
is thrown back exclusively upon literary tradition, 
and distils his truth mostly, not entirely without a 
political tendency out of a hundred inaccurate, if 
not definitely mendacious, and mutually contradictory 
reports. The deeds of the artists we can perceive for 
ourselves ; of the deeds of the kings an echo reaches 
us we hear mostly what the victors have let their 
scribes put down. 

The image is older than writing, and for long 
stretches of time the sole provider of information and 
evidence. 

Now it should be remembered that works of art do 



144 



THE ART HISTORIAN AND PSYCHOLOGY 
not speak they sing, and can therefore only be 
understood by listeners upon whom the Muses have 
bestowed their gifts. 

Grillparzer expresses himself somewhere approxi- 
mately thus: ' Whoever writes a history of chemistry, 
must be a chemist, and can only as a chemist become 
a historian. 3 I wonder if this does not apply to art his- 
tory ? And Schiller says: "There exists only one vessel 
for the reception of works of imagination, namely, 
imagination'. 

Since artistic activity, whatever else it may be, is 
in every case a process of emotionally spiritual nature, 
the science of art is bound to be psychology. It may 
also be something else, but it is psychology in all cir- 
cumstances. Since, however, all psychology depends 
upon experience of what has happened to oneself 
spiritually and emotionally, it follows that only an 
artistically gifted spectator can penetrate into the 
nature of artistic production. We approach thereby 
the deduction that only the practising artist is en- 
titled to judge. This deduction is erroneous. 

The productive artist is incapable of assuming, with 
regard to his own work and the work of others, the 
receptive, observing attitude from which enlighten- 
ment may be expected. He produces naively and un- 
consciously ; his experiences lie too deep for it to be 
possible for him to bring them to the light. Accord- 
ingly, his confessions in letters, autobiographies or 
theoretical expositions are valuable only as indirect 



ART AND ERUDITION 

evidence as such they possess, however, great value: 
but they must be subjected to interpretation. The 
producing artist cannot be accepted as a judge, since 
he cannot free himself from his own artistic formula, 
which is tied from the point of view of time, place 
and individuality ; and he is the less capable of doing 
this, in the measure that his own gifts are original, 
fertile and thirsting for expression. And, after all, he 
has other and better things to do than to indulge in 
philosophizing ; and he may even have a presentiment 
that knowledge does away with ability to produce, "or 
at any rate weakens it something that is noticeable 
in the individual development of many artists in 
civilized ages. And yet there is more to be learnt 
from the stammering utterances of artists than from 
the well-constructed, systematic treatises of the 
aestheticians. 

In front of art the thinkers are mostly blind, and the 
practising artists mostly dumb. There only remains 
the artistically gifted, but non-productive, spectator 
as the one who is capable of insight and deeper un- 
derstanding, and called upon to provide enlighten- 
ment. 

Artistic production and contemplation of art are 
activities which have more in common with one an- 
other than is usually assumed. The creative imagina- 
tion stands in the same relationship to the receptive 
as the cog-wheel to the cog-rail. The lover of art 
shares with the creative artist an abnormally sensitive 

146 



THE LOVER OF ART 

receptiveness. His emotional life reacts quickly and 
violently to optical signals. The ability to produce, 
the capacity to realize vision are absent in him, 
whether it be that his interior images are not distinct 
enough, or that his hand lacks skill. His is a platonic 
talent, so to speak, even a * Raphael without hands'. 
He possesses a reliable memory for visual experiences, 
and indeed not because he has, shall we say, eagerly 
learnt things by heart, but because extending his 
emotional sympathy, and living in a state of excite- 
ment which softens the wax of the tablets of memory 
he has experienced delight in contemplation. Form 
and colour are retained in his memory for decades; 
not in the sense, it is true, that he would be able to 
produce, reproduce or even describe them, but cer- 
tainly in the sense that an appearance, presenting it- 
self once again, is recognized by him and greeted as 
something familiar. Emotion and the senses have a 
much better memory than the intellect. 

It can be contested that there exists a science of art ; 
at least the concept 'science' can be defined so nar- 
rowly that art as a subject accessible to her is elimin- 
ated. 

Every scientific effort presupposes a terminology 
which covers the field of knowledge concerned, and 
through which the scholar can convey to other scho- 
lars his reasons, deductions and conclusions in an un- 
ambiguous manner. Even the greatest optimist will 
not maintain that art criticism can comply with this 



147 



ART AND ERUDITION 

demand. If I judge a work of art, and employ words 
in order to give reasons for my opinion, I am far re- 
moved from thinking that I have expressed with those 
words that which has caused my judgment. The modi- 
fications, blendings and shades in emotional life, 
which truly determined my opinion, are not to he 
seized with verbal pincers. 

An honest striving after knowledge may be recog- 
nized as science, even here, where the results remain 
questionable, and do not make themselves plain to 
reason. 

Science, in conformity with its nature, is striving 
for the goal of establishing necessity ; it hates accident. 
It must not rest content with establishing: 'this is the 
case' ; rather, it must prove, 'this is the case for such 
and such a reason'. The historian comes, however, 
everywhere upon accident, while he does not notice 
necessity which yet is his constant concern but in- 
vents and imagines it. That Rubens and Van Dyck both 
worked in Antwerp at the same time, that Schon- 
gauer had died by the time that Diirer went to Col- 
mar in search of him, these are accidents in the sense 
in which one may at all speak of accident ; and that 
such accidents have had a share in determining the 
development of artistic production is not to be denied. 
In order to avoid coming up against accident, one has 
set up the ideal of a 'history of art without names', 
being moved by the desire to recognize a course of 
events which is governed by law, and which takes 



148 




8 



w 



I 

w 

H 



O 



2 




Q 



a 

o 

O 



I 


s 



'HISTORY OF ART WITHOUT NAMES' 
place independently of the intervention of individual 
artists here and there, now and then. 

I will leave undecided the old question, debated 
by philosophers concerned with history, as to the 
greater or smaller effect exercised upon the course of 
events by heroes who 'accidentally 5 have emerged at 
a given time and in a given place. In any case, even if 
I aim at a 'history of art without names', I am yet 
bound to have classified the works of art according to 
time and place, and as far as possible to have brought 
them into harmony with biographical tradition, be- 
fore I attempt to say something about the general de- 
velopment of the will for art. To someone who knows 
nothing about Correggio's life, it might happen that 
he would assign the works of this master to the ijth 
century, and in no case would he be able to classify 
them as, in their essence, affording evidence regard- 
ing Parma and the time about 15-20. When the Mon- 
forte altarpiece by Hugo van der Goes arrived at 
Berlin, a sensitive art historian to whom 'history of 
art without names' was a desirable goal terrified 
me by his remark that the picture obviously was a 
work of the i6th century, and hence could not pos- 
sibly be by Hugo van der Goes, who, of course, died 
in the ith century. 

The boundaries of the style of a given period may 
only be drawn, it is claimed, after all that was pro- 
duced had been examined. And when would this be ? 

To the art historian there is available evidence of a 



149 



ART AND ERUDITION 

twofold kind. On the one hand we have tradition as 
contained in writings, posthumous fame, records, in- 
ventories, biographies ; and on the other, the surviv- 
ing works. The task consists in building bridges from 
one bank to the other, in bringing the documents into 
harmony with the surviving examples, Were all the 
works by Raphael lost, we would nevertheless, thanks 
to the utterances of his contemporaries, have an idea 
of his importance, his influence and even of his man- 
ner of art. Since we have a considerable part of his 
life's work before our eyes, we enrich and vary that 
idea as a result of contemplation. It becomes the task 
of style criticism to 'attribute' the works, to clas- 
sify them according to time and place, and to fit them 
into the frame of tradition represented by art litera- 
ture and records. 

The community of art scholars consists of two 
groups one may even say, two parties. The univer- 
sity chairs are mostly occupied by people who like to 
call themselves historians, and in the museum offices 
you meet the ' experts'. The historians strive generally 
from the general to the particular, from the abstract 
to the concrete, from the intellectual to the visible. 
The experts move in the opposite direction, and both 
mostly never get farther than half-way incidentally, 
without meeting each other. 

The ideal set up by the historians is called 'history 
of the spirit' . It seems indeed most desirable that all 
visible results of artistic activity be considered as ex- 



HISTORIANS AND 'EXPERTS' 
pressions of the continuously changing spiritual forces 
of humanity, in connection with other expressions to 
be found say in politics, literature, manners and cus- 
toms, and economics. It seems perfectly possible that 
the experts thus, people who take the individual 
work of art as their starting-point through fixing 
their aim high may approach that exalted goal ; and it 
is certainly to be desired that they and the historians 
should join forces. When one passes from theory to 
practice there is, however, lack of mutual confidence. 
The historians look down upon the pettifogging fuss 
with detail of the ' experts', who, for their part, ac- 
cuse their colleagues at the universities of facing the 
works of art with prejudiced views. Both reproaches 
are justified. 

On the defects peculiar to connoisseurship spiri- 
tual narrowness, subjectivity questionable from the 
scientific point of view, an inclination to be guided 
by sentiment, also uncertainty I shall yet have to 
speak, from abundant experience. As regards the 
performance of the historians, however, I may illus- 
trate straightaway the dangers connected with their 
activities. 

An aspiring scholar, intending to show himself 
worthy of university honours, plans a book on 'Art 
during the Period of the Counter-Reformation 3 . By 
assiduous reading he informs himself about the poli- 
tical, religious and philosophical aspects of the move- 
ment and acquires an idea of some of the features 



ART AND ERUDITION 

characteristic of the spiritual life of the period. He 
then turns to the works of art, and in them finds proof 
and confirmatory evidence of his conception of the 
spirit of the period, gained by reading. Thanks to the 
quantity and variety of the buildings, sculptures and 
paintings, dating from the time, there is much that 
can be found. And he who seeks, and knows from 
the start what he is going to find, cannot possibly go 
wrong. Moreover it is in the nature of a work of art to 
speak ambiguously, like an oracle. 

The freshness of observation, the capacity for an 
unprejudiced reception of artistic impressions, may 
well be endangered through such efforts. 

If the scholar takes the concept of 'the Baroque' as 
his starting-point he will feel inclined to fit all forces 
stirring in the ijth century into a scheme, and this 
cannot be done without applying violence. Even Rem- 
brandt is forced into the bed of Procrustes. To be sure 
the searching eye, guided by a knowledge of history, 
can bring hidden things to light, but the mobility, re- 
ceptiveness and lack of prejudice of the contempla- 
tion of art are as a result narrowed down and suffer 
grievously. Blindness to everything that slips through 
the wide meshes of the hunting net of expectations, is 
inevitable. 

The expert is at pains to complete and differentiate 
the materials for study ; the historian has little interest 
in the increase of the works of art which he has to con- 
sider, particularly as he draws his conclusions much 



152 



QUESTIONS OF METHOD 

more easily from ten than from a hundred pieces of 
evidence. One can perfectly well write an excellent 
book on Raphael by exclusively noticing the narrow 
circle of absolutely certain works, and without at- 
tempting to effect additions by means of style criti- 
cism, paying more heed to the purity than to the com- 
pleteness of the entire picture. But the clearer the 
notion at which the biographer arrives, the more 
forcibly is it going to act as a magnet, which attracts 
some examples out of the troubled mass of question- 
able works. Whoever knows something, does also 
recognize it, and whoever does not recognize, reveals 
thereby that he does not know. The historian be- 
comes, whether he wants it or not, an expert. 
Throughout the literature of art history it can be 
traced, how scarcely anyone of the scholars, inclined 
towards the history of the spirit, has on principle 
evaded determination of authorship. And, in view 
of the performance of Carl Justi it is permissible to 
express the hope that the difference between his- 
torian and expert may be bridged over. 

A fundamental evil from which art-historical erudi- 
tion suffers, seems to me to spring from the method 
which, put to good use in other disciplines, has mis- 
takenly been associated with the contemplation of 
art. Art history, which is frequently described as 
young, or even as still in the nursery, appears to the 
typical university mind as a playful child in need of 
education, who must first of all learn seriousness from 



ART AND ERUDITION 

the grown-up branches of learning. I remember a very 
learned colleague once saying: "Unfortunately I have 
no time, being busily engaged on other important 
work; otherwise I would take up the Hubert and 
Jan van Eyck question and settle it.' Such optimistic 
determination may, in other domains, be conducive 
to untiring, systematic and fertile performance ; with 
regard to art, it proves a complete failure. The method 
which we employ if indeed the word method be 
here applicable must be won from the object ; that 
is, irrational art. One should look at works of art 
without any intention of deriving knowledge, and re- 
joice if sometimes, as of itself, a confirmation or 
enrichment of our knowledge comes in a flash ; and 
one should not approach them with the determination 
to solve a problem. One must let them speak, one 
must converse with them, but one must not interro- 
gate them. To an inquisitor they refuse any informati on . 
A continuity of knowledge, such as elsewhere fer- 
tilizes the whole of scientific discipline, is almost 
completely unavailing with regard to art. It is true 
that art historians read a lot, copy or assert the re- 
verse of that which they have read which is not 
much more difficult than to copy but if one en- 
quires by whom and in what manner our knowledge 
has been enriched, one usually comes upon the lover 
of art who, independently of predecessors, has naively 
faced the works of art. Everyone must here start 
scratch and be able to forget what he has read. 



XXII 
THE STANDPOINT OF THE SPECTATOR 

IF I look at a work of art of the i th century, I am 
incapable of eliminating what I have perceived in 
works of later date, and must inevitably judge from 
the standpoint of my own time. Any attempt to avoid 
this 'injustice* could at best be made by a dry-as- 
dust spectator, whose *just 5 appreciations would be 
worthless. If we concern ourselves intensely with the 
art of the past, we acquire some practice in displacing 
our standpoint and in being more or less mobile in 
adopting now this, now that, standard. The art his- 
torian resembles at times the traveller who has been 
everywhere, is knowledgeable wherever he goes, but 
is nowhere at home. The fixed and inalienable point 
to which we always return, is and remains our home 
in time and space. 

Every age receives fresh eyes. The Italian sees 
things differently from the German. I see differently 
from you, see differently to-day from yesterday. 
From this one might deduce that the art of days that 
are gone is strange, dead, incomprehensible and in- 
accessible to us. And this is the case in a higher 



THE STANDPOINT OF THE SPECTATOR 
degree than is usually admitted. The historian, so to 
speak, corrects his natural sight as well as may be 
through spiritual spectacles; from affectation the 
layman pretends familiarity, where shyness and un- 
easiness predominate. He who would be able to ex- 
pose the real feelings of museum visitors, would 
acquire a terrifying idea of the ineffectualness of 
ancient art. 

Everyone begins by understanding and enjoying the 
production of his own time. At least, this is the 
healthy, normal and natural start. Our respect for 
celebrity causes us then to rise to an understanding 
of the old masters. More than one collector has ap- 
proached Primitive art in stages. Adolf Thiem for in- 
stance, a gifted and independent lover of art, bought 
pictures by Menzel and Daubigny when he was forty ; 
by Van Dyck, when he was sixty: by Memling and 
Dirk Bouts when he was seventy. And many collec- 
tors have travelled in the same direction. 

The sense of strangeness with regard to the works 
of past ages a sense which even the most assiduous 
historian never quite overcomes is to this extent an 
advantage, that a certain distance makes us observe 
that which is characteristic of time and place more 
distinctly than he who is close up. When we arrive in 
China we are, it is true, unable to tell one Chinaman 
from another, but that which is Chinese is all the 
clearer in our eyes. Without any question, the work 
of Botticelli seemed to the contemporaries of the 



AESTHETIC NIHILISM 

master more natural and less severely stylized than to 
us. One can raise, but not answer, the question 
whether this has brought about an increase or de- 
crease of artistic value. 

We change the standards and thereby arrive at re- 
sults which are different from one another. All com- 
parisons may become fertile and instructive. One may 
compare Jan van Eyck with Roger van der Weyden, 
or with Manet. In every case it is useful to arrive at 
clearness as to the standard which has been applied. 

A demand is often made with an intention to in- 
struct, and seeming to proclaim highest justice that 
every artist is to be measured by applying to him his 
own standards; but this demand, strictly speaking, 
argues thoughtlessness. It is as if one were to measure 
a yard with a yard. 

The historian considers it as a duty of neutrality to 
extend sympathy to every expression of art, and up to 
a point he succeeds in this. He should, however, at no 
time forget that a universal receptiveness, carried to 
an extreme, leads to aesthetic Nihilism and, indeed, 
a decadence of taste among the learned scribes can 
frequently be observed. 

In surveying artistic activity from one point of van- 
tage namely, our home in time and space we gain 
a general picture which, it is true, is distorted in 
point of perspective, but homogeneous. 

Each generation chooses its favourite from amongst 
the succession of old Masters, in conformity with its 



THE STANDPOINT OF THE SPECTATOR 
method of vision and its own production, always in- 
clined boldly to contradict the previous generation. 
Enthronement does not take place without a tumult 
and not without making the seat free by deposing 
someone else. Bocklin has thus been deposed in 
favour of Manet, and Velazquez in favour of Greco. 
With the passing of time the exaggerations of excited 
propagandists are smoothed down. But we never re- 
vert to the opinions of our fathers and ancestors. 

Drastic instruction regarding the change of taste is 
obtainable by throwing a comparative glance at prices 
paid for pictures at sales by auction. In the year 1 85-0, 
at the sale of the property left by the late King Wil- 
liam II of Holland a sale which spelt fatality for Hol- 
land as a collecting country the following prices 
were realized. 

JanvanEyck, The Annunciation fl. 5-375* 

(now National Gallery, Washington) 
Jan van Eyck, The Lucca Madonna fl. 3000 

(now Staedel Museum, Frankfurt) 
Jan Both, Italian Landscape fl. 10,400 

Jacob Ruisdael, Landscape fl. 12,900 

Hobbema, Landscape fl. 27,000 

Andrea del Sarto fl. 30,35-0 

Leonardo da Vinci fl. 40,000 

Jean Kobell fl. 4900 

Koekkoek fl. 35-00 

Carlo Dolci fl. 5-900 



FLUCTUATIONS OF TASTE 

These are particularly startling examples. A Koek- 
koek brought more than the Lucca Madonna by Jan 
van Eyck. 

Within a generation judgment varies, but not in the 
degree that can be expected from the comparison of 
written and oral utterances. Rembrandt is greater 
than Gerard Dou, Jan van Eyck greater than Carlo 
Dolci. Nobody will to-day contradict such a state- 
ment ; only everyone will feel shy of making it. The 
wide domain of banality is also the domain of that 
which has general validity. 



XXIII 

ON THE VALUE OF THE DETERMINATION OF 
AUTHORSHIP 

IF the determination of the authorship of an indi- 
vidual work of art most certainly is not the ulti- 
mate and highest task of artistic erudition ; even if it 
were no path to the goal: nevertheless, without a 
doubt, it is a school for the eye, since there is no for- 
mulation of a question which forces us to penetrate so 
deeply into the essence of the individual work as that 
concerning the identity of the author. The individual 
work, rightly understood, teaches us what a compre- 
hensive knowledge of universal artistic activity is in- 
capable of teaching us. 

Goethe's works were published under his name; 
nothing is attributed to him or declared not to be by 
him. One might imagine that the understanding of 
Goethe's language, spiritual nature and development 
would be greater than it is, if scribes would have had 
gradually to put together his asuvre. They would 
scarcely have performed their task with complete suc- 
cess, but they would have learnt a good deal as a re- 
sult of their efforts. 



160 



POWER OF PERSONALITY 
Over long stretches of time the determination of 
authorship seems to be impossible. Many produc- 
tions, notably of architecture, can be fixed in time 
in the case of architecture the localization is always, 
and in the case of sculpture often, available but they 
are not recognized as the expressions of individual 
talents. Anonymity is a symptom of deficient know- 
ledge, even if the deficiency often is inevitable. 
Strictly speaking, every work of man is the product of 
a personality with qualities, existing once and unique. 
Whoever arrives in China, thinks at first that all 
Chinamen look alike; it is only gradually that he 
learns to distinguish individualities. A similar experi- 
ence is that of the connoisseur who approaches the 
'dark' periods. Admittedly a personality reveals itself 
according to the period more or less definitely in its 
activity. The ultimate, the most fruitful question, 
even if it cannot be answered, is and remains that 
which concerns personality. 

Fairly frequently one hears the plausible-sounding 
objection that we know that there were hundreds of 
painters, yet all the existing works are divided up 
amongst comparatively few names. A statistical com- 
putation may serve as a defence against these misgiv- 
ings. It is chiefly the prominent works that have sur- 
vived, and of the surviving ones it is again the best 
ones that are collected, exhibited in museums and 
accessible to art lovers. Finally, I possess hundreds 
and hundreds of photographs of Netherlandish pic- 

161 



THE DETERMINATION OF AUTHORSHIP 
tures of the i th and 1 6th centuries which I cannot 
attribute, of which scarcely two seem to be by the 
same hand. These nameless pieces mostly are value- 
less and devoid of character. From this I think one 
may conclude that the many painters who are un- 
known to us have mainly produced unimportant 
things j and that, on the other hand, the better works 
with which the determination of authorship concerns 
itself are due to relatively few artists. This calcu- 
lation applies to Netherlandish and German painting 
of the i th and 1 6th centuries ; it may not be valid, or 
is perhaps valid in a lesser degree, for other countries 
and other periods. 



162 



XXIV 

ON THE OBJECTIVE CRITERIA OF 
AUTHORSHIP 

SINCE the expert, especially when he assumes the 
part of the didactic writer, experiences the diffi- 
culty or impossibility of convincing others of the 
truth of his verdicts, he attaches great importance to 
objective characteristics. Just as a tired swimmer 
breathing a sigh of relief welcomes firm ground under 
his feet, so does the expert react to inscriptions, docu- 
ments and objective data of different kinds. Without 
being able to swim he would, to be sure, not have 
been able to reach land : on long distances the water is 
so deep that it is necessary to swim. 

I class among objective criteria: 

1 . Signatures and monograms, which give or hint at 
the name of the master. 

2. Documentary information, agreements, inven- 
tories, catalogues which are approximately coeval 
with the works of art described in them ; also the 
references to existing works in early writings, for 
instance the Lives of Vasari and Karel van Mander. 

3. Measurably similar forms which, familiar to us 
from authenticated, signed or recognized works, 

163 



OBJECTIVE CRITERIA OF AUTHORSHIP 
reappear in those to be attributed, say in the 
architecture or ornament. Under this heading is 
to be introduced the similarity of form of the ear, 
the hand, the finger nail, so strongly emphasized 
by Morelli. 

As regards the signatures, it is to be noted that their 
evidence may be misleading. The name may have been 
added later, bona fde or mala jide, the genuine may 
have been removed and replaced by a forged one. A 
test with spirits which are resisted by the original 
layer of colour has not infrequently led to the re- 
sult that the signature vanishes. Graphological tests 
assist the chemical ones. But even the 'genuine' in- 
scription that is, the one which was added by the 
master himself immediately on the completion of the 
work must not in all circumstances be regarded as 
valid and binding. A copyist may have taken it over 
from the archetype. When production was organized 
in a workshop a circumstance always to be borne in 
mind the master occasionally provided pictures, 
wholly or in part painted by his assistants, with his 
signature, as it were with a Trade Mark. ' Genuine' 
Bellini signatures occur on pictures which evidently 
are the work of other masters. Nobody troubled very 
much in the past about such a thing as spiritual own- 
ership. Even the most venerable of all inscriptions, 
the celebrated stanza of four lines on the Ghent altar- 
piece, has very recently, for powerful reasons, been 
declared suspect. 



164 



VALUE OF RECORDS 

The inscription may be false, and its statement 
nevertheless accurate so that in this case, a rare one 
no doubt, the objective criterion is not even decisive 
in a negative sense. 

Reliability is not to be expected from statements in 
inventories which, incidentally, contain authors' names 
only in exceptional cases, and from early writings. In 
the Prado at Madrid there is a Madonna y with the 
solemn information on the reverse of the panel to the 
effect that the city of Louvain in 1^88 offered this 
work by Johannes Mabeus (Jan Gossaert, called Ma- 
buse) as a present to King Philip EL I have not let this 
document prevent me from transferring the Madonna 
picture to Bernard van Orley. It is true that if it had 
been possible to fit this picture in among the works 
by Gossaert, then I would have welcomed the state- 
ment referred to as a valuable piece of evidence. 

The inventories of princely galleries such as those 
of Margaret of Austria, Vicereine of the Netherlands, 
or of King Charles I of England, and also, say, the 
notes made by Marcanton Michiel in North Italian 
houses are to be utilized sceptically and to be taken 
seriously only to the extent that facts derived from 
style criticism do not contradict them. 

Again, as to the measurably similar forms, the ob- 
jectiveness of this criterion is questionable, and its im- 
portance as a clue is being exaggerated. 

Enthusiasm is a state of mind natural to the lover of 
art indeed, to him almost something normal. But it 



OBJECTIVE CRITERIA OF AUTHORSHIP 
does produce a confusing effect. Morelli, who called 
himself Ivan Lermolieff, has written some notable sen- 
tences about Otto Miindler, whom he knew person- 
ally and valued highly. Miindler, says Morelli, relied 
upon his memory and his intuition ; he made his de- 
cisions on the strength of the accidental impression 
produced by the whole. Enthusiasm sometimes lets 
down the critic badly. This may be read in the Intro- 
duction to Ivan LermoliefFs volume Kunstkritische Stu- 
dien in den Galerien zu Munchen und Dresden (1891). The 
Italian, masquerading as a Russian, emphasizes the 
shortcomings of his predecessors in order to recom- 
mend his analytical and 'scientific' method as a pro- 
gress, an antidote. Again the ominous word 'acciden- 
tal' is being used. I do not know why the impression 
of the whole of the picture should depend more on 
accident than the impressions of the individual por- 
tions, on the basis of which Morelli claimed and be- 
lieved to judge. 

It could probably be proved statistically that Mor- 
elli by applying his method which he, not exactly 
logically, described as an 'experimental' one made 
as many mistakes as Miindler depending on intuition. 
Nay, he would have made even more mistakes if he 
had applied his method consistently. The decisive fac- 
tor is something that he too owes to intuition ; and if 
we look closer, we shall find that he has utilized the 
much-praised method the observation of measurably 
similar forms, notably the ears, the hands, the finger- 



166 



THE THEORY OF MORELLI 

nails less for the purpose of arriving at a verdict, 
than in order to provide evidence subsequently. He 
points to the individual forms in order to convince 
the reader of the justness of his attributions: but he, 
like every successful expert, has formed his opinion 
from the 'accidental' impression of the whole picture. 
He had a presentiment of this, and has even hinted at 
it, when on one occasion he assesses the value of his 
method fairly accurately as being an ancillary de- 
vice, a means of checking. 

The criterion of similarity of form is completely 
unavailing, once we are faced with the task of differ- 
entiating original from copy thus to answer a ques- 
tion which, in the practice of connoisseurship, is a 
particularly frequent and burning one. 

The verdict may be accurate, although the reasons, 
the attempt to present it as a compelling truth, estab- 
lished by analysis, appear misguided. It is noticeable 
that gifted experts in particular, who make their 
decisions with inner certainty, have little inclination 
to provide 'proof 3 : they probably feel rather like 
Nietzsche, who said 'Am I then a barrel, carrying my 
foundations with me?\ False attributions are often 
presented with an excessive display of acuteness, and 
of arguments which sound irrefutable. False Raphael 
pictures are accompanied by whole brochures. The 
weaker the inner certainty, the stronger the need to 
convince others and oneself by lengthy demonstra- 
tions. 



167 



OBJECTIVE CRITERIA OF AUTHORSHIP 
Enthusiastic lovers of art at the same time mere 
amateurs have contributed most and in the best 
fashion towards artistic reconstruction: they were, 
however, also exposed to the danger of making mis- 
takes. Coldly analytical scholars make fewer mistakes ; 
they perform, however, less in the way of positive 
perception; they discover less, with weaker flair. 
Morelli himself was, after all, an amateur in the best 
sense of the word ; as a scholar he was rather affected. 
Morelli has ably provided psychological reasons for 
his method. Above all the painter renders the human 
figure in pose and movement, as well as the face, 
more particularly mouth and eyes, under the stress of 
emotional tension, in order to convey his vision to the 
spectator. In so doing he penetrates relatively deeply 
into the complexity of that which is individual ; but 
he lapses into convention and routine when he draws 
parts of the human body, such as the ear or the hand, 
which seem to be of secondary importance as bearers 
of expression. The hand speaks more through its 
movement than through its shape. Moreover, pre- 
cisely ear and hand are complicated formations, mas- 
tered by the draughtsman only with difficulty : so that 
the artist was tempted, by clinging to a formula, to 
evade the trouble of studying the given form in each 
case. Even great portrait painters of the i yth century, 
Van Dyck for example, have paid little attention to 
the individual shape of the hand. 

For other reasons the critic of style may be recom- 

168 



THE CAST OF DRAPERY 

mended to observe the drapery folds. The painter 
steps on to a domain of comparative freedom when, 
in conformity with his temperament and his condition 
of spirit, he lets the pliable textile undulate, ripple, 
break, swell, roll up, swing and flow out. 

Costume, more particularly when elaborate and 
idealistic, made expression by means of sonorous 
melody possible for the artist. In studying the cast of 
drapery we almost become graphologists, and can 
deduce the personal temperament and even the mo- 
mentary mood of the author from the flow of writing, 
from the arid, angular, measured, sober or exuberant, 
rushing, dramatically mobile and extravagant play of 
line. Evidence of the expressive force which can be 
breathed into the material of garments is provided by 
Griinewald and Hugo van der Goes. 

The graphologist takes as his starting-point the fact 
that a writer must leave letters as much of their given 
form as legibility demands, but that he, free from this 
compulsion, is capable of spreading himself in the 
flourish, and in so doing expresses character, caprice 
and mood visibly. Also in the work of art, you can 
tell the difference between flourish and conventional 
sign. In the case of materials, costume, cut and sewn 
in such and such a fashion, resembles the letter ; the 
play of folds, fluttering sashes, ends and ribbons re- 
semble the flourish. 

Mr. Berenson, who began as a grateful pupil of 
Morelli, has in a fragmentary article, published in 

169 



OBJECTIVE CRITERIA OF AUTHORSHIP 

1902, under the title Rudiments of Connoisseurship, ar- 
ranged the 'tests' under three headings, according as 
they are of use for the determination of authorship : 

First Group: Ear, Hand, Drapery fold, Landscape. 

Second Group: Hair, Eye, Nose, Mouth. 

Third Group: Cranium, Chin, Structure and Move- 
ment of the Human Figure, Archi- 
tecture, Colour, Chiaroscuro. 1 

This scheme is notable and is based on accurate 
thought: but it must not be applied as having univer- 
sal validity. Tested in the practice of a connoisseur 
who has pre-eminently concerned himself with the 
Italian Quattrocento, it will, if used in connection 
with other periods, other manners of art, be partly 
unavailing. A Netherlandish painter of the 1 6th cen- 
tury may treat the landscape more variably, may in- 
vest it with more expression than architecture, so 
that, so far as he is concerned, architectural form be- 
comes usable as a piece of evidence rather than land- 
scape. 

It must always be nicely calculated in each case 
whether there is predominance of habit and routine, 
or of observation of nature, individualizing charac- 
terization and expression of emotion. The more tense 
the will to achieve artistic form, the greater the varia- 
tion. 

1 Bernhard Berenson, The Study and Criticism of Italian Art, second 
series, p. 144. 



170 



THE INTUITIVE VERDICT 

The paradoxical idea that the master Is recogniz- 
able where he has least drawn upon his force of ex- 
pression, partly holds good. Argumentation providing 
reasons may successfully refer to the similarity of the 
ears ; the act of intuitive arriving at a verdict springs 
from the impression of the whole. In the one case a 
master betrays himself; in the other he reveals himself. 



171 



XXV 

ON INTUITION AND THE FIRST IMPRESSION 

EVEN if attention deservedly goes to all the cri- 
teria which, with more or less justification, are 
described as the 'objective*, seemingly scientific ones, 
and occupy a space disproportionately large in writ- 
ings on art, decision ultimately rests with something 
which cannot be discussed. To be sure, when we 
come upon the concepts of intuition and self-evidence 
and every statement based upon style-criticism ul- 
timately reaches and is wrecked by these concepts 
we resign as scholars and even as writers. A purely 
emotional sense of conviction comes into play, and 
pushes itself into the place of terse deduction. Per- 
haps every verdict, formulated on grounds of style 
criticism, is nothing but a supposition ; perhaps only 
probability may be arrived at along this path. 

Style-criticism inevitably reckons with probabili- 
ties, builds up hypotheses. In order to make fruitful 
use of such sensitive and delicate means, it is neces- 
sary to possess imagination and sincerity, a quality 
which is often unavailing. The vain desire for a c cer- 
tain* result of one's studies is often stronger than the 



DEGREES OF PROBABILITY 

love of truth. The scholar is able to provide reasons 
possessing a certain amount of probability for a c de- 
termination* ; proceeding, however, in the next chap- 
ter, he treats his supposition as an ascertained fact, 
and builds upon it further 'determinations'. We must 
insist that one should remain conscious of the degree 
of probability in each case, and proclaim it. 

A hypothesis is something different from a sup- 
position: it is an experiment. One may, tentatively, 
suppose even that which is improbable to be true, 
and draw deductions therefrom. A supposition gains 
in security if it can support tests of weight. 

One should retain, and steel, one's courage of sub- 
jective opinion, but one should also sceptically and 
coolly put this opinion to the test. As in the case of a 
woman beloved, one should honour naivete, but not 
let oneself be ruled by it. 

The way in which an intuitive verdict is reached 
can, from the nature of things, only be described in- 
adequately. A picture is shown to me. I glance at it, 
and declare it to be a work by Memling, without hav- 
ing proceeded to an examination of its full com- 
plexity of artistic form. This inner certainty can only 
be gained from the impression of the whole ; never 
from an analysis of the visible forms. 

This decision from feeling depends upon compari- 
son, but not so much upon the recollection of such 
and such an authenticated signed or universally ac- 
cepted work, as rather on an unconscious comparison 



173 



ON INTUITION AND FIRST IMPRESSION 
of the picture to be ascribed with an ideal picture in 
my imagination. To gain, retain, refine and revive this 
ideal picture is the important thing, and hence it is 
advisable to devote as much time as ever possible to 
the contemplation, in full enjoyment, of the best and 
the authenticated works by a master; and on the other 
hand to devote little time to the problematic ex- 
amples. Many experts act inversely, to their own de- 
triment: they waste time and strength in examining 
dubious and insignificant pictures, and run the risk of 
confusing their taste and distorting their standards. 

An ideal must not be fossilized: it must ever be 
kept capable of enrichment and change. It comprises 
not only such works as have been seen, but also such 
concealed possibilities as are contained in the gifts 
of a master. The idea of a master's capabilities be- 
comes often all too early cut-and-dried: it should 
never be regarded as unchangeable. 

If one has made a mistake which is something that 
occasionally happens even to the gifted connoisseur 
then one must radically and decisively evacuate the 
falsely judged work of art from one's memory and 
submit to a purge in the guise of the contemplation 
of indubitable works by the master. I remember the 
tragic case of an excellent and conscientious expert 
who once made a mistake. He was unable to summon 
sufficient courage and self-control to confess his mis- 
take to himself; he searched for 'proofs' of his false 
attribution and as a result ended up in a false position 



FASCINATION OF STYLE CRITICISM 
with regard to the master, to whom he once by mis- 
take had assigned something. As a result of this one 
mistake, which in itself was no disaster, he was bereft 
of pure and clear notion, and his judgment, at least so 
far as this master was concerned, lost certainty. And 
that was a disaster. 

One should avoid as far as possible to link up an 
attribution based on style-criticism with another such 
attribution in other words, to forge chains since, 
of course, the risk of mistake is always there, and 
steps must be taken in advance to ensure that error 
does not produce error. A return to the secure start- 
ing-point remains imperative, to a centre from which 
attributions issue like rays. 

Intuitive judgment may be regarded as a necessary 
evil. It is to be believed and disbelieved. Every sudden 
idea, however vague, may serve as basis for a fruitful 
hypothesis ; only one must be ready to drop it as soon 
as it has proved itself incapable of sustaining weight. 

In this mixture of bold initiative and equally deter- 
mined resignation, of enthusiasm and scepticism, lies 
the fascination exciting, keeping the spirit fresh 
and mobile of work on the basis of style-criticism. 

Intuition resembles the magnet needle, which 
shows us our way whilst it oscillates and vibrates. 

In the case of some masters it is easy to find the 
place of security, where an idea of full complexity can 
be gained and the ideal given shape in our imagination. 
In the Hospital at Bruges we not only see works by 



ON INTUITION AND FIRST IMPRESSION 
Memling, but experience his artistic activity, walk in 
his footsteps, measure the possibilities and limitations 
which were contained in his gifts. And it is with a 
similar profit that we leave the Frans Hals Museum at 
Haarlem, since there too the growth, the changes, the 
direction of an individual development, and the scope 
of a master are confided to us. Somewhat recklessly I 
venture to claim that we learn to paint like Memling, that 
is, to form the same visions as he. This imaginary 
pupilage, which naturally has nothing to do with 
realization for, of course, we do not become cap- 
able of successful forging obtains for us the inner 
certainty with which we decide: this must be by 
Memling, or that cannot be by him. 

If someone tells me that he owns a Still Life by 
Frans Hals, signed and dated 16^0, I conjure up 
without ever having seen a Still Life by Frans Hals an 
idea which serves me as a standard as to whether I 
accept or reject the picture when it is shown to me. 

The work of art which I attribute, and my ideal pic- 
ture of the master whose name I pronounce, stand to 
each other in the relationship of lock and key. The 
expert's weapon and possession are less photographs, 
books, or a dictionary of characteristics, than con- 
cepts of visual imagination, gained in pleasurable con- 
templation and retained by a vigorous visual memory. 
The capacity of memory is limited. Even a Wilhelm 
Bode, whose gifts as a connoisseur were of an unex- 
ampled manysidedness, was unavailing in many direc- 

176 



THE EXPERT AT WORK 

tions. The reliable and successful experts are specia- 
lists. One must summon courage to say 'I do not 
know' and reflect that he who attributes a picture 
wrongly reveals his ignorance of two masters of the 
author, whom he does not recognize, and of the 
painter whose name he proclaims* 

You cannot tell by the look of a verdict, based on 
style-criticism, whether it is correct or not. But with 
time its healthiness reveals itself by its capability of 
reproduction. A false verdict shows itself to be sterile. 
With the true something could be done, it was pos- 
sible to build on it, and usually it was subsequently 
confirmed by knowledge gained along other paths, 
and from a different quarter. 

The first impression is deeper than all subsequent 
ones, of different kind and of decisive importance, 
The first contact with a work of art leaves a profound 
imprint, if only because it is connected with excite- 
ment. The receptiveness of the eye is heightened by 
that which is new, strange, unexpected, different. And 
if the contact be repeated, it is the moment of recog- 
nition which produces the strongest effect. It seems 
therefore advisable to look at a picture periodically 
for six seconds rather than once for a whole minute. 
Inexperienced beginners, in order to study a picture 
thoroughly, stare at it so long that they no longer see 
anything: that is, no longer receive the impression of 
something arresting. The eye tires if it stays too long 
in the same place ; that which is peculiar and specific 



177 



ON INTUITION AND FIRST IMPRESSION 
assumes more and more the colour of that which is 
normal and incapable of being otherwise : the grace 
and advantage of the first impression are lost. Young 
art historians, who assiduously and intensively busy 
themselves with one master, without having seen 
much by others, lose the eye for the outline of their 
hero. Did not Montaigne in his wisdom think it 
worth while to note: 'When we want to judge the 
tonality of colour of a scarlet cloth, we must let our 
glance glide over it quickly and repeatedly.' 

Every verdict on art is the result of a comparison, 
mostly made unconsciously. A heightening of the 
impression is obtained by means of contrasted effect. 
If I have seen a picture by Gerard Dou and then look 
at Rembrandt certain qualities of Rembrandt emerge ; 
if, however, from Titian I turn to Rembrandt, I re- 
ceive a different impression. To experiment in this 
fashion is advisable as an exercise. The greater the dis- 
tance as regards time, place or individual character 
between the works of art which we confront with 
another, the more distinct is the impression of that 
which pertains to time and place ; the closer they are 
to one another, the easier does it become to observe 
subtle differences, to draw, say, the dividing line be- 
tween the master and his skilful imitator. 

He who knows but one master knows him insuffi- 
ciently. This inadequacy is often enough to be noted 
in works denoting a writer's debut, and particularly 
in theses for a doctorate. 



178 



XXVI 
PROBLEMS OF CONNOISSEURSHIP 

CHARLATANISM, the professional malady of ex- 
perts, springs from the unstable nature of artistic 
judgment. The moment I formulate a statement in a 
way which goes beyond inner certainty, honesty 
begins to waver. 

Dealers and collectors are not served by supposi- 
tions; they demand a positive decision. The expert 
not infrequently gets into a difficult position, since 
more is expected of him than he can honestly give. 
Let us say that he has recognized a picture as a work 
by Rembrandt. Out of confidence in him somebody 
acquires it at a high price. Later he arrives at the con- 
viction that he has made a mistake. Even if his love of 
truth now overcomes his vanity, he is yet reluctant to 
harm someone who has believed in him. An expert of 
determined character did once, in such a situation, 
take over the doubtful picture at his own expense, but 
declared another time coldly and resolutely that the 
financial risk had to be borne by the person who had 
consulted him. Most people have less character; they 
do not confess their mistake or they try to confuse 



179 



PROBLEMS OF CONNOISSEURSHIP 
the hard facts, more particularly as they know from 
experience that their clients never forget a financial 
loss, whereas grateful memory is developed on a sing- 
ularly slight scale. 

Every work of art has a financial value, which 
largely depends on the view taken of its authorship. 
This value also depends on its artistic value, which is 
difficult to assess, and in any case can be sent consider- 
ably up or down through the verdict of the expert. 
The expert comes up against financial interests and 
gets regrettably caught up in them. 

I had an excellent friend who actually committed 
himself to the view that the science of art could be 
taken in hand seriously only after all works of art had 
become public property. 

Since nobody can be called to account or produce 
proofs, since everything depends upon confidence and 
blind faith, it is authority which is demanded, claimed 
and striven for at times even created artificially. 
The dealers have a natural interest in proclaiming the 
infallibility of the science, to which they appeal. A 
pearl merchant will always contend that it is child's 
play for him to tell genuine pearls from false. 

The expert appears to the layman to be a magician 
and a worker of miracles. He thinks this part suits him 
and he becomes accustomed to indulge in the attitude 
of the conjurer. He is inclined to assert himself 
through rhetorical turns of speech; exclaiming, for 
instance, 'I put my hand in the fire that this is so 5 , or 



180 



DIFFICULTIES OF THE EXPERT 
else, 'Whoever does not see this, must be blind'. At 
times he tries to provide a basis for, and to strengthen 
his authority through, the appearance of heavy intel- 
lectual work and laborious research since you may 
take credit for your industry, but not for your gifts, 
and many people like to take credit. 

At the same time, let us be lenient towards human 
weaknesses. Satisfaction of his vanity, the exalting 
consciousness of authority and the power that goes 
with it, must compensate the expert for much that is 
disagreeable in his questionable profession. Honest 
recognition of positive performance hardly ever 
comes his way, least of all from his professional col- 
leagues, who quote him only when they contradict 
him. Anything true that to-day he has been the first to 
find, is already to-morrow common property and at 
everybody's disposal. Mistakes survive, on the other 
hand, under his name and call up memories of him. 
Dubious things, which he was unable to attribute, are 
over and over again submitted to him with a silent re- 
proach; while the works to which he, without being 
contradicted, has assigned such and such a name, dis- 
appear without further ado, and without earning for 
him any gratitude. 

The quality of the works of art which drift about in 
the market is declining. The number of the dealers 
and agents who want to live by the sales in the art mar- 
ket grows continuously. The difference in value be- 
tween a picture by Rembrandt and one by Ferdinand 



181 



PROBLEMS OF CONNOISSEURSHIP 
Bol is increasing. The hunt for valuable things be- 
comes ever madder and more relentless. Connoisseur- 
ship becomes more and more specialized, takes on the 
character of a mystery, so that even a highly regarded 
and experienced dealer can no longer say to his cus- 
tomers: 'I regard the picture as a work by Titian and 
assume the guarantee ; there is no need for an expert 
opinion'. All these are circumstances which contri- 
bute to an increase in the power of the expert, and 
to the danger of misusing this power. 

'Expertizing' is felt to be mischievous, but as 
things are it is bound to be ineradicable and a neces- 
sary evil. The need to establish whether a picture 
really is the work of Rembrandt by consulting an 
authority, a disinterested and conscientious writer of 
expert opinions, appears urgent. The difficulty lies in 
the regrettable uncertainty as to who is a well-in- 
formed and honest writer of expert opinions. All sug- 
gestions made, and measures taken, in order to com- 
bat the degeneration of ' expertizing' have done more 
harm than good. Museum officials have thus in many 
places been forbidden to give written opinions, 
which means that a number of the best experts have 
been excluded, and the field has been thrown open to 
unofficial, professional writers of expert opinions. As 
a result the average standard of truth of the expert 
opinions has declined. The official may pronounce 
himself only verbally. The verbal opinion is naturally 
formulated with less sense of responsibility than the 



182 



THE FUTURE OF EXPERT OPINIONS 
written opinion; and, moreover, it is usually dis- 
torted when subsequently handed on. The French in- 
stitution of the experts as government officials has 
certainly proved its worth in the administration of 
justice, since a financial guarantee is linked with an 
attribution put forward ; but it has not succeeded in 
asserting itself against free, un-attached, specialized 
connoisseurship ; it has been unable to replace it or 
eliminate the latter. 

There is no choice but optimistically to rely upon 
the fact that ignorance and unscrupulousness will 
gradually be discovered in the circles of the collectors, 
and that the dealers as a result will be induced to 
exercise circumspection in the choice of the writers 
of expert opinions. 

The complaints regarding frivolous and untruthful 
expert opinions are all too justified. They have caused 
a reaction, so that timorous minds nowadays go to ex- 
tremes in judging negatively or with reserve. The 
people concerned say 'no' in order not, at all events, 
to be confused with the 'yes-men'. Now prudence is 
not only the mother of wisdom but also the daughter 
of ignorance. What must be done is to steer the right 
course between the rocks of a conciliatory com- 
plaisance on the one hand and a negative attitude, on 
principle, on the other. 



183 



XX VU 

THE ANALYTICAL EXAMINATION OF 
PICTURES 

I HAVE compared intuitive judgment to swimming 
in a deep river, and have admitted that occasionally 
we, even in shallow waters, tread on firm ground, 
namely, as soon as we dissect the work of art. You 
cannot explain a witticism without murdering it. And 
the position is the same with regard to the work of 
art. Nevertheless, the fear of elusive, mysterious and 
incalculable intuition is over and over again con- 
ducive to reconsideration, testing and checking by 
means of dividing and dissecting visual action. By 
establishing the causes of the total impression 
through analysis we, in any case, enrich our know- 
ledge. And one should not underestimate knowledge. 
He who knows most, sees most. One should not, 
however, on the other hand over-estimate know- 
ledge. It is of no use to him who cannot see. 

We address questions to the work of art. In order 
to be as complete as possible, it may be useful to have 
a questionnaire ready. 

184 



THE PROBLEM OF ORIGINAL SIZE 
First of all a query which is often forgotten: Is the 
work of art preserved for us within its original boun- 
daries, that is complete ? Is it a whole, or only a frag- 
ment, say the wing of an altarpiece ? Our judgment as 
regards the composition depends upon the answer. If 
it is a question of an easel picture, painted on wood or 
canvas, we examine the panel to see if it is present in 
its original extent, and investigate similarly the can- 
vas. The panel was usually, notahly in the Low Coun- 
tries during the i jth century, reduced in thickness by 
working neatly wedge-fashion towards the four edges. 
The uniform, prismatic cut is notable as a character- 
istic. The painters of the i jth century laid the gesso 
ground on the wooden panel which was already en- 
closed by its frame. The doughy matter as a result 
formed a ridge along the edges of the frame. If this 
ridge is visible at all four sides, then we may be sure 
of possessing the picture to its original extent. The 
canvas shows frequently old colour on the edges bent 
across the chassis. This is an indication that the canvas 
has subsequently been stretched on a new chassis, and 
that the picture surface has been reduced in conse- 
quence. 

Like naturalists we study the materials the species 
of wood, the texture of the canvas and the pigments. 
We gain as a result details of evidence helpful to- 
wards localizing and dating. In the Low Countries and 
in Northern Germany one used almost exclusively 
oak panels, mostly of slight thickness; in Southern 



ANALYTICAL EXAMINATION OF PICTURES 
Germany mostly lime wood or especially in the 
Alpine districts the streaky wood of the conifers ; 
in France, apart from oak, the wood of nut-trees; in 
Italy poplar wood in relatively thick boards. If a pic- 
ture by Diirer is painted on oak, we may surmise 
that it was executed in the Netherlands, that is in 
15-20 or 1 2 1. 

Pictures painted on wood have been transferred to 
canvas by means of a procedure which, used in 1 8th- 
century France but sparingly, has of late become 
fairly frequent, notably in Russia. If the climate is dry 
for example in the latter country as well as in Am- 
erica generally as a result of excessive heating the 
panel warps and causes blistering of the layer of 
colour. In order to prevent the colour from flaking 
off one resorts to the radical, at times very risky, pro- 
cedure of planing the wood on the reverse completely 
away, and substituting for it an elastic canvas. If the 
gesso ground which of course is handled with deli- 
cacy is of a certain thickness, then a relatively 
favourable result may be achieved. The picture, 
mounted on canvas, in that case retains its texture 
and lustre ; in other words, qualities which are char- 
acteristic of the panel pictures. Frequently, however, 
transferring entails a partial destruction, a perfora- 
tion of the original body of colour, and disfiguring 
restoration becomes necessary. Of the pictures that 
we possess from the brush of Jan van Eyck, no fewer 
than four have undergone the operation, surviving it 

186 



JAN VAN EYCK AND OIL PAINTING 
more or less satisfactorily ; namely, the two wings of 
an altarpiece in the Metropolitan Museum of Art in 
New York, the Annunciation in the National Gallery 
in Washington, the Crucifixion in Berlin and the 
Madonna with Saints belonging to Baron Robert de 
Rothschild of Paris. 

One may read everywhere that Jan van Eyck in- 
vented oil painting. Posthumous praise, more or less 
erroneous as to content, is never groundless : it points 
to an extraordinary performance, or at any rate to an 
epoch-making event, which then was personified. 
Now the pictures of Jan van Eyck undoubtedly look 
different from the ones that were produced before, 
or by others at the same time. And a notable point 
the Netherlandish followers of Jan van Eyck, who 
took the method over from him, were unable to 
utilize it in the same degree as he: could not as per- 
fectly bring it into harmony with their method of 
vision. We must therefore assume a personal ven- 
ture during the universal crisis in the conception of 
the world about 1420. I should like, however, to 
avoid the expression 'invention*. In any case it would 
be a mistake to assume that Jan van Eyck achieved the 
novel effect because, and after, he had invented some- 
thing. On the contrary he developed the new tech- 
nique because the traditional methods did not corre- 
spond to his vision. The find was result, not cause. The 
decisive and primary thing was the wish to achieve 
clair-obscur, richness of detail, gradation of light, a 

187 



ANALYTICAL EXAMINATION OF PICTURES 
mature which glowed jewel-like and was translucid 
in other words, what the painter had been the first to 
perceive and find beautiful. That which is called oil 
painting is something that genius detected in nature : 
and it found the means to realize its vision. 

What is known as oil colour was not used for paint- 
ing on canvas in the Low Countries before i3- The 
painted 'cloths' which were produced there before 
that date are painted in water-colour, and cannot 
be confused with pictures transferred from panel to 
canvas. 

It is easy to tell the difference between the rela- 
tively opaque, cool tempera, which obtained almost 
through the whole ijth century in Italy, and the 
deeply lustrous, liquid, repeatedly stratified 'oil 
colour 5 , which, ever since the days of Jan van Eyck, 
was favoured in the North of Europe. 

A chemical investigation of pigments permits de- 
ductions regarding the age of a picture; since the 
emergence into use and the discovery of certain pig- 
ments are historically demonstrable. As far as I am 
aware, this research has been carried farthest by Dr. 
A. P. Laurie in his volume The Pigments and Mediums 
of the Old Masters (1914). 

The scientific methods lately applied with eager- 
ness X-ray photography, irradiation by means of 
the quartz lamp, enormous enlargement, photo- 
graphy with powerful side-lighting supplement the 
report given by the naked eye, and often perform use- 



188 



RECENT SCIENTIFIC METHODS 
fill services. More particularly do they supply a diag- 
nosis of the medical case, without which a prudent 
restorer should not proceed to the operation. The 
Manuel de la conservation <Les peintures, published in 
1 93 8 by the Office International des Musdes, conveys very 
graphically information concerning all the devices of 
physical and chemical investigation. In so far as it is a 
question of the actual materials, the refinements of 
observation denote a progress which should be grate- 
fully welcomed. But when it is a question of artistic 
effect, there exists the danger that the scholars, who 
busy themselves so intensely with that which is in- 
visible to the naked eye, lose the capacity to receive 
an impression of that which is visible. Insensitive 
observers acquire the right to take part in a discussion 
about artistic matters : they take the watch to pieces 
in order to study the works. And the watch no longer 
goes. 

It is possible to judge from the manner of painting, 
the individual handling of the brush, in cases where the 
method of work is patent in deep and distinct traces: 
as for instance in the late works by Rembrandt or in 
the coloured sketches by Rubens. Temperament, elan, 
or tiredness betray themselves in the manner vehe- 
ment, decisive or else cautiously feeling its way 
in which paint has been applied. The nearer we get 
to modern times, the more openly does the painter 
reveal himself in the flow of handwriting. Dr. Laurie, 
in his book just referred to, demonstrates very in- 

189 



ANALYTICAL EXAMINATION OF PICTURES 

struct! vely, on the basis of some examples dating 
from the i yth, i8th and i^th centuries, the handling 
of the brush by means of considerable enlargements. 
But even where the colour surface, apparently self- 
contained after the fashion of enamel, seems to tell 
nothing about the method of execution, as for in- 
stance in the panel pictures of the i^th century, 
patient observation is capable of establishing a good 
deal concerning the method of execution. The sur- 
face is mostly not so uniformly smooth as it seems at 
first sight. Some pigments are spread, often as if they 
had run out, with heavier body than others over the 
picture surface. Each workshop had its own special 
procedures. Individual methods of applying the 
colours can be made out, even if, within the crafts- 
manlike formula of working which obtained during 
the i^th century, they reveal themselves compara- 
tively indistinctly. 

The X-ray photography at present employed with 
passion, often usefully, not infrequently to no good 
purpose reveals that which, invisible to the naked 
eye, lies under the top layer of colour; makes it pos- 
sible to recognize pentimenti that is, artists' correc- 
tions on second thoughts ; and is capable of teaching 
the experienced connoisseur a lot. But it confuses the 
inexperienced one and lures him to false deductions. 
The condition of the picture is of the greatest im- 
portance. We must possess a clear notion as to how 
much of the original work is present; how much may- 



190 



IMPORTANCE OF CONDITION 
be is wanting, or replaced or covered up by retouch- 
ing ; what is rubbed off; what is altered by decomposi- 
tion, by darkening or by opaque layers of varnish. A 
priori it is to be expected that no old picture should 
stand before our eyes in flawless condition, exactly as 
it issued from the workshop of the master. These 
questions, if ever asked, are but imperfectly an- 
swered by most people, even by otherwise excellent 
connoisseurs ; and hence their judgment often has an 
insecure basis. The study, unfortunately becoming 
ever more popular, of photographs at the writing- 
desk can least of all confer or strengthen the capacity 
to judge the condition of a picture correctly. The 
best training is provided by frequent visits to the 
studios of the restorers. 

A characteristic of good condition is the uniform 
effect of the whole of the picture. The experienced 
eye runs over the surface, enjoys the unbroken har- 
mony or becomes irritated and suspicious by con- 
tradictions. Opaque passages alongside of transparent 
ones; clear and distinct ones alongside of murky 
ones; delicate drawing alongside of careless; in a 
word, discord of every kind allows us to deduce the 
partial destruction of the original, defects and re- 
painting. Uniform rubbing of the entire picture 
surface is seldom met with, since, of course, the 
pigments have opposed more or less vigorous re- 
sistance to the destructive forces. 

In order to establish the degree of darkening, con- 



191 



ANALYTICAL EXAMINATION OF PICTURES 
fusing and discolouring through layers of varnish 
which have gone opaque or are in a definitely un- 
healthy condition, it is advisable to examine the 
lightest portions say, in the white draperies for 
the purpose of measuring how closely the highest 
light in the picture approaches pure white. In this 
fashion one obtains an approximate idea of the degree 
of disfiguration. 

The method, invented by the late Herr Petten- 
kofer, of using spirit vapours in order to make the 
unhealthy 'dead' varnish once more transparent, to 
'regenerate' it, is in many cases a useful procedure, at 
least for the purpose of gaining a clear idea of the 
condition of the original layer of colour. Lately this 
method, previously eagerly employed, is neglected in 
favour of X-ray photography and irradiation by means 
of the quartz lamp. The most recently invented 
method is naturally for choice regarded as a magic 
formula. 

The layer of varnish even if it gives the picture a 
warmth of tone which was not intended by the old 
masters cannot under all conditions be regarded as 
denoting a decrease of value. It gives the picture on 
the one hand a self-contained effect and restfulness, 
and on the other a picturesque 'mellowness' in 
other words qualities which at any rate to our eyes, 
the eyes of our time, are occasionally advantageous to 
the effect of the picture. There exists such a thing as 
unintentional increase of the picturesque effect. Not 



192 



THE 'CRAQUELURE* 

everything that the centuries have done to the layer 
of colour cometh of evil not even the craquelure, 
which is almost inevitable. It softens and reduces, in 
a welcome fashion, hardness, smoothness and empti- 
ness, entirely apart from the fact that wrinkles and 
symptoms of age once and for all are indissolubly 
linked up with our idea of the venerable art of the 
past. There is such a thing as patina, as aerugo nobilis, 
also on pictures. It is said that colours and tones, as 
years go by, amalgamate. 

As an object of study the craquelure is instructive, 
since it can tell us something about the date of the 
picture. Notably anyone who wants to be able to 
expose forgeries will do well to take a serious interest 
in this remarkable feature. The eye must train itself to 
tell the difference between that tattered condition of 
the gesso preparation and the layer of colour, which 
has come into being organically, and that which has 
been produced wilfully ; the necessarily natural-born 
change from that which has been produced artificially. 
Nature, in alliance with time, has more phantasy than 
the human spirit. Hence the natural craquelure throbs 
with rich variety, whilst monotony and pedantic re- 
petition mark the arbitrary, intentionally irregular 
one. 

Arc-like circular cracks, which recall the spider's 
web, are notably characteristic of the canvas pictures 
of the 1 8th century. 

The unscholarly pictorial technique of the i9th 



193 



ANALYTICAL EXAMINATION OF PICTURES 
century has often brought it about that the tattered 
layer of colour shows broad channels. This pecu- 
liarity has often made it possible to recognize for- 
geries or modern copies. But that this characteristic 
is not infallible became patent when the Madonna with 
the Sweet Pea at Cologne, on the strength of the 
cracks, was declared most erroneously to be a 
forgery dating from the i^th century. 

There are even cases in which the craquelure tells us 
something about the personal manner of painting of 
such and such a master, and is to be taken into account 
as a criterium of authorship. Palma Vecchio's device 
is the enamel of the body of colour, whereby he 
achieves an extraordinary lustre of the flesh and dis- 
tinguishes himself among his Venetian contempor- 
aries. The shadows of the flesh show with him a gritty 
decomposition, which produces the effect of pictur- 
esque softness and mellowness. 

Anton Graff's pictures show, as a result of special 
experiments in pictorial technique, a shrivelling of 
the colour surface which recalls rough leather. 

In the pictures by Pieter Pourbus I have hardly ever 
been able to discover craquelure. 

The criteria, obtained through science, help in any 
case to fix the time and place of the work of art and, 
as a result, indirectly further the determination of 
authorship, since through the classification concerned 
the circle of the masters to be considered is narrowed 
down and the discovery of the author is facilitated. 



194 



A TENTATIVE QUESTIONNAIRE 

A questionnaire, aiming at completeness with re- 
gard to pictorial expression and language of form, is 
something I do not desire to draft. It must be adapted 
to the character of the work of art from case to case. 
One cannot address the same questions to a Botticelli 
as to a Manet. A scheme with indications may here 
suffice. 

1 . Iconography. 

How has the subject been treated previously else- 
where? The relationship with tradition. From the 
legend of such and such a saint it is often possible 
to draw conclusions regarding the locality and 
date of the picture. Some saints were venerated 
only or predominantly in certain cities. 

2. Composition. 

Symmetry, more or less reduced in rigidity. The 
disposition of the figures in the plane or the depth 
of space. The relationship of the figure to space as 
indicated by landscape or figures. 

3. Architecture, Ornament. 

From the style of the buildings one may not in all 
circumstances deduce the date of the picture : it is 
only the terminus post quern which is at all times 
fixed. In the i jth and i6th centuries architectural 
forms of the past notably those of the Roman- 
esque period were often imitated in order to 
provide historical colour to the rendering of a 
sacred or legendary subject. 



195 



ANALYTICAL EXAMINATION OF PICTURES 
4. Language of Form. 

Proportions of the figures, motifs of movement, 
expression of sentiment, colour. 

. Costume, Arms and Armour. 

Knowledge of the history of costume can help con- 
siderably in dating a picture. It is to be noted that 
the Old Masters by no means always cling to the 
costume of their own period; but on the contrary, 
in order to suggest distance of time, and while 
knowing little about older costume, have more or 
less indulged in phantasies. 

In an exemplary fashion and with the simplicity 
peculiar to him Ludwig Scheibler has characterized 
the Cologne masters of the i $th century analytically 
in his memorable doctoral dissertation of 1880. If, 
however, someone imagined that he would only have 
to learn the letterpress of the book by heart in order 
to be able to determine the authorship of Cologne 
pictures he would be guilty of a sorry mistake. It is 
one's own impression of the entire picture which 
decides ; the dissecting contemplation serves at most 
as check and argumentation. 



196 



xxvra 

ON THE USE OF PHOTOGRAPHY 

PHOTOGRAPHY and the publishing of pictures 
are continuously on the up grade; conveniently 
accessible archives contain enormous quantities of 
reproductions, while the possibilities of travel are 
restricted for many students. As a result style-criti- 
cism is being practised in an ever increasing degree on 
the basis of photographs. The evil consequences of 
this condition of things are concealed from no one. 
The very fact of possessing a photographic reproduc- 
tion or the certainty of being able to obtain one 
reduces the interest which is devoted to the original. 
One should picture to oneself how the lover of art 
must have felt when he found himself face to face at 
Castelfranco with Giorgione's altarpiece, at a time 
when no photographs of it existed, and when he 
looked upon this first contact with the picture, as 
maybe, also the last one. How his emotion must have 
increased receptiveness ! 

It is true that the photograph has become indispen- 
sable, and an invaluable auxiliary ; but its use must be 
governed by discretion and moderation. It must not 

197 



ON THE USE OF PHOTOGRAPHY 
push itself into the place of the original. We must 
have a clear perception of that which it can perform. 

The risk of confusing original and copy has been 
immensely increased in the case of the facsimile re- 
productions of drawings, which seemingly are indis- 
tinguishable from the originals. The technical me- 
thods which are necessary in order to achieve such 
similarity, entail drastic intervention by means of 
retouching. The simple photograph, as supplied by 
the camera, is to be preferred to the facsimile repro- 
duction as a dependable, even if incomplete, report. 
Colour plates of pictures are to be used with pro- 
found suspicion. 

The ordinary photographic print, the half-tone 
block made from it not to speak of the lately all- 
too-fashionable cylindrical photogravures with their 
sham chiaroscuro at least do not belie their own in- 
sufficiency. The colour is lacking, and a great deal 
is lost as a result. The gradation of tone can, thanks 
to technical improvements, be reproduced with some 
measure of success. As to size and proportion, false 
ideas are conveyed to us. And these shortcomings are 
not even always made good to some extent by an in- 
dication of the real size. 

Apart from colour the reproduction lacks also the 
texture of the pigments, their lustre, their brilliance, 
their smoothness or roughness, their grain, their im- 
pasto. The indivisible effect, which springs from the 
whole, cannot be conveyed when the reproduction is 



198 



INADEQUACY OF PHOTOGRAPHS 
so fragmentary. The important preliminary question 
of the condition of the picture can be answered from 
a reproduction only in cases of drastic disfigurement. 
All these disadvantages apply also to lantern slides, 
which are so freely used in university teaching. 

Photographs should be used in order to awaken and 
strengthen recollection of the originals : as a basis of 
judgment they are to be excluded as far as possible. 
They will render good service in presenting an argu- 
ment, and in order to provide supplementary reasons 
for an opinion formed in front of the picture itself. 



199 



XXIX 
ON PERSONALITY AND ITS DEVELOPMENT 

OUR courage to proceed to the determination 
of authorship whether we go by intuition or 
by analysis and 'objective 5 criteria we derive from a 
belief that creative individuality has an unchangeable 
core. We start on the assumption that the artist 
whatever he experiences, whatever impulses he re- 
ceives, however he may change his abode at bottom 
remains the same, and that something which cannot 
be lost reveals itself in his every expression. This be- 
lief is often shaken by practical experience, but re- 
mains indispensable as a compass on the journey of the 
critic of style. If we stand in front of two works by 
the same master which, although both authenticated 
and for certain reasons indubitable, yet differ greatly 
from one another, then the question as to what can, 
after all, be determined as the common denominator 
does take us into the very depth of things, and into 
the very core of personality. 

In spite of many disappointments we persevere in 
our endeavour to discover something that is un- 
changeably solid, and in so doing often get into the 
position of a man who peels an onion and in the end 
realizes that an onion consists of peelings. 



200 



CHARACTER IN LITERATURE 
On reading novels and autobiographies one cannot 
help noticing the fact that the less important figures 
now on, now off the stage are drawn definitely 
and distinctly, whereas the hero who in novels not 
infrequently is identical with the author strikes us 
as indefinite, changeable, incalculable, not to say de- 
void of character. The more there is drawing within 
the contour, the less effective and expressive does the 
silhouette become. A description of character which 
always tends somewhat to caricature achieves success 
relatively easily, if it does not go below the surface. 
One can go so far as to claim that all human instincts 
and impulses are concealed in every human being, and 
are at war with one another ; that, according to cir- 
cumstances, this or that impulse moves towards the 
outside and becomes noticeable, in action and be- 
haviour, as a characteristic. The better you get to 
know a human being, the more surprises do you ex- 
perience from him. But however sceptically you may 
be prejudiced, you must, however, presuppose a dis- 
position of character in the individual relation of the 
impulses to one another, in the predominance of such 
and such instincts and inclinations ; and a direction of 
development caused thereby. 

I try to set up a scale in which the qualities, re- 
vealed actively and passively by a human being, appear 
arranged according to their stability, as springing 
more or less compulsorily from his disposition. I 
choose for the diagram in question, the shape of the 



201 



ON PERSONALITY AND ITS DEVELOPMENT 

star. In the centre is the solid core, a point- From 
this centre there issue rays, which undulate and oscil- 
late all the more vigorously the greater their distance 
from the focus. If I then draw concentric circles 
round the centre, I obtain zones, of which the inner- 
most one, with relatively straight rays close to the 
focus, contains the least mobile qualities ; the outer- 
most one, on the other hand, contains the qualities 
which are most powerfully shaken by experiences, 
by contacts, by demands. The ray passes thus from 
necessity to caprice. 

The difficulty begins as soon as we have to deal 
practically with the cut-and-dried scheme. We ex- 
pect that, say, courage to live, phlegm, melan- 
cholia, ethical forces, strength of will, timorousness, 
as well as intellectual faculties, belong in the inner- 
most zone ; whereas everything that has been learnt 
or taken over is to be relegated to the outermost zone. 
The average human being, however, acts and behaves 
predominantly in accordance with habit, norm, gen- 
eral custom; and it is only in exceptional situations, 
and for unusual reasons, that he reveals, surprising us, 
his personal character. 

It is a question by itself whether and to what a de- 
gree the artist gives visible expression in his work to 
his deep-rooted qualities, from the innermost zone. 
Habit may reveal itself more distinctly than individual 
impulse in specific and significant features. We often 
decide the question of authorship on the strength of 



202 



THE GREAT MASTERS 

characteristics whose connection with the core of 
personality is scarcely provable. The great master, the 
genius, brings more out of the depth into the light 
than the master of low or medium rank. Of course, 
no one is going to contradict certain prophecies, based 
upon the knowledge of independent, creative per- 
sonalities. Michelangelo can have produced nothing 
petty, Raphael nothing coarse, Holbein nothing vague, 
Diirer nothing frivolous. But already if someone were 
to lay it down that a master so delicate, so concerned 
about dignity and decorum as Van Dyck, could never 
have aimed at brutality of effect, it would be possible 
to contest this by reference to certain works by the 
master. In this case we might console ourselves with 
the reflection that Van Dyck, with his adaptability, 
his assiduity, his consciousness, does not belong to the 
circle of the great. But who belongs to it ? And who 
belongs to it during the whole of his development ? 
Once Rembrandt's emotional purity, spiritual free- 
dom and inability to compromise have revealed them- 
selves to us from the works of his late period, we 
stand startled and puzzled before certain cheaply 
striking and mannered productions of his early period. 
His genius, his personality rose but slowly, freed it- 
self from and left behind but gradually, such for- 
mulae, tendencies and aspirations as were limited in 
time and space. 

The strong man grows from his own strength, be- 
comes increasingly the one that he is ; the weak man 



203 



ON PERSONALITY AND ITS DEVELOPMENT 
resembles a plastic material which is being shaped. 
Hence it is a problem, in one case to get to understand 
a personality sympathetically, in the other to note the 
surrounding circumstances, the style of the period, 
the demands that were being made. In the first place 
it is principally intuition which decides as regards the 
'must'; in the second it is analysis and manysided 
knowledge which exercise similar action with regard 
to intention. The experts of the second rank can deal 
successfully mainly with artists of the second rank. 
The expert's relation to genius is that of the faithful 
disciple, to talent that of a cunning detective. 

We are entitled to expect that in all the productions 
of a master the degree of spiritual giftedness appre- 
hended in a work of art as level of merit remains a 
constant. In conformity therewith the connoisseur 
makes his decision in a positive or negative sense, and 
traces in awkward beginnings the possibilities con- 
tained in a given disposition. For each master he 
draws in his imagination a boundary line which may 
be reached and not crossed. That which is positive 
is less deceptive than that which is negative. A man of 
brains is far more likely to say something silly once, 
than a fool is likely ever to say anything intelligent. 

If we conceive individual nature not as something 
that exists but something that grows and that, of 
course, is what we have begun to do then we get 
over many difficulties. The great masters begin un- 
demonstratively, as it were iri a chrysalis: they start 



204 



THE ARTIST'S EVOLUTION 

on a line marked by their predecessors. We may re- 
call the discussions as to the boundary line between 
Giorgione and Titian, or as regards Albrecht Diirer's 
Bale period. The controversy about such problems 
continues for decades without being settled. If all 
pictures by Rembrandt had been lost, except one of 
1627 and one of 1660, it would be impossible to 
connect them with one another solely on the basis of 
style criticism. Only when we are familiar with the 
chain of many links, which makes up the ceuvre and 
that is the case with Rembrandt can we join begin- 
ning to end. 

The personality forms itself gradually, and we must 
see to it that it forms itself before our eyes. All con- 
noisseurship aims at biography. The Ariadne thread 
of biographical dates makes it possible for us to find 
our way. The chronological order helps us consider- 
ably. That which a master once has achieved cannot 
be completely lost by him. Every creation can be 
regarded as the result of all preceding ones. 

Greco can only be understood if we know that he 
went from the Near East to Venice, and from there to 
Toledo. 

Sooner or later, the master finds the path which is 
in conformity with his nature, and thereupon follows 
it more or less in a straight line. Capacity for evolu- 
tion is a characteristic of powerful gifts. Those who 
are great find themselves at the end far away from 
their starting-point. 



ON PERSONALITY AND ITS DEVELOPMENT 

Historians have learnt to reckon with the possi- 
bility of change, but not with its necessity. A master 
cannot, strictly speaking, produce the same in 1 5-20 as 
in 15-10, unless he imitates himself, in which case 
paralysis, dullness and ossification are bound to become 
noticeable. When growing ceases, decay begins. 

The position is seldom so favourable as in the case 
of Memling and of Frans Hals. It is generally not made 
so easy for us to form an idea of the development of a 
master in one locality. We are faced with necessity of 
building up through our imagination, as far as pos- 
sible, for each master something like the Hospital of 
St, John or the Frans Hals Museum. At one time there 
was only one work the Portinari altarpiece avail- 
able to tell us something about Hugo van der Goes. 
One point in his career was thus fixed. I then compare 
the Portinari altarpiece, whose date is known, with 
earlier and with contemporary works by other mas- 
ters especially such as were settled in the same 
neighbourhood and already in so doing gain an idea 
of the direction which the personality of the Ghent 
master was bound to follow. I am able to carry a line 
through the point denoted by the Portinari altarpiece 
in two directions, and thus to further the aggregation 
of other works. It is an advantage if, as in this case, 
the fixed point has its place approximately in the 
middle of the artist's career. 

It is to be presupposed that Hugo van der Goes, as 
an independent master, highly capable of develop- 



2o6 



EVOLUTION OF HUGO VAN DER GOES 

ment, enlarged and widened his form more and more ; 
and this makes us inclined to consider pictures of 
small size like those in the Vienna Gallery as com- 
paratively early works. But, as I remarked when 
speaking of size and proportion, the painters were 
not free in the choice of dimensions. The Vienna 
panels do not show a natural, congenital smallness; 
they suggest a reduction of size, due to compulsion. 
They grow in one's recollection. Their content of 
form and knowledge of form are in conformity with 
larger proportions. Hence we begin to doubt whether 
we really are concerned with youthful productions. I 
feel more certain in regarding the St. Anne, the Virgin 
and Child, the modest work in the Brussels Gallery, as 
a relatively early production, because in this instance 
the volume of sound seems to be in conformity with 
the size of the instrument. If I place the Portinari 
altarpiece and the Monforte altarpiece thus two 
triptychs of approximately similar dimensions next 
to one another, I believe I am entitled to deduce that 
the Monforte altarpiece is the later work, because in 
it the language of form seems even more of a match 
for the monumental proportions. This is, by the way, 
not the general opinion. 

A statement 'this work, from its character of style, 
is the work of a youthful master 3 is audacious, and 
unreliable. One may say, with more justification and 
definiteness : 'This picture by the master was painted 
earlier than that one; it is, comparatively speaking, a 



207 



ON PERSONALITY AND ITS DEVELOPMENT 
youthful work/ Only with difficulty should one de- 
cide to say: "The master cannot have painted this pic- 
ture' ; it is already easier to opine 'he cannot have 
painted it in 1470'. The more we know about the 
master's destiny, the more extensive the material of 
examples we have brought together, the more does 
the circle of possibilities and mistakes contract, the 
greater become the calm and determination with 
which we classify and build up. 

General rules as to individual growth, valid always 
and everywhere, may only be formulated with extreme 
caution and reserve; they assist the critic of style 
solely in conjunction with far-reaching knowledge of 
biographical dates. Titian, Rembrandt and Frans Hals 
have followed paths which run approximately parallel. 

Youth, quickly changing, is bold and shy, arrogant 
and dejected; the age of man's maturity witnesses 
solid work for the benefit of the outside world and 
the reaching of an understanding with conditions that 
exist; old age if untroubled by illness, want or care 
has reached clarity and is equably cheerful. Youth 
learns to look, old age to overlook. At first hesitation 
while walking briskly; then an unperturbed advance; 
finally, rest. Healthy natures show, more or less clearly, 
this sequence. 

One may form for oneself an ideal of organic awak- 
ening, maturing and decaying ; one may postulate for 
each age certain instincts and impulses as predomin- 
ating; one must, however, also bear in mind the 



208 



FORCES AFFECTING DEVELOPMENT 

many forces which cut across a normal development, 
such as illnesses, opposition from the world outside, 
uprooting, the growth of bitterness owing to lack of 
success. Prejudice frequently enough confirmed, 
for that matter causes us to presuppose that form, 
as an individual artist develops, becomes ampler and 
poorer in detail. It is, however, perfectly possible for 
the taste of the period which terrorizes notably the 
feebler talents to direct precisely an opposite course 
to be followed; as was for instance the case in Hol- 
land during the second half of the iyth century. The 
strong men pursue their path in opposition to the 
general movement, as for example Frans Hals and 
Rembrandt, who at the end, isolated and uncompre- 
hended, rose above their contemporaries. Other great 
masters, like Raphael or Titian, seem to cover long 
distances in step with their own generation, though in 
such cases it is difficult to define how powerfully they 
themselves determine the taste of the period. Many 
masters have died prematurely, having had no time to 
age organically. 

That the mature master works with wise superio- 
rity, relying more upon experience and memory than 
upon observation, and taking a general view of things ; 
that the ageing master may reach the point at which he 
becomes his own imitator: this is a law of artistic na- 
ture with which we have to reckon. 

All human activity is governed by the law of inertia. 
It is only possible for strong forces such as fanaticism, 



209 



ON PERSONALITY AND ITS DEVELOPMENT 
coercive richness of imagination pertaining to genius, 
ambition, dissatisfaction with one's own perform- 
ance to paralyse the deep-rooted compulsion to 
repeat a movement, to follow the same path once 
again ; and such strong forces generally wane with the 
passing of years. Every action demands, when per- 
formed for the second time, a lesser expenditure of 
force than the first time. Habit runs through all ar- 
tistic activity, and more particularly when a master of 
mature years can look back upon success, upon re- 
cognition. Repetition, as the inner tension decreases, 
is eventually conducive to mannerism. 

We speak of mannerism as opposed to style when 
we come upon forms that are conventional, imitated 
from the artistic production of others or oneself, not 
derived from vision: that is, upon forms that strike 
us as artificially made, instead of natural-born. An in- 
structive example of the groundlessness of manner- 
istic motifs is offered by the fluttering terminations of 
draperies, favoured in Antwerp about 1^20; they 
wave and whirl without reference to an air current or 
any other motive power. 

The assumption of a normal development proves 
fruitful even in cases where, as frequently experienced, 
observation testifies against it. Questions such as why 
Lucas Cranach was not impelled towards the grand 
style, the picturesque; why Albrecht Altdorfer in his 
maturity became a miniaturist; which forces coun- 
teracted the natural unfolding such questions are 



210 



METHOD OF DEDUCTION 

conducive to the study of surroundings and the condi- 
tions of the times, and facilitate the construction of 
biographies. The general tragic German fate counter- 
acted the organic individual development, particu- 
larly fatally in the case of Cranach, but more or less in 
the entire production of his time. Only the genius of 
Griinewald seems not to have been hemmed in by the 
repressive forces. 

Generally speaking the art historian cannot exercise 
sufficient caution in dealing with the concept of law ; 
he should content himself with deducing points of 
view from his observations and from these points of 
view he then gains further observations. The natur- 
alist may argue: 'if this is what happened in that case, 
the same thing must happen in this case' . We, how- 
ever, must limit ourselves to saying: 'if this is what 
happened in that case, the same thing 12707 ^PP 611 * n 
this case, and we will now see if that is so 3 . 

At times the following phases outline themselves: 
first, clinging to inherited form, tradition and theory ; 
then, awakening of the individuality coupled with in- 
dependent observation of nature ; finally, autonomous 
handling of the possessions of form thus acquired. The 
path runs from the manner of others, via observation 
of nature, to one's own manner. The frequently de- 
fective knowledge with regard to youthful works may 
seemingly restrict the number of cases in which this 
scheme is applicable. We know, for instance, nothing 
about the beginnings of Lucas Cranach, who perhaps 



21 I 



ON PERSONALITY AND ITS DEVELOPMENT 
reveals himself to us only in the second phase of his 
development. 

The works belonging to the old age of the greatest 
masters all share a sublime and transfigured timeless- 
ness. 

Only if the destiny of a master with all its changes 
is known to us in every detail and our knowledge 
never extends thus far would we be in a position to 
apply general rules to the individual case without vio- 
lating the latter. 

The ultimate wish, hardly ever fulfilled in the case 
of the art historian, is directed towards the discovery 
of the law in conformity with which personality began 
to be formed. We should like to deduce from the 
seed, by which we understand the early work, all 
possibilities of development. The more original the 
work of a master, the closer do we expect to ap- 
proach this goal, which is never reached but must 
never be left out of sight. 



212 



XXX 

ON THE ANONYMOUS MASTERS, THE 

MEDIUM MASTERS AND THE LESSER 

MASTERS 

THE great masters, with whom historians for 
choice concern themselves in the hope of com- 
ing across those forces which were the decisive ones 
in history, are exceptions apart at the same time 
those who produced some effect already upon their 
contemporaries be it the effect of admiration or 
amazement so that repute beyond the grave has pene- 
trated into early writings. It is true that traditional 
fame must not, without further ado, provide the his- 
torian with his standards not to speak of the lover of 
art. Accents have been distributed from prejudice, 
according to the standpoint of the chronicler. Vas- 
ari's partiality for Florence is even now productive of 
confusion. 

Some masters have by signing their works pro- 
vided for their fame beyond the grave. Martin Schon- 
gauer may not be greater than the Master of the Haus- 
buch ; he has, however, as a historical personality got 
in before the anonymous artist, precisely because with 



THE ANONYMOUS MASTERS 
his Initials he made things easy for the historians and 
forced himself upon them. It is the great artists con- 
cerning whom we learn, at any rate, something with 
regard to time and place, life and influence; so that 
we construct the edifice of style-criticism with the 
aid of a biographical scaffolding. 

There are, however, cases in which we build with- 
out a scaffolding, as it were stitch and crochet, in- 
stead of carrying out an embroidery on a given 
ground. The study of masters who owe their exis- 
tence and provisional names to style-criticism, can 
be described as the march past of connoisseurship. In 
this endeavour we must take as our starting-points the 
well-known masters, whose historical position is 
firmly established and who, like milestones, make it 
possible to assign places to the anonymous in their 
vicinity and between them. Let me quote as an instance 
a panel picture, on oak, displaying a composition 
which in part goes back to Roger van der Weyden, 
and a language of form which recalls Memling. The 
subject is St. Donation, the patron of Bruges. Hence it 
is possible to deduce: Bruges, second half of the i jth 
century. I look, not without success, for pictures by 
the same hand in Bruges churches and find one bear- 
ing the date 1480. The dated picture looks earlier than 
the one first considered. I now have at my disposal 
not only the characteristics of a personal style but also 
an idea of the direction taken by the painter's evolu- 
tion, and can, with growing certainty, increase the 

214 



'MASTER OF THE DEATH OF THE VIRGIN' 
ceuvre of the anonymous master and put it into order. 
Under all circumstances we work with a yearning 
for biography. 

The 'Master of the Death of the Virgin', before his 
real name was known, stood before the eyes of art 
lovers as a personality definitely outlined. Time and 
place, the direction taken by his development, and a 
considerable ceuvre had been deduced without the help 
of documentary or literary tradition. As his ceuvre 
grew, the more data and supports emerged, enabling us 
to evolve the hypothesis that he was identical with 
Joos van Cleve, a painter about whom something was 
to be found in early writings. That which had been 
conquered by means of style-criticism tallied happily 
with the biographical data. Finally, the Cleve arms in 
an altarpiece, together with the initials J. v. B. (Joos 
van Beke), transformed the surmise into ascertained 
fact. 

We experience difficulty in keeping up with the 
great, and worry lest we be hoaxed by the lesser men. 
The talent of medium or lesser strength disguises it- 
self, masquerades, intentionally does now this, now 
that. Genius changes from inner necessity, talent for a 
reason. Evolution in one case follows its course in ac- 
cordance with laws, which it is one of the tasks of a 
biographer to discover; in the other by fits and starts, 
with sudden changes, whose causes the biographer en- 
deavours to establish. 

The modest artists find mostly, after a period of 



THE MEDIUM MASTERS 

feeling their way, a manner to which they cling com- 
placently, especially if some success is vouchsafed 
them. They make least trouble for style-criticism. 
The assiduous and ambitious men of medium stature, 
especially in a critical period of universal change of 
style, are capable of driving the expert to despair. The 
task of bringing their czuvre together is often insoluble, 
unless inscriptions or * objective' data of another 
kind lend their assistance. An example of histrionic 
capacity for change is afforded by Bernard van Orley. 
As regards the medium and lesser masters, one should 
always bear in mind their colleagues of the same gen- 
eration who, equal in artistic importance, work under 
similar conditions and in the same atmosphere ; as re- 
gards the great masters one must not lose sight of their 
imitators and copyists and also the forgers. 

In itself the 'attributing 3 of the insignificant works 
of art does not appear too important ; what mainly 
gets the sublime sport going, and indeed may turn it 
into a profitable profession, is the insatiable hunger 
for names on the part of the collectors and dealers. 
You may do your best by talking to these people and 
pointing out to them that every work of art, even the 
poorest one, is due to one human being who has borne 
a name; and that it depends on accidental circum- 
stances whether the name is known or not. The de- 
lusion that something notable clings to each name is 
ineradicable. Whoever pays a lot of money for a Rem- 
brandt demands to be covered by authoritative judg- 

216 



STUDY OF LESSER MASTERS 

merit. The unconditional respect for names, even ob- 
scure ones, is at all events a bad symptom so far as 
taste and feeling for quality are concerned. 

The attention devoted to the lesser masters has 
proved profitable and fruitful inasmuch as the per- 
sonalities of the great artists, as a result, have been de- 
fined more clearly and decisively. Much has been 
gained for the understanding of Rembrandt after his 
pupils and followers, one after the other, have metho- 
dically been put on their feet, both biographically and 
from the point of view of style-criticism. A happy 
cleansing of his CBUVTC has thus been carried out. 

General validity attaches to the maxim that it is 
easier for the expert to say 'this picture is by such and 
such a hand' than to gain the conviction that it is not 
by that hand. We judge with greater certainty posi- 
tively than negatively. 

The study of the lesser masters furthers knowledge 
of the general level, of the style of the period. We 
learn to know the starting line of the great masters, 
and see how it is set off lustrously against the dark 
background of average activity. 



217 



XXXI 
THE STUDY OF DRAWINGS 

NOTHING is to be recommended more strongly 
to the aspiring connoisseur of pictures, and in- 
deed urged upon him, than assiduous study of draw- 
ings. Whoever turns from the pictures of a master to 
his drawing has the feeling that a curtain rises before 
him, and that he is penetrating into the inner sanc- 
tuary. For more than one reason a drawing is superior 
to a picture as evidence, as an autograph. It came into 
being relatively quickly, as a result of spontaneous 
action, and did not have to take the long and toilsome 
road through craftsmanlike procedure ; it is in conse- 
quence less closely tied to teaching, tradition and 
studio convention. When drawing, the artists of the 
i ^th and 1 6th centuries were more of artists ; when 
painting, more of craftsmen. The drawing stands in 
the same relation to painting as a mountain brook to a 
canal. In many cases the draughtsman is not comply- 
ing with any wish from outside, does not carry out an 
order: he feels himself free in mood and fancy, alone 
with himself, as it were, as speaking a monologue. 
Moreover, a drawing has hardly ever suffered from 

218 



OLD MASTER DRAWINGS 

distortion, subsequent alteration, restoration or falsi- 
fication. Everything lies there open to the day, as at 
the moment of its birth. 

To draw, in a higher degree than to paint, involves 
selection, decision, elimination, spiritual interven- 
tion: hence it is inestimable as an immediate, per- 
sonal, intimate utterance of individuality. 

It is regrettable, and a hindrance, that the totality 
of surviving drawings should be so very unequally dis- 
tributed among the centres of art periods and masters. 
Scarcely any drawings by Frans Hals and Velazquez 
have been traced, and only a few by the great masters 
of the i .fth century, while, when we come to Raphael, 
Leonardo, Diirer and Rembrandt, the number of ex- 
isting and known drawings is great: indeed, it has im- 
measurably enriched the idea of these masters, and in 
many respects provided the foundations for it. 

In the past the artists stood closer to nature when 
they drew than when they painted. The intentions of 
the draughtsman were now in this, now in that direc- 
tion. In many degrees it is a question of attempts, 
means of orientation, preparations, models, studies, 
designs, gaugings, notations of sudden ideas, ideas of 
pictures; but also of self-contained works of art 
which were retailed, sought after and collected, like 
the coloured drawings of Aelbert Cuyp. 

In the course of evolution the drawing freed itself 
more and more from painting. At a primitive stage a 
picture was nothing but a drawing that had been 



219 



THE STUDY OF DRAWINGS 

coloured, that had been completed by the indication 
of the local colours. A drawing by Rembrandt differs 
from a picture by this master more strongly than say a 
drawing by Roger van der Weyden differs from one of 
his panel pictures. Painting has gradually realized the 
specific possibilities given to its means. Drawing, on 
the other hand, was carried away into the movement, 
becoming pictorial with richness of tonality and in- 
creased looseness of stroke ; on the other hand it de- 
veloped the special style conformable with its means, 
in the sketch, in the rapid notation, the writing down 
of the sudden flash of an idea for a picture. 

In the highest degree personal, and original in the 
narrowest sense, are such impressions as we possess 
by Rembrandt, which mostly bear no relation to pic- 
tures, do not exist as auxiliaries or designs, but on the 
contrary have been done purposelessly, out of the 
sheer abundance of vision. 

As to knowledge of form and the measure of ex- 
pressive power, the evidence of the drawing is more 
definite than that of painting, which can conceal short- 
comings and cloak defects. Many painters are badly 
given away by their drawings. 

To establish the relationship between drawing and 
painting in each case is highly instructive. Rubens has 
utilized studies from nature of individual figures in 
order to give his painting steadiness and firm struc- 
ture. In the i th century the drawing is concealed be- 
neath the layer of colour ; it stands in the same rela- 



220 



DRAWING AND PAINTING 

tion to painting as the skeleton to the flesh ; it is pre- 
sent in painting as something immanent, just as paint- 
ing is immanent in Rembrandt's drawings. 

Fortunately we possess a solidly instructive book in 
Josef Meder's volume Die Handzeichnuag (1919), 
which deals precisely with the art and technique of 
drawing. 



221 



xxxn 

INFLUENCE 

METAPHORICAL technical terms have the un- 
pleasant quality of being used for many pro- 
cesses or conditions which are different from one 
another, and eventually also for such as the metaphor 
does not fit at all. Every effect, which is produced by 
one master upon another, is called influence. Under 
the sign of this image we perceive how green waters 
pour into blue ones, with the result that the colour of 
the stream is changed. Thereby the idea of a mechan- 
ical occurrence is awakened. You should, however, 
definitely distinguish between occurrences that are 
mechanical fundamentally, additions or minglings; 
chemical reactions ; and psychical the latter so com- 
plicated that no term is adequate to their multiplicity. 
That which is called 'influence 7 is a psychical occur- 
rence. From case to case, in boundlessly numerous 
modifications, something occurs or fails to occur 
when two artists or two manners of art collide. 

The masters of the past lived in a condition of com- 
munity, akin to that of the guilds ; they worked at 
times conjointly, and helped each other out. The in- 
clination to segregate oneself, to cultivate individu- 
ality as a priceless possession, to isolate oneself, to 



222 



THE MODERN ARTIST 

retire to an island, is connected with the striving 
for originality characteristic of modern times ; while in 
our days impulses of variegated multiplicity are con- 
veyed to an artist in confusing quantity by exhibitions, 
museums, photographic reproductions and the teach- 
ing of the universities. Ambition causes the modern 
painter to try to free himself from the eclecticism 
forced upon him. Tradition was formerly a coercion 
from which genius alone, and even genius but gradu- 
ally, freed itself. The painter took over ideas and lan- 
guage of form from his teacher, and felt himself content 
with his inherited possessions, provided for and armed 
to perform that which was asked of him. He had what 
he needed when he made himself independent, and 
nothing caused him to look out eagerly for impulses 
and artistic experiences. This was the normal condi- 
tion of a master who still stood very close to crafts- 
manship. 

Where something can pour in, there a vacuum must 
exist in other words, say, an ideal not realized at 
home or in tradition, ambitious discontent with that 
which was one's own, a need, a power of attraction. 
Often the readiness to absorb something foreign was 
heightened by travels. 

With regard to certain masters we are fully in- 
formed and are able to measure in detail the effect of 
the journeys, of the visual experiences abroad. Jan van 
Scorel, van Dyck and above all Diirer travelled with 
open eyes, with the wish to learn and to enrich them- 



223 



INFLUENCE 

selves. A close examination of individual destinies 
which in the case of Diirer can succeed, especially 
thanks to the penetrating analysis of Professor Hein- 
rich Wolfflin is conducive to conclusions of general 
validity. 

Diirer and Jan van Scorel lived in a period of thirst 
for knowledge, in a restless time, in which the ex- 
change of intellectual possessions and achievements 
between the peoples, especially across the Alps, was 
longed for and assiduously practised. That which can 
be learnt was overvalued perspective and the theory 
of proportion, for instance. Rationalists are more eager 
to learn and more capable of learning than nai've people . 

In all cases the vessel which receives must first be 
examined, and then only the force pouring into it. To 
what degree is the vessel empty and in need of con- 

O -I J 

tent ? Is the master still growing ? Do the forces which 
hitherto have nourished him no longer nourish him 
sufficiently ? Is the field ploughed and ready to receive 
the seed ? These, and similar questions, must be asked 
from the fundamental consideration that an artist is 
always capable of realizing possibilities that are con- 
tained in his disposition whatever he imitates, which- 
ever leader he follows, whatever school he passes 
through, whatever the track along which he climbs 
upwards. 

The vessel becomes capable of reception by expel- 
ling something. By this I want to say that whoever 
acquires something, gives something else away ; he can- 



224 



COMPULSION AND CHOICE 

not pile up acquisitions. The adaptable, receptive 
nature makes sacrifices all the time. It is true that the 
Academicians in Bologna imagined that they could add 
or multiply values. Reynolds sought success by com- 
posing like Van Dyck and painting like Rembrandt. 

The relationship of the forces on the one hand and on 
the other must be weighed. If a weak talent encoun- 
ters genius, then it misunderstands although it may 
give itself up completely ; if strong gifts collide with 
still stronger ones, then there is a possibility of under- 
standing many things, and of robbing the examplar 
of that which can be utilized. 

Compulsion and choice at times stand out in vigor- 
ous contrast to one another. Van Dyck laboured in his 
youth under the terror which issued from Rubens and 
sacrificed something of his individuality ; later in Italy 
he chose Titian, the Venetian, as a congenial exem- 
plar, and in this fashion freed himself from the Flem- 
ish tradition. With the aid of Titian he set personal 
qualities free, and could now satisfy his desire for 
grandezza and distinction. He was at first like wax 
which is impressed; then he seized the initiative; 
finally he combined and became fossilized in routine. 
Whether it be passive or active, personality reveals it- 
self in all phases also in its capacity to give itself up to 
something, to disguise itself, to utilize and to blend. 

In a similar fashion the relationship of Diirer to 
Schongauer and to Mantegna should be analysed. The 
engravings of Schongauer revealing the manner of 



INFLUENCE 

the Late Gothic goldsmith in its highest perfection 
were bound to affect the youthful Diirer as a powerful 
magnet. When his own forces burst the narrowness 
of this form of style, Mantegna severe in his great- 
ness and solidly constructive offered precisely that 
which the German artist, conformably with his nature 
and his then stage of maturity, could absorb from 
Italian art as a nourishment that was wholesome and 
favourable to his growth. 

Little noted, and yet very notable, is the absence of 
effect. Bruegel was in Rome and seems to have per- 
ceived neither Raphael nor classical statues. The vessel 
does not admit not only because it is closed, but also 
because it is fulL That which was intuitively expected 
is welcomed. That which someone loves, respects, 
understands, imitates then, the way in which he mis- 
understands it, and that which he overlooks all this 
completes in an instructive fashion the idea of a person- 
ality . A painter chooses as his leader a master in whose 
works he has found his unsubstantial dreams realized. 

Some masters were imitated because their method 
of painting called forth admiration: but it was the 
creators of types, the story-tellers who, stimulating 
and affording an example, exercised a much stronger 
influence. This influence reached far and wide, especi- 
ally once the picture-print had intervened as a vehicle 
of popularization, conveniently offering, as it were, 
an excerpt of the imitable. Masters who have height- 
ened and extended their domination through engrav- 

226 



MAIN CURRENTS OF ART 

ing original and/or reproductive are Schongauer, 
Diirer, Raphael, Rubens and Watteau. 

The Middle Ages appear cosmopolitan so far as art 
is concerned, since the world domination of the 
Church did not allow the individual tendencies of the 
nations and peoples quite to develop. Just as the 
scholars of all countries could understand each other 
by means of Latin, so did the artists express themselves 
in a language of form which was homogeneous, though 
rich in vernacular. Later on, as the conception of the 
world became more mundane, and as the national 
states emerged, the North separated from the South. 
Personality, in freeing itself, drew also the character- 
istics of the race, of the people into the light. Now as 
the Italians with an unspoilt eye turned to reality, they 
concentrated themselves with fanatical onesidedness 
on the human body whose loveliness, strength, dignity 
and nobility were glorified in proud self-conscious- 
ness. In the North, on the other hand, more attention 
was turned to the human soul, and to the intercon- 
nection of body, space, light, and atmosphere. 

The accurate and conscientious observation prac- 
tised by the Netherlandish masters, coupled recipro- 
cally with the detailed method of painting, fixed on 
the object and invested the entire production with 
something reminiscent of still life, a comfortable, 
contemplative narrowness and stiffness. The pictures 
from the North struck the South as being pious, with 
ascetic vigour or collected devotion. The successes 



227 



INFLUENCE 

gained in Italy by the professional landscape painters 
from the North such as Paul Bril, Elsheimer, Pous- 
sin, Claude confirm a superiority in this field which 
was recognized, even if too high a value was not 
placed on it. Of "noble 5 art there was demanded that 
which the Italians could give with unsurpassable 
mastery, and therefore Netherlandish, German and 
French painters in the 1 6th century went to the for- 
eign school, where they strove to learn picture-build- 
ing with human bodies in movement as their material. 
The plants transplanted in the North burst forth in 
strange flowers. 

About 145-0 Memling left the Middle Rhine, Schon- 
gauer Alsace, Diirer's father, the goldsmith, Hun- 
gary and a ll for the Netherlands in search of the 
'great masters' ; but Diirer in 1495" went to Northern 
Italy. It is profitable to follow the exchange which 
took place between North and South, of motifs, lan- 
guage of form, artistic knowledge, technical means. In 
each individual case it is different relationships and 
different consequences and results which ensue. We 
must measure capability of reception, readiness, the 
strength of the innate gifts which are capable of 
digesting the alien substance. Gossaert, Jam van 
Scorel, Frans Floris, Rubens, and van Dyck stayed in 
the South, eager to learn, and in each case analysis 
proves a different relationship of gain and loss, of 
nourishment and poison. Rubens provides the image 
of an unique victory. He avoided no danger of eradi- 



228 



EXAMPLE OF THE GREAT 

cation and overcame every one of them ; he gave him- 
self up consciously to Italian Baroque, and eventually 
freed himself from it unconsciously. 

The great masters, mighty in their rule and sweeping 
us along, exercise seductive attraction on their own 
generation and those that follow, if and as long as they 
express themselves in harmony with current taste, 
and remain comprehensible up to a point. In their 
full maturity they are capable either of taking or of 
giving. They work then in exalted seclusion and isola- 
tion, and their followers, who always misunderstand 
or caricature a little, desert them. It is the gift of 
a genius to see things which his contemporaries do 
not see. The latest works of Rembrandt and Titian 
have been least imitated. Frans Hals and Rubens, 
too, attracted more adepts when they began than at 
the end of their career. Those who ran after them 
could not keep pace, but lagged behind. 

Their achievement and quick push forward mean 
that the great masters move in advance of their time 
with such speed, and so far, that the importance of 
their example is recognized only long after their 
death. Thus it is, strictly speaking, only in the igth 
century that Velazquez has exercised influence, and 
Frans Hals, at any rate in his late manner, has not been 
appreciated or imitated at an earlier date either. Goya 
is in the same position. The case of Greco is some- 
what different, since this old master was discovered 
as an ancestor rather than as a teacher. 



229 



XXXIII 

ARTISTIC QUALITY: ORIGINAL AND 
COPY 

ON reading impatiently or with a patient smile 
elaborate dissertations which 'prove' that a 
picture is by such and such a master, at the end we 
generally, after many arguments and references, come 
upon the word 'quality'. This means that the decisive 
point, which also brings things to a dead end, has 
been reached. You derive the impression that the en- 
tire letterpress lengthy, spasmodic, crowded with 
quotations is just counsel's pleading, while the judge 
who condemns or acquits solely uses the word 'qual- 
ity 5 . The concept of 'quality' arrests the flow of words 
of even the most garrulous. 

When an impression fills us with pleasurable satis- 
faction with 'disinterested pleasure' as the aesthe- 
ticians say it springs from a pure, individual and 
hence uniform vision and also from a successful reali- 
zation of this vision, thanks to which the emotional 
values are communicated to us without any con- 
siderable loss. We hear an individually coloured voice 
which says something that we know, but says it in 

230 



REACTIONS OF THE CONNOISSEUR 

such a fashion that we think we hear it for the first 
time. 

A pleasurable sensation of a definite degree and a 
definite kind is, in our experience, associated with 
the works of this or that master. We stand in expec- 
tation of some such kind of delight as a work of art 
can produce a sense of elevation, of shock, of reve- 
lation, of disclosure, of rapture, or whatever the case 
may be and we decide on authorship and authenticity 
according as such an experience does or does not 
take place. 

The connoisseur of wine determines with full cer- 
tainty brand and vintage from a particular flavour: in 
the same way, the connoisseur of art recognizes the 
author on the strength of the sensually spiritualized 
impression that he receives. Sometimes it is a ques- 
tion of lovely equipoise, sometimes of stark, exciting 
vividness, sometimes again of an intensification of the 
sense of life, or a sense of pathos, of boundless abun- 
dance, of heroic exaltation and every time the accent 
is unmistakable. Always quality shows itself in this, 
that emotional values experienced by the artist in his 
vision are interpreted in visible terms. 

I must anticipate the query whether the quality of, 
say, a still-life or a slight scene from passing life is due 
to a successful interpretation of spiritual emotion. 
The answer is, Yes ; only it is necessary fully to under- 
stand what has happened. When Chardin saw a fruit 
this visual experience filled him with delicate enjoy- 



231 



ARTISTIC QUALITY: ORIGINAL AND COPY 
ment, and this sensation invested eye and hand with 
the capacity to paint as he did. Again, when Pieter de 
Hooch painted the roof of a house, so deeply did he 
feel the marvel of light in the wealth of colours and 
tonal values, that he was able to communicate to the 
spectator, in full purity and completeness, the delight, 
the peaceful pleasure that the world, as he had seen it, 
had given him. 

It is only the line put down at the dictate of feeling, 
only the brushstroke guided by instinct nothing that 
is taught, calculated, selected or painstakingly im- 
proved upon which communicates the vibrations of 
feeling and thereby that experience for the sake of 
which art has value to us. No wonder that the differ- 
ences between good and bad, measured or weighed, 
appear infinitesimal. 

An unsatisfactory column, shall we say, is the one 
which is drawn as with the ruler. To the good archi- 
tect the column is an organism with a soul, suffering, 
triumphant, carryingand burdened ; and in the scarcely 
measurable, delicate life of the outline there are ex- 
pressed strength, tension, pressure and resistance. 

The notion of quality is brought out by a compari- 
son between an original and a copy better than by the 
best definition. Such a comparison makes us pene- 
trate deep into the essence of artistic production. The 
understanding of Holbein's art was furthered with 
quick strides, thanks to the 'Holbein War' the occa- 
sion when, in 1871, the Darmstadt Madonna was 



232 




By giacious permission of His Majesty The King 
24. AFTER JAN VAN EYCK, CANON VAN DER PAELE 
Hampton Court Palace 




22. AFTER HANS HOLBEIN 
MADONNA OF THE BURGOMASTER MEYER 

Dresden GalJery 
Painted about 1632-1638 




23. JAN VAN EYCK, CANON VAN DER PAELE 
Detail of the Altarpiece in the Bruges Museum 



DRESDEN OR DARMSTADT? 

placed alongside of the Dresden picture, with the re- 
sult that the celebrated Dresden version was recog- 
nized as a copy dating from the seventeenth century. 
Adolph Bayersdorfer's brochure Der Holbein-Streit 
(1872) can still be read with profit, even for other 
aspects of the subject. 

All opinions even the mistaken ones expressed 
on the two versions are instructive. You find them 
reprinted in a publication by G. T. Fechner entitled 
Ueber die Echtheitsfrage der Holbein-Madonna (1871)* 
And, incidentally, a striking demonstration is thus 
provided of the supremacy of experts with a historical 
point of view over artists who go by a canon of beauty 
which belongs to the nineteenth century. 

Bayersdorfer who, in the discussion over the Hol- 
bein Madonna at Dresden, formulated the arguments 
against the authenticity of the picture in the most 
pointed fashion described notably the colouring as 
irreconcilable with the manner of Holbein. At a 
period of advanced 'pictorial' vision that is at the 
beginning of the seventeenth century when the copy 
was painted the painters strove after a harmony of 
colour which entailed the sacrifice of the local col- 
ours. Holbein, on the other hand from the period to 
which he belonged and particularly from his personal 
bent had an objective respect for the local colours, 
and has never falsified the flesh tints through col- 
oured reflections and greenish half- tints. If it is ob- 
jected that for once he might have painted differently 



233 



ARTISTIC QUALITY: ORIGINAL AND COPY 
from his wont, you overlook the fact that end and 
means, spirit and pictorial technique, form and col- 
our, all having an identical origin, are bound to agree 
in a given picture ; and indeed, in the Darmstadt ori- 
ginal the discrepancies, patent to every sensitive eye 
at Dresden, are in no wise to be traced. If, however, 
many amateurs and artists about 1870 found the Dres- 
den picture 'more beautiful* than the Darmstadt one, 
this verdict of taste is to be explained through the fact 
that those judges, as regards their visual convention, 
were still closer to the time about 1630 than to the 
time about 1^30. 

The copyist, by contrast to the creative master, 
takes as his starting point a picture, not life ; and is 
concerned with a vision already realized. In a way, of 
course, there exists no such thing as absolutely ori- 
ginal production. Strictly speaking it is a question of 
difference of degrees. Even a great and independent 
painter has not only seen nature but also works of art, 
paintings by other masters and his own. He depends 
upon a tradition of art. To some extent every painter 
is an imitator and copyist, if only in this, that he paints 
his picture from his own nature studies, drawings, 
sketches. In the professional routine no one can escape 
recollection of the work of others and of his own earlier 
work. The artist in fact is not only father and mother 
to his production, but also the accoucheur. 

We might endeavour to set out a synopsis of de- 
grees. As works to be classified as original in the 

234 



SYNOPSIS OF DEGREES 

highest degree you might put down, say, drawings by 
Rembrandt notably those of his late period or 
drawings by Griinewald. By their very nature, draw- 
ings rank higher than paintings in the table of prece- 
dence we are now establishing. Directness and spon- 
taneity are indissolubly linked with originality. The 
lowest rank is that of copies in the narrow and rigor- 
ous sense of the word. And there are endless quan- 
tities of intermediate stages. 

The truly creative master struggles with the task of 
projecting on the picture surface the vision which 
exists in his imagination. He can approach his goal but 
never reach it, and herein lies the stimulating, excit- 
ing fascination of his activity and also, to be sure, the 
tragedy of his destiny. The copyist faces a task which 
is laborious and soporific but, in his view, perfectly 
feasible. Before his eyes he has the artist's projection, 
which constitutes the real task, and he requires only 
keenness of perception and skill in order to perform 
his work. 

Whoever copies need not be incapable of indepen- 
dent creation. It is even conceivable that the artist in 
question may be more highly gifted than he who pro- 
duced the archetype. The decisive point is, however, 
that the servitude and duty of the copyist's task stamp 
his performance with the character of subordination 
and lack of freedom ; that his mental attitude, who- 
ever he be, is essentially different from that of the 
creative artist. As soon as he copies the painter re- 



ARTISTIC QUALITY: ORIGINAL AND COPY 

nounces his own method of vision. The creative 
master stakes the whole of his intellectual and spiri- 
tual forces, the copyist only memory, eye and hand. 
Whoever feels the difference between growing and 
making is not going to be easily deceived. An original 
resembles an organism ; a copy, a machine. 

Generally the decisive hall-mark of an original is 
the perfect harmony it establishes between pictorial 
imagination and form, between conception and exe- 
cution, between formative intention and the means of 
expression. An original is in harmony with itself. The 
further in time a copyist is parted from the production 
of the archetype, the less is he capable of reaching this 
homogeneousness or even of mimicking it since, 
even if he were able completely to suppress his 
method of vision, he lacks the pigments and mediums 
which went to produce the original. 

Now there do, of course, exist contemporary 
copies, and workshop replicas. With regard to these, 
the decision is often difficult. Even the ambiguous, 
evasive expression 'replica', to which experts often 
resort from uncertainty or may be politeness suggests 
the possibility of producing a perfectly successful fac- 
simile. Even assuming that a master has, with his own 
hand, copied one of his works (usually such tasks were 
no doubt allotted to his assistants) it is to be expected 
that he would be unable in the repetition to reach the 
freshness and vitality of the first creation. It is true 
that masters such as Gerard Dou and Gerard Ter- 



236 



DIFFERENT KINDS OF COPIES 
borch, who work with phlegmatic accomplishment, 
methodically and coolly, probably do not betray 
themselves as copyists of their own works. 

Copies have been, and are being, commissioned and 
produced with motives and intentions that are differ- 
ent from one another. Academic training through the 
imitation of classical masterpieces ; the desire to re- 
tain the duplicate of a sold picture such cases entail 
the striving after accurate reproduction and the com- 
plete effect of the archetype. 

At times independent masters have produced para- 
phrases of older works without suppressing their own 
manner of expression, as for instance Delacroix when 
he translated Rubens into his own language of art. 
And Rubens acted similarly even if not so con- 
sciously when he paraphrased pictures by Titian. 
The conflict, inevitable in these cases, between concep- 
tion and pictorial treatment results in ambiguous or 
hybrid effects. Formerly copies were frequently made 
light-heartedly, without the ambition to approach the 
archetype, notably in order to put on record a com- 
position, to make a reproduction, before the inven- 
tion of photography. In such cases it is easy to con- 
vince oneself that the master concerned cannot be 
credited with the invention of the work. We feel at 
once that we have before us a corrupted, poorly inter- 
preted text. 

As for the faithful, excellent and successful copies 
and only these are disturbing to style-criticism, 

237 



ARTISTIC QUALITY: ORIGINAL AND COPY 
notably if the archetypes are unknown or not available 
for comparison there are some general considera- 
tions which will help us to avoid mistakes. 

Copying is a business which calls for feminine de- 
votion, readiness for sacrifice, patient and never failing 
attention. Copies often are slower in coming into 
being than originals. It is hardly to be avoided that eye 
and hand grow tired at the wearisome task, that every- 
thing does not succeed uniformly. And it is to be ex- 
pected that the copyist concentrates his undistracted 
attention preferably upon the essential passages , which 
are of decisive importance for the total effect, and, on 
the other hand, devotes less care to trifling details. If 
it is a question of a Madonna, he will be more sympa- 
thetic, take more time and trouble, when painting the 
head of the Virgin than when dealing with the land- 
scape background or the ornaments of the architec- 
tural setting. Relaxing and waning of accuracy are not 
infrequently to be observed in the corners or along 
the edges of the copy. 

A second consideration seems to contradict the first 
one. I remember in this connection a conference 
which I attended at the Government Printing Office 
in Berlin. A decision was to be taken regarding new 
currency notes. An art lover on the Committee made 
an eloquent plea for sacrificing the portraits admit- 
tedly mostly lacking in taste in favour of purely 
ornamental designs. At this the representative of the 
Reichsbank protested energetically. Precisely the por- 

238 



THE PROBLEM OF THE CURRENCY NOTE 
trait heads, he urged, were indispensable, a part of the 
picture on no account to be left out. This was because 
it had been noticed that the Bank cashiers, who 
quickly had to make up their mind as to the genuine- 
ness of notes, instinctively looked first at the head, 
the expression of the face, and reached their decision 
accordingly. I have never forgotten this experience 
from the practical domain of detection. 

For it is a fact: a tiny alteration of the ensemble 
of forms scarcely measurable, nay, so small that a 
copyist or forger hardly can avoid it brings about in 
the face and notably round the mouth a considerable, 
immediately apprehended variation of spiritual ex- 
pression. An equally slight alteration of an ornamental 
design is scarcely capable of proof, and in any case not 
noticeable at first glance. 

The portrait need be no masterpiece, yet in every 
case the psyche of a human being impresses itself upon 
the cashier who has seen thousands of notes; and 
the character image, now so familiar, is immediately 
missed by him in the forged notes. 

The expert reacts in no way differently from this 
bank cashier. There exist accomplished copies of en- 
gravings before which one glance at a human face 
saves you from error more certainly than the most 
painstaking comparison of other parts of the plate. 

The copyist is least successful when striving after 
the interpretation of spiritual expression ; and when 
it is a matter of accessories, his eagerness flags. 



239 



ARTISTIC QUALITY: ORIGINAL AND COPY 
I have already noted that drawings tell you more 
about the essence of original work than pictures. Hence, 
when we study drawings, the fundamental difference 
between archetype and copy becomes most easily 
apprehended. Graphology can teach us much about 
the difference between true-born and imitated form. 
The copyist draws warily, directing his eye alter- 
nately at his exemplar and the copy, and is even for 
this reason incapable of achieving the boldly flowing 
sweep of the archetype. 

Even the best copyist cannot avoid misunderstand- 
ings. The master who works direct from nature and 
realizes his own vision, has taken in much more than 
he notes down what he gives is an excerpt, a short- 
hand note, an abbreviation. He may indicate, say, the 
contour of a hat with a slight stroke of the pen ; but he 
has seen and knows the building up of the other side 
and the interior of the piece of head-gear, as also the 
material it is made of. Something of this expert know- 
ledge guided his hand as he drew the contour. The 
copyist has before his eyes the result of a visual action 
in which he takes no part. A tiny projection or twist 
of the original line has a cause which the copyist does 
not know, a significance of which he is ignorant. 

I will try to illustrate by means of an example the 
kind of mistake which a copyist is liable to make. Be- 
fore me there lie two drawings, one the archetype, 
the other a close imitation. In the foreground, out of 
the earth, there rises a stone across whose base there 



240 



THE COPYIST'S MISTAKES 

extends a wavy mass of sand. The copyist has errone- 
ously taken the slight, undulating line for the lower 
edge of the stone, which now in the copy is not con- 
tained in the soil but, on the contrary, stands on the 
ground with an impossible jagged contour. The en- 
semble of forms seems in each case to be almost 
exactly the same, yet the total effect is completely 
different, since the copy has wiped out the special 
illusion caused by the position of the stone behind the 
wavy mass of sand. 

Silhouettes which overlap, foreshortenings and 
concealments of forms, are means of suggesting the 
third dimension on the flat surface. Since the copyist, 
unlike the creative draughtsman, has not seen the 
volumes in space he is incapable of understanding 
more especially those lines which take us in the direc- 
tion of depth. If an outline collides with an outline in 
a different direction, if such an outline is cut through, 
then form becomes partly concealed, turning from 
the picture-space towards the depth. Three-dimen- 
sional appearance, cubic mass and the movement of a 
body in space are conjured up by insignificant con- 
tractions, interruptions, and the end and beginning 
of the stroke of the pen. There is marvellous vitality 
in the intermittent handling of Leonardo's sketchy 
drawings, which nobody can copy without neutral- 
izing the staccato or else if the imitation be mechani- 
cally cautious without losing the suggestive effect. 
In considering gaps and omissions, say, in Rem- 



241 



ARTISTIC QUALITY: ORIGINAL AND COPY 
Brandt's drawings, it is to be borne in mind that the 
master drew for himself and that completeness did 
not come within his intentions. Such forms as adjoin 
the gaps must give the spectator the possibility of 
building bridges in his imagination. To the extent 
that the connection is not provided, it must be 
created. A drawing by Leonardo or Rembrandt, as 
outcome of a vision, is for the spectator nothing but a 
means to let that vision revive. The examination is an 
experiment aimed at establishing whether this en- 
semble of strokes and dots produces the vision. 
Truly it is not a slight performance which in this 
fashion is demanded of the judge. 

The hall-mark of originality is the individual char- 
acter which is peculiar to the work and all its com- 
ponent parts ; in a manner of speaking, the resemb- 
lance to the creator of that which is created. Now 
there also exists an originality of inferior rank ; one 
that has been made, alongside of the one which has 
grown naturally. Especially at a time when native 
strength and personal individuality are held in high 
esteem as being of basic importance for the true 
artist, it is the desire of everyone who handles brush 
and pencil to be himself an original, and when, as is so 
often the case, the necessary resources are insufficient, 
the person concerned wilfully resorts to that which is 
bizarre; and to differentiate this from that which is 
original is far from easy and, indeed, frequently im- 
practicable for contemporaries. Something original is 



242 



RELATION OF ORIGINAL TO BIZARRE 
strange when first seen, shocking and unpleasant; 
something bizarre is striking and entertaining. The 
former is something enduring and permanent and only 
gains in impressiveness ; the latter is a thing of fashion, 
is ephemeral, causes satiety and vanishes before long. 
The relation of original to bizarre is that of the man of 
genius to thefantaisiste. Whoever is creative in a truly 
original sense especially if he be a man of genius 
aims at being self-sufficient ; whoever indulges in the 
bizarre, endeavours to impress his contemporaries or 
to amaze them. 

We often hesitate to use the derogatory word 
'copy* and look in vain for another term. In periods 
of strictly binding iconographical tradition in the 
middle ages and as late as the fifteenth century com- 
positions and pictorial motifs were repeated, without 
any intention of achieving an artistic effect. ' What' , 
not 'How*, was the important thing, and it was far 
from rare that refinement, animation, happy re- 
arrangement of a traditional pictorial idea, were 
realized at a later stage. A free translation into a new 
language was given of the text. More particularly icon- 
like, hieratic images enjoyed a long lease of life, re- 
tained their standing for reasons which had little to do 
with artistic value. At times there may have attached 
to them the prestige of special sanctity or miraculous 
powers ; and they may for that reason over and over 
again have been welcome and desirable to the pious. 

In order to gain the correct standpoint in regard to 



ARTISTIC QUALITY: ORIGINAL AND COPY 
such phenomena, we must free ourselves from certain 
prejudices. The concept of the 'artist', with his 
jealous claim on intellectual ownership, came gradu- 
ally into being from the sixteenth century onwards. 
Previously, it was a matter of illustrating the story of 
the Gospels and the Legends for the benefit of the 
faithful, and of making the saintly character tangible. 
The situations could only be understood, and the per- 
sons recognized, if they were to be seen in familiar 
guise. Owing to the need of making himself under- 
stood, the artist's possibilities of expression were re- 
strained and limited. Neither the painter nor the com- 
munity had ever seen St. Peter, but they knew pic- 
tures of him. Illusion and vague belief connected such 
images with archetypes, which reached back to the 
period of the saint. Veneration was extended to au- 
thentic, true portraits of the Saviour and to Madonna 
pictures which St. Luke, as was thought, had painted 
from life. 

If one realizes the spiritual relationship of the pious 
to the devotional image, then one no longer wonders 
at the faithfulness to tradition which, taking medieval 
iconography as its starting-point, kept alive composi- 
tions, motifs and types. There were times when the 
new pictorial conception could not count on under- 
standing, indeed was regarded as blasphemous. In all 
probability Jan van Eyck, Hugo van der Goes and 
Griinewald did not only satisfy but also disturb their 
contemporaries. No doubt, bearing this in mind, the 



244 




i s . BRUGES MASTER OF 1499, MADONNA WITH DONORS 

Paris, Louvre 
The composition in part goes back to Hugo van der Goes 




26. LUCAS CRANACH, PORTRAIT OF PHILIP OF POMERANIA 

Rheims, Museum 
A record from life 



OBSTACLES FACING ORIGINALITY 
breaking down of the pious convention through mas- 
ters of genius appears all the more worthy of admira- 
tion. Originality was neither asked for nor encour- 
aged; on the contrary It had to overcome strong 
resistance. 



XXXIV 

DEDUCTIONS 'A POSTERIORI 5 FROM COPIES 
REGARDING LOST ORIGINALS 

OUR heritage of works by the Old Masters shows 
many gaps. Destructive action has, in the course 
of centuries, intervened more radically in one place 
than in another. Italian art cities, like Florence or 
Siena, had in essentials remained untouched until the 
greed and eagerness of collectors and dealers not in- 
deed destroyed but carried off the artists' works. 
North of the Alps, on the other hand, icono- 
clastic movements, wars and revolutions have made 
terrible ravages ; nor have they halted before ecclesi- 
astical property, which in the South has remained re- 
latively unharmed. Some countries, districts and 
cities have been looted almost down to the last ves- 
tige, for example France and Holland, and certain 
German cities like Ulm and Augsburg. If style- 
criticism has been successful in assembling and classi- 
fying at any rate a modest remnant of Dutch panel 
painting of the i $th century, this is due to the fact 
that a few examples, more particularly of small 
dimensions, were sent out of the country and thus 
saved from falling into the hands of the iconoclasts. 



246 



THE ARCHAEOLOGICAL METHOD 
In the case of the great Flemish and Dutch masters of 
the i ^ th century we possess only a fraction of that 
which originally existed of their production. Many 
works mentioned in records or in old writings can no 
longer be traced. 

Since there was much copying and imitation going 
on in the Netherlands as I have described, and in the 
spirit that I have indicated we must conclude that 
there exist copies that do not automatically reveal 
themselves as such precisely because the archetypes are 
lost ; and their number is greater than is generally sup- 
posed. 

We turn for instruction to the archaeologists, 
whose efforts, when based on style-criticism, are al- 
most exclusively directed towards deducing lost ori- 
ginals from copies, towards the reconstructing of the 
archetype from one or more imitations. Thus to 
transfer the method, perfected by the archaeologists, 
into the study of ith century painting has occa- 
sionally been productive of valuable results. The 
analysing eye sees through the disfiguring garments 
with which the master in charge of the immediate 
execution has cloaked the body of the archetype. 

In each case questions such as these should be asked: 
'Is the artist to whom we owe the execution also the 
one whom we may credit with the invention? Does 
the manner of painting, chronologically and as regards 
artistic value, accord with the conception? Can we 
notice, within the composition, a contradiction, a 



DEDUCTIONS <A POSTERIORI 5 
break, a sudden transition, a lack of logic ? Do the 
parts fit in with each other and the whole ? A negative 
answer to these questions, or to one of them, justifies 
the deduction 'copy' or 'free copy' that is, a mix- 
ture of copy and independent work. We aim at dis- 
covering the author of the archetype and, in so far as 
Flemish ijth century is concerned, we can restrict 
our search to a relatively small circle, since experi- 
ence has taught us that, as creators of types, there are 
but few masters to be considered, namely, Jan van 
Eyck, Roger van der Weyden, Dirk Bouts, Hugo van 
der Goes and Jerome Bosch. 

I will try to illustrate by one example how such an 
investigation can be carried out successfully. 

In the Louvre there is a Madonna picture which has 
achieved undeserved fame because a French scholar, 
on the basis of the initials J. P., claimed it for Jean 
Perreal. This idea was mistaken. The initials relate to 
the couple shown as donors. The picture is Flemish, 
painted about 1500 the costume also tells us that 
and it is by an unimportant painter, whose petty man- 
ner may be recognized in some other panels. 

The Madonna sits enthroned in the middle; the 
male and the female donor are shown as half-lengths on 
the left and right behind the sides of the throne. If 
anything has been created independently by the 
painter, then it is the couple of donors who gave him 
the commission. Now it is the portraits which are the 
weakest part of the whole ; they enable us to measure 



248 



A PICTURE IN THE LOUVRE 

a spiritual narrowness, which also reveals itself in the 
dry and pedantic elaboration of the decoration of the 
metal throne and of the piece of brocade. The face of 
the Madonna is empty and inexpressive possibly 
'beautified' by a restorer. On the other hand in the 
play of folds in the Madonna's garments, and in the 
Infant Christ with its vigorous movement, will and 
temperament are active as expressive forces which lie 
beyond the possibilities of the painter to whom the 
execution is due. The tubes and ridges of the dra- 
pery, boldly crossing, and colliding with, each other; 
a dramatic and imaginative language of form, in a poor 
translation it is true, point to Hugo van der Goes. The 
Infant Christ, lying across the lap of His Mother and 
raising the upper part of the body, partakes in the 

face, in the movement of the lean bodv, and in the 

* j 7 

fingers, spasmodically bent inwards of the emo- 
tional life of the Ghent master, with its intense yearn- 
ing. The motif of mo vement does not seem to be fully 
justified or consistent in this context. One does not 
see who or what has caught the attention of the Child 
and caused its action. The motif is borrowed, taken out 
of a different connection. Probably there were in the 
archetype female saints close to the Madonna, and the 
Child turned vivaciously to one of the saints, perhaps 
St. Catherine. 

We may hence insert, as a welcome increment, 
some parts of this inconsiderable picture into our 
idea of the activity of Hugo van der Goes. 



249 



XXXV 
WORKSHOP PRODUCTION 

A PICTURE is, in accordance with the view which 
nowadays has acquired general validity, the 
creation of one single person, who was the only one 
working at it from its conception down to the last 
brush-stroke and who is held to be solely responsible 
for it. It is with difficulty that we accustom ourselves 
to note, and to draw deductions from the fact, 
that this was not always so. The painters of bygone 
days were at the head of a workshop ; they worked to- 
gether with journeymen and apprentices. Now this 
circumstance is, it is true, occasionally taken into ac- 
count by style-criticism, and the derogatory term 
'workshop production 5 is introduced especially in 
cases where an original can be compared with a 
weaker replica which is nevertheless identical in com- 
position and pictorial technique. Without a doubt col- 
laboration was far-reaching, and is not restricted to 
the cases in which the dissatisfied eye looks in vain for 
the expected quality, or can establish an inequality of 
merit in the pictorial execution. 'Autograph' quality 
is questionable also in cases where the defects of exe- 
cution are by no means patent. Gifted journeymen 



EVIDENCE OF RECORDS 

and even boy apprentices may, as regards ability, have 
been equal, and even superior, to the head of the 
workshop. It is to be borne in mind that a journey- 
man did not, as an Academy student does nowadays, 
create himself a master the moment he imagined he had 
learnt enough ; on the contrary, economic conditions 
might cause him to persevere as an assistant, all the 
more so as the master had an interest in keeping 
skilled collaborators. 

Jan van Scorel was, as regards artistic gifts, at least 
a match for his third master, the Amsterdam painter 
Jacob Cornelisz. At the age of about twenty-four he 
received from Jacob, for 'ingenious and skilful 5 work, 
a certain sum of money and, in addition, permission 
to paint in his free time some pictures for himself. 
This Karel van Mander tells us. It is scarcely to be as- 
sumed that Jan van ScoreF s collaboration lowered the 
value of the pictures which about i2o issued from 
this Amsterdam workshop. One would rather deduce 
that they showed a less stiff and rejuvenated manner. 
And it is necessary to reckon with similar conditions 
also elsewhere 

Agreements, whose texts still exist, throw a vivid 
light on the methods of work which were customary 
in the studios of the i^th and i6th century. Particu- 
larly notable is the lawsuit about an altarpiece that 
Albert Cornelisz. was to supply about i2o. In the 
agreement with the people at Bruges who had or- 
dered the picture, it was stipulated that the master 



WORKSHOP PRODUCTION 

was to paint everything essential especially the flesh 
with his own hand. Since this was laid down by 
agreement we may conclude that even partially 'auto- 
graph' quality was not supposed to be a matter of 
course, that Albert Cornelisz. in other cases did not at 
all intervene with his handiwork, but contented him- 
self with providing the preparatory drawing and super- 
vising the work. And up to a point such a procedure 
may have been general in the Flemish workshops of 
the 1 6th century. Things do not seem to have been 
very different in Venice, say in the studio of Giovanni 
Bellini. Of the lawsuit at Bruges I have spoken at 
length in the twelfth volume of my Geschichte der Alt- 
niederlandischen Malerei, on the basis of the document 
published by the late Mr. Weale. Many masters were 
something akin to owners of business-concerns, or 
heads of factories. 

Frans Floris was, about 1550, the leading master in 
Antwerp, and an organizer like Rubens half-a-century 
later. As Karel van Mander relates, one counted one 
hundred and twenty painters who had learnt and 
worked under him. Very informative as regards the 
way in which things were run here are further state- 
ments by the same writer. Floris made the prepara- 
tory drawing of the composition with chalk, and then 
let his journeymen do the underpainting and ' con- 
tinue' ; he would tell them to 'bring in these heads in 
such and such a place' . The fact is that he kept in the 
workshop, as models and exemplars, a number of 



2 S 2 




2 7 . WORKSHOP OF LUCAS CRANACH, PORTRAIT OF PHILIP OF POMERANIA 




28. HANS CRANACH, PORTRAIT OF A MAN 

Lugano, Castle Rohoncz Collection 

Signed and dated 1534 H C 



THE APPRENTICES 

studies of heads which he himself had painted on 
wood. Such studies by Floris have survived in con- 
siderable numbers. Nobody took any exception to 
this procedure, and Karel van Mander records with 
praise that, as a result, the pupils achieved sureness 
and independence. 

Completely 'autograph' quality was the exception. 
Even Diirer, so conscious of his duty and conscien- 
tious, stresses in his letters to Heller as something ex- 
traordinary that he had let no apprentice take a hand 
at his work, and hints that he used to execute "ordin- 
ary pictures' (gemein gemal), which paid him better, 
with the assistance of pupils. 

The boy apprentices were there not only to learn ; 
they were also auxiliaries, who were compensated 
with food and lodging for their performance. They 
bound themselves to serve for many years, and the 
master was entitled to hope that as years went by they 
would prove increasingly useful. After all it was im- 
possible that year after year they should all day long 
grind colours, clean brushes and perform other menial 
services ; on the contrary, they must have taken some 
part in the actual painting. They could not possibly 
learn how to paint without painting. 

As to the running of the workshop of Rubens, we 
are well informed. The letters of the master contain 
certain passages which tell us a lot. In the year 1619 
Rubens writes to William, Duke of Neuburg: 'The 
St. Michael is a difficult subject. I fear that I shall 



WORKSHOP PRODUCTION 

have difficulty in finding among my pupils someone 
who will be capable of executing the work, even if I 
provide the drawing. In any case it will be necessary 
for me to retouch the work with my own hand.' The 
picture is in the Alte Pinakothek at Munich. 

The superiority of genius, its inimitability, will 
naturally become patent to sensitive eyes. Rubens, 
especially late in life, painted pictures which in every 
sense are 'autograph*. Style-criticism can in this case 
attempt with some confidence to draw the boundary 
lines of workshop performance and the intervention 
of a collaborator as gifted as Van Dyck. 

So far as medium masters are concerned, it is only 
the part of incapable and perhaps self-willed appren- 
tices that can be segregated: skilful, well-trained ones 
remain concealed. If I describe a picture as a work by 
Bernard van Orley I am, strictly speaking, stating no 
more than that the master has made his style clearly 
and uniformly visible, despite a collaboration not to 
be checked-? by capable apprentices. 

A painter who for financial reasons increases the 
production of his workshop Lucas Cranach in his 
Wittenberg period offers an instructive example 
does not so much raise the apprentices to his own 
level: rather, he descends to theirs; he creates a 
language of form and manner of painting which can be 
taught and imitated, and gives his production an im- 
personal character. 

It is extremely instructive to compare the surviving 



CRANACH'S METHODS 

'autograph 5 portrait studies by Cranach, notably those 
in the Museum at Rheims, with the pictures as exe- 
cuted. They were put to the same use as the 'heads' in 
the Floris studio. The master seized upon the essen- 
tials of the individual face by means of the draughts- 
man's shorthand. His recordings give pleasure through 
their striking unambiguousness and are lacking in 
detail. It is obvious that the master thought of the 
purpose, of the usableness, of the exemplars and 
wished to give the apprentices definite and unmistak- 
able directions. In the pictures as executed their 
relation to the recordings is that of copies the 
scheme is not in the least enriched, and the sim- 
plicity after the manner of the woodcut strikes one, 
when taken over into painting, as empty and rigid. 

If we disregard the portraits, copying in the strict 
sense of the word was not practised in the workshop 
of Cranach. Obviously the master felt responsible for 
the composition, regarded it as a matter of honour to 
provide variants. What we see is over and over again 
the Judgment of Paris, similarly conceived, similarly 
painted with the same types, but with varied grouping 
and changed motifs of movement. The Maitre des demi- 
figuresproceeded in the same way, while Joos van Cleve, 
the Master of Frankfurt, and other Netherlandish 
painters caused more or less accurate copies to be made 
in the workshop. It would be utterly mistaken to de- 
duce 'autograph' execution from the principle of vari- 
ation, pedantically clung to in the Cranach workshop. 



WORKSHOP PRODUCTION 

It has not proved possible to differentiate, in the 
later work of Cranach, between him and his sons. At- 
tempts were, indeed, made to segregate the part 
taken by the elder son, Hans; but these had to be 
abandoned after the discovery of two panels, signed 
by Hans Cranach, which as regards style mark no dif- 
ference from the homogeneous mass of pictures which 
about 1 3 issued from the father's workshop. Both 
pictures are now in the Castle Rohoncz collection at 
Lugano. 

Netherlandish altarpieces with wings, and devo- 
tional pictures, which especially in Antwerp were 
produced in very large numbers for the market and 
for the export trade, were not infrequently else- 
where, more particularly at Cologne and Bruges, 
supplied with portraits of the donors. There exist 
Flemish wings of altarpieces with portraits by Barthel 
Bruyn, and others in the style of the Antwerp Man- 
nerists with likenesses by the master known as 
Adriaen Ysenbrant. 

A salutary education towards scepticism and doubt 
is experienced in studying the inventories published 
by J. Denuce. 1 One gets scared by the multitude of 
painters' names with which we cannot link any con- 
ception of style ; also by the large number of pictures 
which are put down as copies. 

The critic of style must frequently expect that joint 
work by many painters which has become so alien to 

1 J. Denuc<, De Antrrerpsche ' Konstkamers* (The Hague, 1932). 




29. STYLE OF THE ANTWERP MANNERISTS 
WINGS OF AN ALTARPIECE 

London Art Market 
The portrait of the donor inserted by Adriaen Ysenbrant 




By gractous permission of His Majesty The King 

30, HANS HOLBEIN, PORTRAIT OF A MAN 
Drawing. Royal Library, Windsor 



JOINT WORK 

us. Joachim Buecklaer painted the clothes in por- 
traits by Antonio Moro. This is stated by Karel van 
Mander. Houbraken, in speaking of Kneller, remarks, 
probably with some exaggeration, that it was the gen- 
eral rule in England for the master to paint only the 
face and hands, while the clothes and subsidiary de- 
tails were painted by others. 



XXXVI 
ON FORGERIES 

MANY of the principles which I have outlined 
when treating of copies apply also to forgeries, 
only that the intention to deceive causes an ethical 
discord to penetrate into the domain of aesthetics, and 
that a cunning, stealthy attitude of mind replaces the 
circumspectly andhonestlyplodding one of the copyist. 
In face of the disguise, the affectation and hypocrisy 
which defile art, the connoisseur becomes a crimino- 
logist. 

At the leading string of a master the forger moves 
most nearly with security and achieves his aim most 
easily by copying. In so doing he runs, however, the 
risk of being caught out, as the archetype generally 
is known, and a glance at it threatens to expose the 
fraud. For this reason experienced and ingenious for- 
gers aim at extracting from several archetypes an ap- 
parently new whole. In putting together heterogene- 
ous parts they give themselves away. They will imitate, 
say, the i th century manner of painting, but will 
choose a motif of movement characteristic of the i6th 
century; or they place a headgear of the i6th century 
on a cranium with a coiffure of the i jth century. Con- 

2 S 8 




3i. PORTRAIT OF A MAN 
Forgery based on the Holbein drawing, reproduced Plate 30 




32. HANS HOLBEIN, PORTRAIT OF (?) ANTOINE, DUKE OF LORRAINE 
Berlin, Picture Gallery 



CHARACTERISTICS OF THE FORGER 
fusion of styles and disharmony are typical of a forgery 
even more than of a copy. Homogeneousness from the 
moment of its birth the hall m$rk of originality is 
lacking. 

The forger will copy, closely and cleverly, a Hol- 
bein drawing; in the reverse, moreover, so as there- 
by also to cover up his traces a little. The beard and 
coiffure of a given male head denotes the time about 
i 30. The forger places on it a tall cap, of the kind 
that was worn about 1490, and adds a landscape back- 
ground in the style of Memling. 

The forger is an impostor and a child of his time, 
who disowns the method of vision which is natural to 
him. Once the consequences of this disastrous position 
are clearly realized there will be no difficulty in per- 
ceiving the characteristics by which his concoction 
differs from an original. Oscillating between uneasy 
cautiousness and brazenness, afraid lest his own 
voice may grow too loud and betray him, he suc- 
cumbs to the prejudices of taste that belong to his own 
period the moment he will give 'beauty 5 . His pathos 
sounds hollow, theatrical and forced, since it does not 
spring from emotion. 

The greatest difficulty which besets the forger is 
that of achieving the decisiveness of the original work 
a decisiveness which springs precisely from that 
naivete and certainty of instinct which the forger 
lacks. Deliberation and consciousness reveal them- 
selves in artistic form as lack of life or else hesitation. 



ON FORGERIES 

The style of the forger's period betrays itself in the 
expression often through sentimentality, sweetness, c 
desire to please and insipidness. The forger differs 
from the master, into whose skin he slips, also in this, 
that he has but an inadequate knowledge of the object 
that the master in question had before his eyes. He 
does not know how a coat was cut and sewn in the 
i th century. From our archaeological knowledge we 
are in a position to discover his mistake and unmask 
him easily, especially if he has not copied closely but 
has dared to vary. 

The aristocrats among the forgers, a Bastianini or a 
Dossena, did not strictly speaking work by copying or 
combining ; on the contrary, they harboured the illu- 
sion that they had penetrated so deeply into the crea- 
tive methods of previous ages, that they could express 
themselves in the spirit, and in the style, of the Old 
Masters. They dared to push forward, from a *pla- 
tonic' production of which also a gifted connoisseur 
is capable into the real one. Success was granted 
them only for a brief while. They took in only their 
contemporaries, and even these not permanently. 
Their works partake of none of that timid pettiness 
which is characteristic of ordinary forgeries: on the 
contrary, they display boastfully an audacity which, on 
their becoming unmasked, transforms itself into fool- 
hardiness. 

Since every epoch acquires fresh eyes, Donatello in 
1930 looks different from what he did in 1870. That 



260 




33. HANS HOLBEIN, PORTRAIT OF A MAN 
Vienna, Picture Gallery 

1541 




34- PORTRAIT OF A MAN 

Forgery based on the Holbein portrait at Berlin, Plate 32, but utilizing 
the hands in the Vienna portrait by the same master, Plate 33 



THE 'NO' MAN AND THE 'YES' MAN 
which is worthy of imitation appears different to each 
generation. Hence, whoever in 1870 successfully pro- 
duced works by Donatello, will find his performance 
no longer passing muster with the experts in 1930. 
We laugh at the mistakes of our fathers, as our descen- 
dants will laugh at us. 

If only for this reason not to speak of other con- 
siderations it was a silly business when, towards the 
end of 1908, the Cologne Madonna with the Sweet 
Pea was declared to be a work of the early nineteenth 
century. I wrote at the time a brief article against this 
aspersion, and formulated in it the phrase: Forgeries 
must be served hot, as they come out of the oven. As 
the 'No' man imagines that he stands above the 'Yes' 
man and probably also to others seems to stand higher 
critics will always feel the impulse to attack genuine 
works in order to win the applause of the maliciously 
minded. The 'Yes' men have done more harm, but 
have also been of greater usefulness, than the rigorous 
'No' men, who deserve no confidence if they never 
have proved their worth as 'Yes' men. 

After being unmasked every forgery is a useless, 
hybrid and miserable thing. Bastianini was perhaps a 
talented sculptor ; in the style of the past he was, how- 
ever, only able to bring abortions to the world. 

Discussions and polemics regarding forgeries are 
seldom of long duration. This is the typical sequence 
of events: the work emerges from obscurity, is ad- 
mired, then seen through, condemned, and finally 

261 



ON FORGERIES 

sinks into limbo. Behind it are left nothing but silent 
shame among those that were concerned in the epi- 
sode, and superior smirks among those not so con- 
cerned. 

A forgery done by a contemporary is not infre- 
quently successful from being pleasant and plausible, 
precisely because something in it responds to our na- 
tural habit of vision ; because the forger has under- 
stood, and misunderstood, the old master in the same 
way as ourselves. Here is, say, a 'Jan van Eyck' thus 
the great venerable name, and yet something that has 
attractiveness in conformity with the taste linked up 
with our own time : how could it fail to gain applause 
under such circumstances ? To many lovers of art a 
false Memling is the first Memling that gave pleasure. 

I remember how, years ago, an art dealer submitted 
to me drawings after Holbein's Dance of Death, claim- 
ing them as originals from the master's hand and re- 
ferring to the fact that pathos and emotional expres- 
sion made a stronger appeal in the drawings than in 
the woodcuts. The observation was accurate; the 
drawings were, however, imitations of recent date. 
Holbein, in the woodcut, in the design upon a small 
scale, has made the motifs of movement not the 
facial features the vehicle of expression: and in this 
he followed the sure sense of style characteristic of 
him. The copyist took as his starting-point neither 
vision nor the requirements of the woodcut, but the 
intellectual significance of the tragic theme ; and, by 

262 




35". PORTRAIT OF A MAN 
Forgery based on the head of Canon van der Pacle, Plate 2J 




36. PORTRAIT OF A MAN 
Forgery in imitation of Antonello da Messina 



SUCCESS OF FORGERS 

petty strokes of the pen, he heightened the expression 
of fear and distress in the heads. 

Above all things I would not wish that my argu- 
ment produced the impression that I feel sure of my- 
self. This is by no means the case. Not only I, but also 
my teachers for whom I have the greatest regard 
have been taken in though in truth, it seemed im- 
possible to understand, later on, how this had come 
about. 

The eye sleeps until the spirit awakes it with a ques- 
tion. And the question 'Is this work ancient or not? 5 
will at times not be asked, especially not when a 
dealer deserving of confidence submits the object with 
the power of suggestion springing from a good con- 
science and a demand for a high price. 

Novel forgeries tend to be most immediately suc- 
cessful. It is easier to deduce from certain character- 
istics 'This is the work of the forger I know' than to 
argue negatively 'This cannot be genuine '. In order to 
give pictures the look of age, the forgers imitate the 
cracks of the stratum of colour, and this all the more 
keenly since the wrinkles in the skin of the picture are 
noted by less gifted amateurs as the only indications of 
age and genuineness. There exist many genuine pictures 
which show no cracks, but these are never absent in 
false ones. The craquelure caused by age differs more or 
less clearly from the one achieved artificially. The 
primitive method of making cracks by drawing or 
scratching with pencil or brush is held in contempt 



263 



ON FORGERIES 

by the forgers of our days. It is customary to resort to 
the trick of producing false cracks by a chemical action 
say by sudden heating which causes a coating and 
breaks up the colour surface lattice-fashion. 

The natural cracks penetrate into the gesso prepara- 
tion, while the artificial ones reach no farther than the 
colour surface. The imitation, howsoever brought 
about, lacks the capricious playfulness and irregularity 
of the network which has come into being gradually 
and under the influence of climate. The appearance 
changes according to the character of the pigments 
and the greater or smaller body of the impasto of 
colour. In one and the same surface of colour, the 
cracks will now be very noticeable, now not at all or 
very faintly discernible. 

Accomplished forgers make successful use of old 
pictures, which they clean radically often down to 
the gesso preparation in order subsequently to 
superpose their forgery, glazing carefully and treating 
with the utmost delicacy the craquelure, which they 
leave exposed. The connoisseur is in such cases thrown 
back upon his sense of style, since the examination of 
pigments does not help him. 

Genuine old pictures are made more valuable 
through forged signatures. It is, naturally, more con- 
venient and hopeful to supply a good picture by Jan 
van Kessel with a Ruisdael signature, than to produce 
a picture by Ruisdael. Signatures of obscure masters 
have often been cleaned off. 



264 



FALSIFICATION 

More danger has come to attach to the falsification 
a defiling of works of art which is difficult to com- 
bat than to the forgery in the strict sense of the 
word. Let us say that a dealer possesses a Dutch 
i yth century landscape which has suffered greatly. 
From certain indications he considers it though 
wrongly to be a work by Hercules Seghers, all the 
more gladly as the works by this master are scarce and 
valuable. He hands the picture over to an able re- 
storer, supervises its cleaning, and supplies repro- 
ductions of genuine works in support of his attribu- 
tion and in order to instruct the restorer. The latter, 
without any evil intention, is thus inspired to re-inter- 
pret certain passages in the picture 'in the style of 
Seghers'. Under the delusion that he has in front of 
him a work by the master, he restores it. By slow de- 
grees, proceeding from case to case, the bonajide re- 
storation approaches the malevolent falsification. At 
times pictures in poor condition have been shown to 
me. Of one such I will have said for instance: 'This is 
in the manner of Holbein' . And before long it was 
once again submitted to me, neatly completed and 
with beautiful clearness showing the style of Holbein. 

A picture by Vermeer is something exceptionally 
precious. Of this master the dealers are dreaming. As 
regards their conception, his works do not differ over- 
much from those of other and much smaller masters ; 
the magic of light, colour and the individual technique 
of dots give his pictures their singularly exceptional 



ON FORGERIES 

quality and value. More than once has it happened that 
modest Dutch landscapes and scenes from daily life 
have been worked over in an attempt to give them, 
through vivifying dots of light, the appearance of Ver- 
meer's unique handling of the brush. Tame Dutch 
pictures have often, by the addition of bold brush 
strokes, been falsified into works by Frans Hals. 

As the forgers, in conformity with their view of 
their activities, are manufacturers they often produce 
several versions of a fake: and it may be particularly 
noted that duplicates have emerged from the Belgian 
workshops which, during the last few decades, have 
abundantly seen to the supply of early Netherlandish 
panels. Machines are identical, while organisms re- 
semble each other. 



266 



xxxvn 

ON RESTORATIONS 

THE business of the restorer is the most thankless 
one imaginable. At best one sees and knows 
nothing of him. If, out of his own invention, he has 
provided something good he has got mixed up with 
the dubious company of the forgers ; and with the de- 
spised one of the destroyers of art if what he has done 
is bad. His accomplishment remains out of sight, his 
deficiency leaps to the eye. Judgment regarding the 
performance of the restorers is even more unreliable 
than that regarding works of art. And that is saying 
something. 

Restoration is a necessary evil; necessary, inas- 
much as threatening decay can be stopped by the lay- 
ing down of blisters, stabilization of the pigments, 
strengthening of the ground that carries everything. 
Moreover artistic value can be increased through 
cleaning, through the removal of later disfigurements, 
of retouches and of varnish, darkened or even ruined 
and gone opaque. Thirdly and here the intervention 
begins to become of doubtful value the restorer sup- 
plements, fills in holes, from a delusion of being able 
to re-establish the original condition. 

267 



ON RESTORATIONS 

Even the purely preserving action is accompanied 
by risk. The old canvas has, say, decayed; so new 
material is glued to the reverse of the old one. This 
entails ironing, not infrequently to the detriment of 
the impasto of the pigments. 

The removal of the old layer of varnish, be it by the 
dry method through rubbing with the hand, or by 
means of spirits, is not always effected with the neces- 
sary circumspection. Something of the original colour 
can easily be attacked. If the original layer of colour is 
grainy and rough, the darkened varnish has settled in 
the depths and can hardly be rubbed off without in- 
jury to the original paint. Incidentally in many cases 
the endeavour to remove the old varnish radically, 
down to the last vestige, appears by no means so de- 
sirable as to justify running the risk which I have in- 
dicated. 

We can remember many sensational incidents over 
which the newspapers busied themselves. A restorer 
had put right a picture that is he had removed the 
old varnish and perhaps also some repaint. At once 
accusations were heard that he had, 'overcleaned 5 the 
picture, rubbed down its genuine glazes and reduced 
its artistic value. Such an indignation usually ex- 
pressed by people who never had paid any attention 
to the picture before it was ' damaged' is mostly 
unjustified, if only for the reason that the picture may 
have been overcleaned already before it was last 
cleaned, and that, strictly speaking, only someone 



268 




37. NARRATIVE SUBJECT 

Forgery dating from the middle of the l$th century, in subject and costume full of 
childish impossibilities. There exist many forgeries produced by the same worksho^ 




38. BRUGES MASTER, MADONNA WITH TWO FEMALE SAINTS 

Formerly London Art Market 
Much rubbed, painted about 1480 



THE PICTURE AFTER CLEANING 
who was present at the restoration ought to have a 
right to judge. Moreover the hard, cool, and naked 
appearance shown by the picture immediately after 
being cleaned proves in itself nothing against the 
restorer. Our taste depends on convention. We are 
not accustomed to perceive the original condition, 
more particularly so, for example, in a Gallery like 
the Louvre, where almost every picture, under many 
layers of dull varnish, disproportionately warm and 
dark, shows the cheapest form of harmony. A cleaned 
picture, among such as are not cleaned, looks over- 
cleaned. Our eyes are enervated and spoiled. The 
aspect worn by the pictures, when they originally 
left the workshops, would shock us as being crude 
and motley. The earlier restorers knew this well and, 
after cleaning the pictures, used to make them 'old' 
that is warm and 'harmonious' once more, by 
means of coloured varnish. A continuous change in 
the demands, in the prejudices, is to be expected. 
Especially in German museums one has got accus- 
tomed to the appearance of cleaned, and occasionally 
over-cleaned, pictures. 

The activity of the restorer becomes highly pro- 
blematical the moment there presents itself the ques- 
tion of making up for deficiencies that is, of filling 
gaps or revivifying passages which have been rubbed. 
Here the various wishes, demands and aims part com- 
pany. The historian, to whom the work of art is a 
record, opposes, from his standpoint, with full justi- 



269 



ON RESTORATIONS 

fication, that kind of restoration which goes beyond 
preserving and exposing. He demands to see clearly 
what is left of the original, but wishes it also not to be 
concealed from him that something of the original is 
missing. It is precisely the successful re-integration 
that is distasteful to him: the unsuccessful one he, of 
course, detects easily and can make allowance for. 
The picture-owners, in whose service and in confor- 
mity with whose wishes the restorer works collec- 
tors or dealers take up a different standpoint from 
that of the scholar. What they demand is not so much 
the document which has been cleaned and gives reli- 
able information as, rather, the maximum of value 
and, indeed, not only artistic value but also market 
value. Every damage, as long as it remains visible, 
lowers the market value. The restorer must conceal 
such damage. The serious lover of art and the museum 
official, who supervises and directs the work of the 
restorer, are inclined to side with the scholar. And, 
as a matter of fact, the purist faction has lately gained 
adherents. Now and then you find in public galleries 
carefully cleaned pictures whose defective portions 
have been left open say, have been filled in with a 
neutral tint. There is this to be submitted in favour of 
such a procedure, that the best restorer is ineffectual 
when it is a question of filling gaps, especially if it is a 
question of parts which are of fundamental importance 
for the total impression of the picture. 

The decision apparently unavoidable against 



270 



PRACTICAL DIFFICULTIES 

every re-integrating restoration is, however, beset 
with practical difficulties. If part of the original pig- 
ments are missing in a panel of the i4th century, it is 
still possible to derive some enjoyment from what is 
left, and in one's imagination to fill the gaps which 
have remained open. The position changes, however, 
if gaps are visible in the midst of a picture of the i yth 
century. They do away with the illusion of a spatial 
whole, and destroy the effect. In every case it must 
be carefully considered whether a more or less ques- 
tionable addition, made by a restorer, is not to be 
regarded as the lesser evil; just as a surgeon always 
should ask himself whether the success to be expected 
from an operation is so great and so certain that it out- 
weighs the danger entailed by the operation. It is, no 
doubt, possible to choose a middle course, namely, to 
fill in the hole in such a fashion that the defective pas- 
sage does not strike the eye as something that disturbs 
the general effect, but yet becomes obvious if you 
look closely. Such a procedure has the defect of all 
half-measures. 

As long as works of art in private ownership are 
regarded as representing financial values, so long will 
the restorer again and again find himself forced into 
the part of the forger. 

There exist underground connections between the 
workshops of the restorers and of the forgers. Years ago 
I saw in the possession of a London dealer a pretty 
Bruges picture of about 1480 which was greatly 



271 



ON RESTORATIONS 

rubbed, a full-length figure of the Madonna with two 
female saints. A Belgian restorer then got hold of it. 
I do not know whether he restored it: in any case I 
have not seen it again. But a forgery, based on it, did 
turn up, considerably larger and more imposing than 
the archetype. A small misfortune had, incidentally, 
befallen the forger. St. Catherine, receiving the ring 
from the Infant Christ, was depicted; the ring was, 
however, no longer recognizable in the poorly pre- 
served original. In the copy, the Infant Christ busies 
himself quite unaccountably with the finger of the 
saint, since He has no ring to bestow. 



272 




39- MADONNA WITH TWO FEMALE SAINTS 
Forgery based on the picture by the Bruges Master, Plate 38 




40. REMBRANDT 

MOSES SHOWING THE TABLES OF THE LAW TO THE PEOPLE 

not l Moses breaking the Tables of the Law' 

Berlin, Picture Gallery 

1659 



XXXVIII 
ON ART LITERATURE 

LANGUAGE is poor and inadequate, but it is 
nevertheless the only vehicle at our disposal; an 
obtuse instrument which we must untiringly try to 
perfect. At the same time we should bear in mind 
that sharp knives easily become blunt. Emotions re- 
semble butterflies: speared by the needle of the word, 
they lose their life. All that is said on art sounds like a 
poor translation. 

The arts have a common root, are interconnected 
in the depths: poetry, music and the arts of the eye. 
Whoever is bound to convey by word the impres- 
sion of a picture or a piece of sculpture, finds him- 
self impelled towards poetical expression, while his 
intelligence cautions him to avoid poetry. Not wholly 
unjustifiably it has been said by somebody that one 
ought to be musical in order fully to understand for- 
mative art. There is some truth in this sweeping 
maxim inasmuch as music, in preference to all other 
arts, is absolute art, and hence can put us on the 



273 



ON ART LITERATURE 

direct road to that which has the characteristics of 
specific art. 

If one enquires into art literature which has 
assumed such gigantic proportions with reference 
to substance, fertility and permanent value, one will 
find that there remains little enough, nothing more 
than the translation into speech of visual impressions. 
Most certainly the word can never replace the optical 
experience, but he who has some command of lan- 
guage may well be able to help the man who hears 
better than he sees, to 'understand' the work of art 
as the inappropriately rationalistic expression goes. 

The higher the artistic value, the deeper the im- 
pression, the farther does description depart from 
sober and matter-of-fact chronicling, in order to ap- 
proach poetic re-creation, which, it is true, must be 
checked with circumspection if it is not to degener- 
ate into obscurity and empty word-play. You cannot, 
let us say, by means of words produce so graphic an 
impression of the type of woman of Hugo van der 
Goes that a work by this master might be recognized 
as such from that account ; but the description can de- 
finitely, as a guiding interpretation, deepen the im- 
pression and increase the capacity for recognition in 
anybody who has the work before him. And in so do- 
ing it performs at any rate something. We say, for in- 
stance, 'the lean, stark, pale, bony forms have been 
moulded from within by profound and sublime 
thought, by emotional distress and struggle 5 . It is pos- 



274 



VARIOUS WRITERS ON ART 
sible for an imitative artist to reproduce a picture thus 
described, but, since he lacks the inspiring force, he is 
unlikely to get beyond empty masks and caricatures, 
sorrowful expressions bereft of cause, and seriousness 
with no foundation. 

The more deeply observation and notation have 
penetrated into spiritually emotional existence, the 
better will the reader who, however, must not only 
be a reader be enabled to carry out an investigation 
based on criticism of style, and especially to unmask 
copies and forgeries. 

There do not exist many authors whose literary 
capacity is on a level with their understanding of art. 
Among the living men in Germany the one whose per- 
formance, in interpreting by language visual experi- 
ences, stands out, is perhaps alongside Heinrich 
Wolfflin Wilhelm Fraenger. 

The ideal of the art scholar, expressing himself by 
means of language, was perhaps realized in Carl Justi. 
In him extensive knowledge of historical facts united 
itself with the capacity to experience the process of 
artistic activity. I refer the reader to his book on Vel- 
azquez. From the impression which the work of art 
has produced on him, he immediately conjures up 
the situation in which the master found himself; that 
which was demanded of him, that which he wanted. 
He provides psychological interpretations of the vis- 
ible from knowledge of historical circumstances. 

The re-creation of the work of art, in which but few 



ON ART LITERATURE 

have succeeded, is something very different from the 
itemized description which the writers of catalogues 
provide with more or less ability. It has partly become 
superfluous, alongside of the photographic reproduc- 
tion, and is largely ineffectual, since the unfortunate 
reader is hardly capable of constructing a connected 
idea of the whole from a large number of data. The 
enumeration of colour values, which in many cata- 
logues of recent date is meant to supplement the black- 
and-white illustrations, demands from memory the im- 
possible. The courageous and interesting attempt 
made in the catalogue of the Donaueschingen Gallery, 
published in 1921 in which the colours are itemized 
with letters and figures in conformity with Ostwald's 
plates, has remained a curiosity. 

So far as I can see, the endeavour to overcome the 
barren pedantry of the customary catalogue descrip- 
tion has been successful but once, namely, in Rudolf 
Eigenberger's volume on the Gallery of the Vienna 
Academy. Here the mirror of words has, successfully 
and in an exemplary fashion, caught the total effect, 
the artistic significance of each picture ; and the indi- 
vidual data regarding form and colour are not just set 
out one after the other, but on the contrary dovetailed 
into a connected whole. 

A double request is addressed to description and 
cataloguing. For one thing we want to have the subject 
explained, to be enlightened about iconography, and 
to have laid down for us what the master had to do, 



276 



REMBRANDT'S 'MOSES': A CORRECTION 
and what he intended. Secondly, we want a descrip- 
tion of that which is visible, a translation of the im- 
pression received, and a statement of what the master 
has done. Analysing erudition is to be avoided in per- 
forming the second task: the impression must be re- 
ceived at first hand and with our senses fully recep- 
tive ; and it is to be canalized into the form provided 
by language, without prejudice, and without reliance 
on thought. 

An instructive illustration of the interconnection 
between iconographical learning and naive directness 
in the contemplation of a picture was not long ago pro- 
vided by an article by Dr. Heppner. 1 Fault has often 
been found with a picture by Rembrandt in the Berlin 
Gallery which, as was thought and may be read every- 
where, represents Moses, who, in his anger, breaks 
the tables of the Law. The impression produced by the 
movement and the face of the hero did not fulfil the 
expectations raised by the subject. Dr. Heppner has 
now proved that Rembrandt in reality was faced with 
quite a different subject the law-giver who shows the 
tables to the people. We now no longer miss the ex- 
plosion of anger, the powerful action; on the con- 
trary, from our better knowledge, we can appreciate 
the dignified exaltation of the figure. We learn, too, 
that the picture is but a fragment; we can supply from 
our imagination that which is missing, and our im- 
pression and judgment change. 

1 See Heppner in Oud Holland, 1934:, p. 241 . 



277 



ON ART LITERATURE 

A disadvantage of the terminology of art history 
consists in this, that aesthetic concepts are not suffi- 
ciently clearly differentiated from such as relate to 
time. In art literature of the i9th century, certain art 
forms of the ijth century were censured by applying 
to them the expression 'Baroque', and this term was 
retained for those forms at a time when they had be- 
come admired and appreciated. At present the word 
'Baroque' is used, now in a sense aesthetically dero- 
gatory, now in order to convey the neutral notion of 
date, and confusion follows. Similar remarks apply to 
the concept 'Romantic'. One should never use these 
terms without defining them clearly. 

To every description which aims at conveying a 
sense of character, there attaches something of the 
caricature, since the web, which means the totality of 
qualities, is destroyed if you pick out a thread in order 
to show it for the purpose of emphasis. You need not, 
however, fight shy of such distortion, as long as you 
remain conscious of the one-sidedness, and ready to 
neutralize it by other utterances. The contradictions 
which ensue, and which belong to the nature of all 
psychological knowledge, are not to be feared. Just 
as man is full of contradictions, so is everything that 
he does or creates. 

Descriptions or statements, elaborate and aiming 
at completeness, demand too much of the visual mem- 
ory of the reader; it is the aphorisms, throwing light 
like flashes, which are above all effective. The last sen- 



278 



A CONTENTION 'PRO DOMO' 
tence contains perhaps less of a truth that is univer- 
sally valid than a contention pro domo, by means of 
which reasons are given for a personal peculiarity or, 
maybe, an apology is offered for a personal weakness. 



279 



INDEX 



Aertsen, Pieter, 63, 109, 132 

Albert Cornelisz, 25-1, 252 

Aldenhoven, Carl, 14 

Alfonso, King of Naples, 133 

Alma Tadema, 102 

Altdorfer, 57, 121, 123, 210 

Anonymous Masters, 2 1 3-5 

Antwerp Mannerists, 256 

Appearance, 32-8 

Art and Erudition, 1435-4 

Art and Symbol, 3942 

Art Literature, 2739 

Asia, Painting in, 67 

Authorship, Determination of, 160 

Bastianini, 260, 261 

Bayersdorfer, Adolph, 13, 14, 233 

Beauty, 8790 

Bellini, Giovanni, 164, 252 

Berenson, Mr., 169, 170 

Bocklin, 141, 158 

Bode, Wilhelm von, 10, 13, 14, 176 

Bol, Ferdinand, 1 8 1 , 182 

Bosch, Jerome, 119, 248 

Both, Jan, 15-8 

Botticelli, 1^6, 195 

Bouts, Dirk, 49, ro, 93, 119, 123, 

1^6, 248 
Bril, Paul, 228 

Bruegel, 71, 74, 118, 122, 226 
Bruges, Master of 1480, 214 
Bruyn, Barthel, 2^6 
Buecklaer, Joachim, 25-7 

Cennini, 114 
Cezanne, 26, 27, 68 
Chardin, 46, 107, 231 
Charles I, 165- 



Claude, 122, 228 
Clemenceau, 28 
Colour, 43 2 
Composition, 91-6 
Connoisseurship, Problems of, 179- 

183 

Contemplation, Pleasurable, 1 9-3 1 
Cornells van Haarlem, 124 
Corot, 63 
Correggio, 88, 149 
Cranach, Hans, 2,5-6 
Cranach, Lucas, sen., 57, 120, 121, 

2102, 25-46 

Craquelure, 193, 194, 263, 264 
Criteria of Authorship, Objective, 

163-71 

Crivelli, Carlo, 51 
Cuyp, Aelbert, 46, 122, 220 

Daubigny, 15-6 

Deductions a posteriori from copies 

regarding lost originals, 246-9 
Delacroix, 237 
Denuce", J., 2^6 
Determination of Authorship, Value 

of, 1602 
Diderot, 106, 107 
Distant View and Near View, 58-63 
Dolci, Carlo, 158, 15-9 
Donatello, 260, 261 
Dossena, 260 

Dou, Gerard, 5^9, 159, 178, 236 
Drawings, Study of, 2 1 8-2 1 
Diirer, Albrecht, lor, 119, 129, 

130, 134, 136, 139, 148, 186, 

203, 205-, 219, 223-8, 253 
Dyck, Van, 139, 148, 156, 168, 

203, 223, 228, 2^4 



28l 



INDEX 

Edward, Prince, 126 Graff, Anton, 194 

Eigenberger, Rudolf, 276 Greco, o, 5-7, 76, 141, 1^8, 20.5-, 

Elsheimer, 228 229 

Emotion, 30 Greek Art, 33, 42, 88, 92, 105, 135- 

Erudition, see Art and Erudition Grillparzer, i r, 142, 145 

Examination, Analytical, of Pictures, Griinewald, 25-, 26, -7, 121, 140, 

184-96 169, 211, 235-, 244 

Existence, 32-8 

Eyck, Hubert van, i 4 Hals, Frans, 142, 176, 206, 208, 
Eyck, Jan van, 37, 49, i, 61, 66, 209, 219, 229, 266 

116, 117, 119, 123, 127, 132, Heller, 25-3 

133, 139, iJ4, I 57> I i" 8 > 1 8 6-8, Henry Vm, 126 

244, 248, 262 Heppner, Dr., 277 

Heyden, Jan van der, 6 2 

Facius, 116 History in Painting, 1003 

Fechner, G. T. , 2 3 3 Hobbema, 1 2 2 , 1 5$ 

Feeling, 30 Hofmannsthal, Hugo von, 140 

Floerke, Hanns, 92, no Holbein, 126, 127, 203, 232-4, 
Floris, Frans, 228, 2^2, 25-3, 2f 2,5-9, 262, 266 

Forgeries, 2 ^8-66 Hooch, Pieter de, 5-0, 1 1 r , 2 3 2 

Form, 43, 2 Houbraken, 2^7 

Fraenger, Wilhelm, 121, 27^ Huber, Wolf, 120, 121 
Frame, Function of, 946 

Frueauf, Rueland, the Younger, 120 Impression, First, 172-178 

Fry, Roger, n Impressionists, 28, 29, 77, 94 

Fyt, Jan, 132 Individuality and Type, 84-6 

Influence, 222-9 

Geertgen tot Sint Jans, 119 Interest in Things, Objective, 3 2-8 

Genius and Talent, 13442 Intuition, 172-8 
Genre Painting, 108-12 

Gerard, David, 115-, 119 Jacob Cornelisz., 139, 2ji 

Ghirlandaio, 133 Justi, Carl, 1^3, 27^ 

Giorgione, 116, 119, 197, 204- Justus of Ghent, 129 
Giotto, 139 

Goes, Hugo van der, 49, ^o, 149, Kalf, ^o, 132 

169, 206, 207, 244, 248, 249, Keller, Gottfried, 33, 82 

274 Kessel, Jan van, 2 64 

Goethe, 24, 42, 44, 76, 81, 8^, 93, Kneller, 257 

100, 1 06, in, 136, 144, 1 60 Kobell, Jean, 15-8 

Gogh, Van, 141 Koekkoek, 15:8 
Gold, 43-^2 

Gossaert, Jan (Mabuse), 49, i6j, Landscape, 113-23 

228 Lawrie, Dr. A. P., 188, 189 

Goya, 229 Lenbach, 1 3 6 

Goyen, Jan van, 62 Lesser Masters, 217 



282 



INDEX 



Leibl, 6 1 

Leonardo da Vinci, 90, 136, 15-8, 

219, 241, 242 
Liebermann, Max, 22 
Light, 43-^2 
Lochner, Stephan, 49 
Lucas van Leyden, 139 

Aiaitredesdemi-figuresy 25$ 

Makart, 136 

Manet, 37, 98, 99, 15-7, 1^8, 195- 

Mantegna, 225-, 226 

Mander, Karel van, 124, 163, 2i, 

*$*, 25-3, 2 7 
Margaret of Austria, 165- 
Master of the Death of the Virgin, 

(Joos van Cleve), 215-, 2 55 
Master of Frankfort, 25-5 
Master of the Hausbuch, 2 1 3 
Massys, Quentin, 90 
Meder, Josef, 221 
Medici, The, 132 
Medieval Art, 5-7, 65, 67, 109, 114, 

nr, 127, 227 
Medium Artists, 2 1 7 
Meit, Konrad, 80 
Mending, nj, 128, 132 
Menzel, 103, 156 
Michelangelo, 76, 77, 140, 203 
Michiel, Marcanton, 165 
Millet, Jean-Francois, 108 
Momper, Jodocus, 63 
Monet, 28 
Montaigne, 178 
Morelli, 164, 1669 
Moro, Antonio, 127, 257 
Moroni, Giambattista, 127 
Movement, 69-74 
Miindler, Otto, 166 
Murillo, 78 

Neuburg, William Duke of, 25-3 
Neufchatel, Nicholas, 127 
Nietzsche, 21, 167 
Nude, The, 104-7 



Original and Copy, 230 4^ 
Orley, Bernard van, 165-, 216, 25-4 
Ostade, Adriaenvan, 0, ^9, 85-, 1 1 1 
Ostwald, 276 
Ouwater, 93 

Pahna Vecchio, 194 
Patanir, Joachim, 1 1 9 
Perceiving, 19-31 
Perreal, Jean, 248 
Personality, 20012 
Perspective, Atmospheric, 62, 63 
Perspective, Linear, 648 
Petrus Christus, 132, 133 
Pettenkofer, Herr, 192 
Philip of Burgundy, i r 6 
Philip n, 1 6^ 

Photography, Use of, 1979 
* Pictorial', Concept of, ^4-7 
Picture Categories, 979 
Piero della Francesca, o 
Piloty, 102 
Portraiture, 12430 
Potter, 3, ^9, 72 
Pourbus, Pieter, 194 
Poussin, 228 

Quality, Artistic, 230-45- 

Raphael, 88, 89, 92, 147, ijo, 153, 
!6 7 , 203, 209, 219, 226, 227 

Rembrandt, 38, 47, 60, 81, 140, 
141, 1^2, 1^9, 178, 179, 181, 
182, 189, 203, 205-, 208, 209, 

216, 217, 219-21, 22, 229, 23J, 
2 4 I, 242,278 

Restorations, 267-72 

Reymerswaele, Marinus van, 133 

Reynolds, 225- 

Ribera, 81 

Ring, Crete, 17 

Rothschild, Baron Robert de, 187 

Rubens, 45, 72, 101, 122, 136, 139, 

148, 189, 220, 22^, 227-9, 237, 
2J2-4 

Ruisdael, Jacob, 46, 122, 158, 264 



283 



INDEX 



Saenredam, 3 5-, 47 

Sarto, Andrea del, 1^8 

Scheibler, Ludwig, 13, 14, 196 

Schiller, 14$ 

Schinkel, 96 

Schongauer, Martin, 148, 213, 22j 

228 

Schopenhauer, 23 
Schwind, Moritz von, 1 2 1 
Scorel, Jan van, 223, 224, 228, 25-1 
Seeing, 1931 
Seghers, Hercules, 264- 
Self-Portrait, The, 127, 128 
Simone Martini, 139 
Size and Scale, 5-8-63 
Slevogt, 71 
Snyders, jo, 132 

Spectator, Standpoint of the, 155 9 
Steen, Jan, 36, 85-, in 
Still Life, 131-3 
Strigel, Bernhard, 127 
Style, 7^-8 3 

Talent, see Genius and Talent. 
Teniers, o 

Terborch, Gerard, 35-, 59, 102, 1 1 1, 
112, 236, 237 



Thiem, Adolf, 1 5-6 
Tintoretto, 5-9, 6 1 
Titian, 38, 50, 127, 178, 182, 20$, 

208, 209, 225-, 229, 237 
Tonality, 43 5-2 
Toulouse-Lautrec, 71-4 
Truth to Nature, 7^-83 
Type, see Individuality and Type. 

Value, Artistic, 7^-83 
Vasari, 163, 213 
Velazquez, 219, 229, 27-5- 
Velde, Willem van de, 3 
Vermeer, o, in, 265-, 266 

Watteau, 46, 88, 139, 227 
Weale,W. H. J., 2^2 
Weinberger, Martin, 120 
Weyden, Roger van der, ii, 117, 

1^7, 214, 220, 248 
Wilhelm, Master, 135- 
William U, King of Holland, 1^8 
Witte, Emanuelde, 29, 47 
Wolfflin, Heinrich, 224, 275- 
Workshop Production, 2 0-7 

Ysenbrant, Adriaen, 25-6 



284