Art and the Brain

Author(s): Semir Zeki

Source: Daedalus, Vol. 127, No. 2, The Brain (Spring, 1998), pp. 71-103
Published by: The MIT Press on behalf of American Academy of Arts & Sciences
Stable URL: http://www.jstor.org/stable/20027491
Accessed: 09-08-2016 18:18 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

The MIT Press, American Academy of Arts & Sciences are collaborating with JSTOR to digitize,
preserve and extend access to Daedalus

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Semir Zeki

Art and the Brain

Les causeries sur l'art sont presque inutiles.1

?Paul C?zanne

More often than not, [people] expect a painting to

speak to them in terms other than visual, preferably
in words, whereas when a painting or a sculpture
needs to be supplemented and explained by words
it means either that it has not fulfilled its function
or that the public is deprived of vision.2
?Naum Gabo

I.

Much has been written about art, but not in relation to

the visual brain?through which all art, whether in
conception or in execution or in appreciation, is ex
pressed. A great deal, though perhaps not as much, has been
written about the visual brain, but little in relation to one of its
major products, art. It is therefore hardly surprising that the
connection between the functions of art and the functions of the
visual brain has not been made. The reason for this omission lies
in a conception of vision and the visual process that was largely
dictated by simple but powerful facts, derived from anatomy
and pathology. These facts spoke in favor of one conclusion to

Semir Zeki is Professor of Neurobiology and Co-head of the Wellcome Department of

Cognitive Neurology at University College London.

71

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
72 Semir Zeki
which neurologists were ineluctably driven, and that conclusion
inhibited them, as well as art historians and critics, from asking
the single most important question about vision that one can
ask: Why do we see at all? It is the answer to that question that
immediately reveals a parallel between the functions of art and
the functions of the brain, and indeed ineluctably drives us to
another conclusion?that the overall function of art is an exten
sion of the function of the brain. In that definition are the germs
of a theory of art that has solid biological foundations and that
unites the views of modern neurobiologists with those of Plato,
Michaelangelo, Mondrian, C?zanne, Matisse, and many other
artists.
The concept of the functions of the visual brain inherited by
modern neurobiologists was based on facts derived between
1860 and 1970. Chief among these was the demonstration by
the Swedish neuropathologist Salomon Henschen and his succes
sors in Japan and England that the retina of the eye is not
diffusely connected to the whole brain, or even to half the brain,
but only to a well-defined and circumscribed part of the cerebral
cortex. First called the visuo-sensory cortex and later the pri
mary visual cortex (area VI), it therefore constituted "the only
entering place of the visual radiation into the organ of psyche."3
This capital discovery led to a prolonged battle between its
proponents and its opponents, who thought of it as "une
localisation ? outrance";4 they had conceived of the visual input
to the brain as being much more extensive and as including large
parts of the cerebral cortex that were known to have other
functions, a notion more in keeping with the doctrine of the
French physiologist Pierre Flourens. The predecessor of the Ameri
can psychologist Karl Lashley, Flourens had imagined that each
and every part of the cortex is involved in every one of its
activities. It was not until early this century that the issue of a
single visual area located in an anatomically and histologically
defined part of the cortex was settled in favor of the
localizationists.5 There was much else to promote the idea of VI
as the "sole" visual center. It had a mature appearance at birth,
as if ready to "receive" the visual "impressions formed on the
retina,"6 whereas the cortex surrounding it matured at different
stages after birth, as if the maturation depended upon the acqui

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 73
sition of experience; this made of the latter higher cognitive
centers, the Cogitatzionzentren, whose function was to interpret
the visual image received by VI, or so neurologists imagined. As
well, lesions in VI lead to blindness, the position and extent of
which is in direct proportion to the position and size of the
lesion; by contrast, lesions in the surrounding cortex resulted in
vague visual syndromes, referred to first as mind blindness
(Seelenblindheit) and then as agnosia, following the term intro
duced by Freud. Together, these facts conferred the sovereign
capacity of "seeing" on VI, leading neurologists to conceive of
it as the "cortical retina," the cerebral organ that receives the
visual images "impressed" upon the retina, as on a photographic
plate?an analogy commonly made. Seeing was therefore a pas
sive process, while understanding what was seen was an active
one; this notion divided seeing from understanding and assigned
a separate cortical seat to each.
This concept left little room for the fundamental question of
why we see. Instead, seeing was accepted as a given. Asked the
question today, few would suppose that it is so we can appreci
ate works of art; most would give answers that are specific,
though related in general to the survival of the species. The most
general of these answers would include all the specific ones and
define the function of seeing as the acquisition of knowledge
about the world.7 There are, of course, other ways of obtaining
that knowledge?through the sense of touch or smell or audi
tion, for example?but vision happens to be the most efficient
way of obtaining it, and there are some kinds of knowledge,
such as the color of a surface or the expression on a face, that
can only be obtained through vision.
It takes but a moment's thought to realize that obtaining that
knowledge is no easy matter. The brain is only interested in
obtaining knowledge about those permanent, essential, or char
acteristic properties of objects and surfaces that allow it to
categorize them. But the information reaching the brain from
these surfaces and objects is in continual flux. A face may be
categorized as a sad one, thus giving the brain knowledge about
a person in spite of the continual changes in individual features,
viewing angle, or even the identity of the face viewed; or the
destination of an object may have to be decided by its direction

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
74 Semir Zeki
of motion, regardless of its speed or distance. An object may
have to be categorized according to color, as when judging the
state of ripeness of an edible fruit. But the wavelength composi
tion of the light reflected from an object is never constant; it
changes continually, depending upon the time of day, without
entailing a substantial shift in its color. The ability of the brain
to assign a constant color to a surface or a constant form to an
object is generally referred to as color or object constancy. But
perceptual constancy is a much wider phenomenon. It also ap
plies, for example, to faces that are recognizable when viewed
from different angles and regardless of the expression worn.
There is also what I shall call situational constancy, when the
brain is able to categorize an event or a situation as festive or
sad and so on, regardless of the particular event. There is even
a narrative constancy when, for example, the brain is able to
identify a scene as the "Descent from the Cross," regardless of
variations in the detail or the style of the painting. The brain, in
each case, extracts from the continually changing information
that reaches it only what is necessary for it to identify the
characteristic properties of what it views; it has to extract con
stant features in order to be able to obtain knowledge about
them and to categorize them. Vision, in brief, is an active pro
cess that depends as much upon the operations of the brain as
upon the external, physical environment; the brain must dis
count much of the information reaching it, select only what is
necessary in order to obtain knowledge about the visual world,
and compare the selected information with its stored record of
all that it has seen. A modern neurobiologist should approve
heartily of Matisse's statement that "Voire, c'est d?j? une opera
tion cr?atrice, qui exige un effort."8
How the brain achieves this remarkable feat remains a puzzle;
indeed, the question has only been seriously addressed in the last
thirty years, which have witnessed a prolific output of work on
the visual brain. Among the chief discoveries is that it is com
posed of many different visual areas that surround VI.9 Each
group of areas is specialized to process a particular attribute of
the visual environment by virtue of the specialized signals that
each receives from VI.10 Cells specialized for a given attribute
such as motion or color are grouped together in anatomically

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 75
identifiable compartments within VI, and different compart
ments connect with different visual areas outside VI, thus con
ferring their specializations on the relevant areas.11 VI acts much
like a post office, distributing different signals to different desti
nations; it is but the first, though essential, stage in an elaborate
machinery designed to extract the essential information from the
visual world. What we now call the visual brain is actually VI
plus the specialized visual areas with which it connects, directly
and indirectly. We therefore speak of parallel systems devoted to
processing simultaneously different attributes of the visual world,
a system comprising the specialized cells in VI plus the special
ized areas to which these cells project. Vision, in brief, is modu
lar. The reasons for evolving a strategy to process in parallel the
different attributes of the visual world have been debated, but it
seems plausible to suppose that they are rooted in the need to
discount different kinds of information when acquiring knowl
edge about different attributes.12 With color, it is the precise
wavelength composition of the light reflected from a surface that
has to be discounted, whereas with size it is the precise viewing
distance, and with form, the viewing angle.
Recent evidence has shown that the processing systems are
also perceptual systems in that activity in each can result in a
percept without reference to the other systems; each processing
perceptual system terminates its perceptual task and reaches its
perceptual endpoint at a slightly different time than the others,
thus leading to a perceptual asynchrony in vision?color is seen
before form, which is seen before motion, with the advantage of
color over motion being on the order of 60-100 ms.13 Thus
visual perception is also modular. In summary, the visual brain
is characterized by a set of parallel processing-perceptual sys
tems and a temporal hierarchy in visual perception.14
These findings lead me to propose that there is also a modu
larity, a functional specialization, in visual aesthetics. When
area V4, the color center, is damaged, the consequence is an
inability to see the world in color.15 But other attributes of the
visual scene are perceived normally. When area V5, the motion
center, is damaged, the consequence is an inability to see objects
when in motion, although other attributes are seen normally.
Damage to a region close to V4 leads to a syndrome character

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 76 Semir Zeki
ized by an inability to see familiar faces. There are other specific
syndromes?for example, the inability to recognize certain cat
egories of objects?and neurology is continually uncovering new
syndromes of selective visual loss. I do not mean, of course, to
imply that the aesthetics of color are due solely to the activity in
V4 or the aesthetics of kinetic art are due solely to the activity in
V5; I am suggesting only that the perception of color and of
motion is not possible without the presence and healthy func
tioning of these areas. It does little good to ask a patient with a
V4 lesion to appreciate the complexities of fauvist art or a
patient with a V5 lesion to view the works of Tinguely. These
are aesthetic experiences of which such patients are not capable.

II.

The definition that I have given above of the function of the

visual brain?a search for constancies with the aim of obtaining
knowledge about the world?is applicable with equal vigor to
the function of art. I shall thus define the general function of art
as a search for the constant, lasting, essential, and enduring
features of objects, surfaces, faces, situations, and so on, which
allows us not only to acquire knowledge about the particular
object, or face, or condition represented on the canvas but to
generalize, based on that, about many other objects and thus
acquire knowledge about a wide category of objects or faces. In
this process, the artist must also be selective and invest his work
with attributes that are essential, discarding much that is super
fluous. It follows that one of the functions of art is an extension
of the major function of the visual brain. Indeed, philosophers
and artists often spoke about art in terms that are extremely
similar to the language that a modern neurobiologist of vision
would use, except that he would substitute the word "brain" for
the word "artist." It is striking, for example, to compare Herman
von Helmholtz's statement about "discounting the illuminant,"
in which a colored surface is viewed (in order to assign a con
stant color to a surface), with the statement of Albert Gleizes
and Jean Metzinger in their book Cubism. Discussing Gustave
Courbet, they wrote, "Unaware of the fact that in order to
display a true relation we must be ready to sacrifice a thousand

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 11
apparent truths, he accepted, without the slightest intellectual
control, all that his retina presented to him. He did not suspect
that the visible world can become the real world only by the
operation of the intellect."16 I interpret "intellect" to mean the
brain or, better still, the cerebral cortex. In order to represent
the real world, the brain (or the artist) must discount ("sacri
fice") a great deal of the information reaching it (or him),
information that is not essential to its (or his) aim of represent
ing the true character of objects.
It is for this reason that I hold the somewhat unusual view
that artists are neurologists, studying the brain with techniques
that are unique to them and reaching interesting but unspecified
conclusions about the organization of the brain. Or, rather, that
they are exploiting the characteristics of the parallel processing
perceptual systems of the brain to create their works, sometimes
even restricting themselves largely or wholly to one system, as in
kinetic art. These conclusions are on canvas and are communi
cated and understood through the visual medium, without the
necessity of using words. This may surprise them since most of
them, naturally enough, know nothing about the brain, and a
good many still hold the common but erroneous belief that one
sees with the eye rather than with the cerebral cortex. Their
language, as well as the language of those who write about art,
betrays this view. But however erroneous their views about the
seeing organ or the role of the visual brain may be, it is sufficient
to glance at their writings to realize the extent to which they
have defined the function of art in a way that a modern neuro
biologist would not only understand but feel very sympathetic
to. Thus Henri Matisse once said, "Underlying this succession of
moments which constitutes the superficial existence of things
and beings, and which is continually modifying and transform
ing them, one can search for a truer, more essential character,
which the artist will seize so that he may give to reality a more
lasting interpretation."17 Essentially, this is what the brain does
continually?seizing from the constantly changing information
reaching it the more essential one, distilling from the successive
views the essential character of objects and situations. Similar
statements abound, and it is sufficient to give just one more
example. Jacques Rivi?re, the art critic, wrote: "The true pur

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
78 Semir Zeki
pose of painting is to represent objects as they really are, that is
to say, differently from the way we see them. It tends always to
give us their sensible essence, their presence, this is why the
image it forms does not resemble their appearance ..." because
the appearance changes from moment to moment.18 A neurolo
gist could hardly have improved on that statement in describing
the functions of the visual brain. He might say that the function
of the brain is to represent objects as they really are, that is to
say, differently from the way we see them from moment to
moment if we were to take into account solely the effect that
they produce on the retina.
To summarize, therefore, both the brain and one of its prod
ucts, art, have the task of, in the words of artists themselves,
depicting objects as they are. And both face the problem of how
to distill from the ever changing information in the visual world
only that which is important in order to represent the perma
nent, essential characteristics of objects. Indeed, this was almost
the basis of Kant's philosophy of aesthetics, to represent perfec
tion; but perfection implies immutability, and hence arises the
problem of depicting perfection in an ever changing world. I
shall therefore define the function of art as being a search for
constancies, which is also one of the most fundamental functions
of the brain. The function of art is therefore an extension of the
function of the brain?the seeking of knowledge in an ever
changing world.

ill.

Plato was among the most prominent of those who lamented the
poverty of art. Without saying so, and indeed without ever
referring to the brain, he implicitly compared the limitations of
art to the infinite capacities of the brain. His most explicit
statement in this regard occurs in Book X of The Republic,
where he dismisses painting as a mimetic art, one that could
represent only one aspect of a particular example of a more
general category of an object. To him there was the general ideal
of a given form, which was the embodiment of all the examples
of that form; then there was a particular form that was but one
example of the more general, "universal" form; and finally,

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 79
there was painting, which captured but one facet, one image, of
one particular form. "The Greeks," Sir Herbert Read tells us,19
"with more reason, regarded the ideal as the real, and represen
tational art as merely an imitation of an imitation of the real."20
Plato's contempt for painting was really linked to his theory
of forms and ideals. The example he gives in Book X is that of
a couch. To him, there is only one real couch, the one created by
God; this is the idea of a couch, and it has a universal existence.
One can therefore obtain real knowledge only about this one
ideal couch. Of particular couches?as made by a craftsman
(?T||LLioupYoQ, or represented in a single view in a painting, or
reflected in a mirror?there can only be an opinion, and an
unverifiable one at that.21 Put in mathematical terms, we can
only obtain real and reliable knowledge about ideal circles,
triangles, and straight lines. Viewing painted circles and straight
lines without reference to the Ideal leads only to a superficial
impression and an opinion, which may turn out to be true or
false. Plato implied that, at least to get nearer to the Ideal,
painting should change direction in order to represent as many
facets of an object or situation as possible, since this would give
more knowledge about the object. What Plato only implied,
Schopenhauer made explicit many centuries later when he wrote
that painting should strive "to obtain knowledge of an object,
not as a particular thing but as Platonic Ideal, that is, the
enduring form of this whole species of things," a statement that
a modern neurobiologist could easily accommodate in describ
ing the functions of the visual brain.22 Indeed, to a neurobiolo
gist, a brain that is not able to do this is a sick, pathological
brain. Painting, in other words, should be the representation of
the constant elements, the essentials, that would give knowledge
of all couches; it should represent constancies. As John Con
stable put it in his Discourses: "The whole beauty and grandeur
of Art consists ... in being able to get above all singular forms,
local customs, particularities of every kind ... [The painter] makes
out an abstract idea of their forms more perfect than any one
original," the "abstract idea" presumably being Constable's term
for the Platonic Ideal.23
There is something unsatisfactory about the Platonic Ideal
from a neurobiological point of view, because the Ideal has an

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 80 Semir Zeki
existence that is external to the brain and without reference to it;
we can only have an opinion of that which we perceive, "whereas
knowledge is of a super-sensible eternal world."24 Implicitly
more dependent upon brain function, and thus more acceptable
neurobiologically, are the views of Kant and Hegel. Their views
exalt art, which is seen as being able to represent reality better
than the "ephemera of sense data," since the latter changes from
moment to moment. Hegel deals with the Idea that is derived
from the Concept. In a painting, the brain, having "accumulated
a treasure," can "now freely disgorge [it] in a simple manner
without the far-flung conditions and arrangements of the real
world." By this process of "disgorging," and thus of externaliz
ing and concretizing, the Concept becomes the Idea. The Idea,
then, is merely the external representation of the Concept that is
in the brain, the Concept that it has derived from ephemeral
sense data. It is, in fact, the product of the artist. Art, including
painting, therefore, "furnishes us with the things themselves, but
out of the inner life of the mind"; through art, "instead of all the
dimensions requisite for appearance in nature, we have just a
surface, and yet we get the same impression that reality af
fords."25 It is through this translation of the Concept into the
Idea that Dutch painting, for example, "has recreated . . . the
existent and fleeting appearance of nature as something gener
ated afresh by man."26
Although the views of Plato and Hegel may appear antipodean,
the difference between the two is in fact neurobiologically irrel
evant if we try to give a neurobiological definition of the Pla
tonic Ideal and the Hegelian Concept. The first step in such a
definition, relevant to Plato's views but less so to Hegel's, is a
neurobiological doctrine: forms do not have an existence with
out a brain. This may seem an audacious statement to make, but
it is supported by numerous clinical and physiological studies
that have shown that individuals who are born blind and to
whom vision is later restored find it very difficult, if not impos
sible, to learn to see even a few forms, and these they soon
forget. The question that the learned Mr. Molyneux asked in
John Locke's Essay Concerning Human Understanding?whether
a man born blind and who had learned to distinguish between
forms by touch alone would be able to distinguish them by sight

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 81
alone when vision is restored to him?has been answered nega
tively many times by clinical studies.27 Physiological studies,
particularly those of David Hubel and Torsten Wiesel, have
shown that even when the genetically determined visual appara
tus is intact at birth, the organism must be exposed to visual
stimuli after birth, after which visual education becomes much
less important.28 There is, in other words, a critical period for
vision, just as there appears to be for emotional development.29
Artists have often wished that they could see and paint the world
as a child does?for the first time, innocently, without what they
suppose to be the prejudice of the developed and possibly even
corrupted influence of a brain that has knowledge of the world.
Picasso admired the art of children and Matisse wanted to paint
like them, as does Balthus;30 Monet wished that he could have
been born blind, with vision restored to him later in life so that
he could see pure form "without knowing what the objects were
that he saw before him."31 They are all yearning for something
that is physiologically almost impossible. The visual apprentice
ship of children occurs at a very early age, before two, and
begins immediately after birth, long before the motor apparatus
has developed sufficiently to be able to execute a painting. In its
conceptual immaturity and technical simplicity, the art of a four
year-old child may be touching and even exciting, but it is the art
of a visual brain that is already highly developed and that has
acquired much knowledge about the world. The innocence that
artists yearn for is, in terms of the brain, a myth.
If neurologically no forms, ideal or otherwise, exist without a
brain that is properly nourished, how can we define the Platonic
Ideal and the Hegelian Concept in neurological terms? I would
propose that both can be equated with the brain's stored memory
record of all the views of all the objects that it has seen, from
which it has formed a Concept or an Ideal of these objects such
that a single view of an object makes it possible for the brain to
categorize that object. Indeed, in Plato's system, we can only
recognize and categorize objects of which our immortal souls
have seen examples constructed by ?r||LiioupYO (see, for example,
Plato's Meno). In this sense, therefore, the Platonic system ac
knowledges the importance of a stored record, though without
making reference to the brain. The recognition that we can only

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 82 Semir Zeki
categorize objects that we have already seen (and therefore have
a general representation of) nevertheless constitutes a far-reach
ing insight and brings Plato's position close to a modern neuro
biological one. Neurobiology would have to depart from the
Platonic system in saying not only that this general representa
tion is built by the brain but also that there can be no Ideals
without the brain.
We know a little, but not much, about the brain's stored
visual memory system for objects. We know that it must involve
a region of the brain known as the inferior convolution of the
temporal lobes, because damage here causes severe problems in
object recognition. Although very much in their infancy, recent
physiological studies have started to give us some insights into
the more detailed physiological mechanisms involved.32 When a
monkey, an animal that is close to man, is exposed to different
views of objects that it has never encountered before (objects
generated on a television screen), recording from single cells in
its inferior temporal cortex can show how they respond when
these same objects are subsequently shown on the screen again.
Most cells respond to one view only, and their response declines
as the object is rotated in such a way as to present increasingly
less-familiar views. A minority of cells respond to only two
views, but only a very small proportion, amounting to less than
1 percent, respond in a view-invariant manner. Whether they
respond to one or more views, the actual size of the stimuli or
the precise position in the field of view in which they appear
make little difference to the responses of the cell. On the other
hand, no cells have ever been found that are responsive to views
with which the animal has not been familiarized; hence, expo
sure to the stimulus is necessary, from which it follows that the
cells may be plastic enough to be "tuned" to one or more views
of an object. In summary, many cells, each one responsive to one
view only, may be involved during recognition of an object, with
the whole group acting as an ensemble. But the presence of that
small 1 percent of cells that respond in a view-invariant manner
suggests also that form constancy may be the function of a
specialized groups of cells, since 1 percent represents an enor
mous number in absolute terms.

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 83
When undertaking their work, artists generally are concerned
not with philosophical views but rather with achieving desired
effects on canvas?by experimenting, by "sacrificing a thousand
apparent truths" and distilling the essence of their visual experi
ence. We are told, for example, that Cezanne's work is "a
painted epistemology" (Erkenntnis Kritik), since C?zanne sup
posedly shared Kant's ideology.33 But C?zanne, in particular, put
paid to all these empty speculations even before they were made
when he said that "les causeries sur l'art sont presque inutiles."34
I agree with Kahnweiler when he states, "J'insiste, en passant,
sur le fait qu'aucun de ces peintres . . . n'avait de culture
philosophique, et que les rapprochments possibles?avec Locke
et Kant surtout?d'une telle attitude leur ?taient inconnus, leur
classement ?tant, d'ailleurs, instinctif plus que raisonn?. "3S The
preoccupation of artists has instead been less exalted and more
similar to the physiological experiments described earlier: expos
ing themselves to as many views of their subject as possible and
thus obtaining a brain record from which they can distill on
canvas the best combination. If, in executing his work, the artist
is indifferent to these polar views?Plato on the one hand, and
Hegel and Kant on the other?so should the neurobiologist be,
if he accepts my equation of the Platonic Ideal and the Hegelian
Concept with the brain's stored record of what it has seen.
Whether art succeeds in presenting the real truth, the essentials,
or whether it is the only means of getting to that truth in the face
of constantly changing and ephemeral sense data, the opposing
views are at least united in suggesting that there is (Hegel) or
that there should be (Plato and Schopenhauer) a strong relation
ship between painting and the search for essentials. And my
equation of both the Hegelian Concept and the Platonic Ideal
with the brain's stored record means that the difference between
the two, from a neurological point of view, is insignificant.
There have been artists who have, again without reference to the
brain or its stored record, tried deliberately and with much
success to contradict the stored memory record of the brain.
Many of the works of Ren? Magritte go against everything that
the brain has seen, learned, and stored in its memory. There is
no Platonic Ideal or Hegelian Concept here because the brain has
no representation of such bizarre scenes. It is an act of the

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 84 Semir Zeki
imagination that fascinates the brain, which tries to make sense
of a scene that goes against all its experience and for which it
can find no solution.

IV.

To a neurobiologist viewing the art scene without being involved

in it, it seems to be cubism that most explicitly, without ac
knowledging it or perhaps being even aware of it, set out to
answer that deep paradox between reality and appearance al
luded to by Plato?although this is my interpretation, not that of
cubists. Cubism, the most radical departure in Western art since
Paolo Uccello and Piero della Francesca introduced perspective
into painting, "was a sort of analysis," a static representation of
the result of "moving around an object to seize several successive
appearances, which, fused in a single image, reconstitute it in
time."36 The aims of cubist painting, which was an attempt "to
discover less unstable elements in the objects to be represented,"
were well stated by the French critic Jacques Rivi?re, and they
read as if they were an account of the aims of the brain.37 Rivi?re
wrote that "The cubists are destined ... to give back to painting
its true aims, which is to reproduce . . . objects as they are." But
to achieve this, "lighting must be eliminated" because "it is the
sign of a particular instant. ... If, therefore, the plastic image is
to reveal the essence and permanence of things, it must be free of
lighting effects. ... It can therefore be said that lighting prevents
things from appearing as they are. . . . Sight is a successive sense;
we have to combine many of its perceptions before we can know
a single object well. But the painted image is fixed. ..." Perspec
tive must also be eliminated because it "is as accidental a thing
as lighting. It is the sign ... of a particular position in space. It
indicates not the situation of objects but the situation of a
spectator. . . . Perspective is also the sign of an instant, of the
instant when a certain man is at a certain point."38 That state
ment is one that a modern neurobiologist would applaud, for the
brain likewise never sees the objects and surfaces that make up
the visual world around us from a single point or in a standard
lighting condition. Instead, objects are viewed at different dis

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 85
tances, from different angles and in different lighting conditions,
yet they maintain their identity.
The solution that cubism brought to this problem was to try
to mimic the functions of the brain, though with far less success.
The precursor of cubist painting is generally agreed to be Picasso's
Les Demoiselles d'Avignon, a forceful painting about which a
great deal has been written, much of it neurologically and visu
ally uninformative. What is especially interesting visually is the
ambiguity in the figures, especially the one seated to the bottom
right; she could be facing us or facing sideways. This ambiguity
was much exploited by both Picasso and the cofounder of cub
ism, Georges Braque. The elimination of the point of view be
came a prominent feature of many of Picasso's portraits, so that
the subject could seemingly be facing one direction or another.
In later representative paintings such as The Violin Player, Picasso
introduced so many different points of view that the painting
ceased to be recognizable to the human brain, the final result
being only recognizable as a violin player through its title. A
brain ignorant of that title can hardly construe this as a violin
player. The brain regularly views objects and people from differ
ent angles and is able to integrate these different views in an
orderly way, allowing it to recognize and obtain knowledge
about what it is viewing. The attempt by cubism to mimic what
the brain does, to create a perceptual constancy for objects
regardless of viewing angle, was, in the neurobiological sense, a
failure?a heroic failure perhaps, but a failure nevertheless. My
neurobiological interpretation is that it is indeed because of this
failure that cubism changed course and entered its later, syn
thetic phase; it is certain that Mondrian saw it that way, for he
abandoned cubism and accused it of "not accepting] the logical
consequences of its own discoveries [and] developing abstraction
towards its ultimate goal, the expression of pure reality/'39 In
the synthetic phase, Kasimir Malevich tells us, "objective nature
is merely the starting point?the motivation?for the creation of
new forms, so that the objects themselves can scarcely, if at all,
be recognized in the pictures."40 But the new forms that synthetic
cubism created were ultimately derived from the forms in nature
that the artist was exposed to, and perhaps the best proof of this
is found in the objective titles given to the paintings. It is diffi

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 86 Semir Zeki
cult for the brain of a spectator to decipher what many of the
creations of synthetic cubism represent. It was probably also
difficult for Picasso himself, which is presumably one reason
why he used objective and recognizable titles to describe his
paintings. Nilsen Lauvrik, who was hostile to cubism, described
Woman with a Mustard Pot as "one of the most engaging
puzzles of a very puzzling art. This is sharply emphasized by the
delight and pride of every spectator who is successful in solving
the puzzle by finding in these enigmatic charts some sort of a
tangible, pictorial justification of the title appended thereto. ... The
discovery of the 'mustard pot' would scarcely have been possible
without the happy cooperation of the title with the spectator's
previous knowledge of the actual appearance of a mustard pot."41

v.

From the neurobiological point of view, representational art wa

a good deal more successful in meeting the brain's incessan
demands for constancy. Here I will consider neurologically t
work of two different artists, Vermeer and Michelangelo, bo
of whom?unknowingly and in different ways?satisfied thi
demand far better than the product of the heroic but neu
biologically flawed experiments of the cubists.
A great deal has been written about Vermeer, "un artiste
jamais inconnu," as Proust astutely called him.42 His techni
brilliance, his use of perspective and rich chromatism are
common knowledge. But in viewing a painting such as Man a
Woman at the Virginal (displayed at Buckingham Palace), it
not these features that attract and move the ordinary viewe
Paul Claudel, among others, has commented on the banality
Vermeer's subjects: an interior, a maid pouring milk, a g
weighing gold, another reading a letter, a music lesson?all da
events seemingly without special significance.43 But there is, in
Claudel's words, something "eerie, uncanny" about them.44 In
good many of his paintings, the viewer is invited to look inside
as if through a keyhole, but not to enter.45 He is a voyeur
peering into the private moments of private, unknown, individ
als; what they are doing, or saying, or thinking is a mystery. I
is this aspect of Vermeer, I believe, that has the immediate pow

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 87
to attract and provoke, and his technical virtuosity is used in the
service of that psychological power, not as an end in itself,
unsurpassed though it may be.
Where does this psychological power come from and what, in
any case, do we mean by psychological power? A painting like
Man and Woman at the Virginal derives its grandeur, I believe,
from the way in which its technical virtuosity is used to generate
ambiguity. I use the term ambiguity here to mean its ability to
represent simultaneously, on the same canvas, not one but sev
eral truths, each as valid as the others.46 These several truths
revolve around the relationship between the man and the woman.
There is no denying that there is some relationship between
them. But is he her husband, her lover, a suitor, or a friend? Did
he actually enjoy the playing or does he think that she can do
better? Is the harpsichord really being used or is she merely
playing a few notes while concentrating on something else, per
haps something he told her, announcing a separation or a recon
ciliation, or maybe something a good deal more banal? All these
scenarios have equal validity in the painting, which can thus
satisfy several "ideals" simultaneously. Through its stored memory
of similar past events, the brain can recognize in this work the
ideal representation of many situations and can categorize the
scene represented as happy or sad. This gives ambiguity, a char
acteristic of all great art, a different, and neurological, defini
tion: not the vagueness or uncertainty found in dictionaries, but,
on the contrary, certainty?the certainty of many different and
essential conditions, each equal to the others, all expressed in a
single painting, profound in its faithful representation of so
much.
The Vermeer painting satisfies Schopenhauer's wish that a
painting should "obtain knowledge of an object, not as particu
lar thing but as Platonic Ideal, that is to say, the enduring form
of this whole species of thing."47 In any of a number of situa
tions, the scene depicted is what one might actually expect.
There is a constancy about it, which makes it independent of the
precise situation and applicable to many. The painting is indeed
"a vision of two distant people 'alone together' in a space moved
by forces beyond the ken of either," a scenario effectively ex
ploited by Michaelangelo Antonioni in some of his films, and

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 88 Semir Zeki
most notably in VAvventura and VEclisst, where once again the
viewer becomes imaginatively involved in trying to guess the
thoughts of the protagonists.48 Though it may come as a sur
prise, there is in this respect and in terms of the brain a certain
neurobiological similarity between the paintings of Vermeer and
those of the cubists, especially the later variety that cultivated an
ambiguity, in the sense that I have used the term. Writing of
cubism, Gleizes and Metzinger tell us that "certain forms should
remain implicit, so that the mind of the spectator is the chosen
place of their concrete birth."49 There could be no more admi
rable description of the work of Vermeer, where very nearly all
is implicit. As with forms and objects in cubist art, the brain of
the spectator is the chosen birthplace of many situations in
Vermeer's paintings, each one of equal validity as the others.
The true solution remains "? jamais inconnu," because there is
no true solution, there is no correct answer. It is therefore a
painting for many conditions.
Situational constancy is a subject that neurology has not yet
studied; indeed, the problem itself has not been addressed. We
have hardly begun to understand the simpler kinds of constancy,
such as form or color, and it is not surprising that neurologists
should not have even thought of studying so complex a subject.
I would guess from the kind of physiological experiment de
scribed above that, in broad outline, exposure of an individual
to a few situations would be sufficient to extract the elements
that would be common to all similar occasions. But what brain
mechanisms are involved remains a mystery today.
It was perhaps the masters of Cycladic art in the sixth century
B.c. who understood earlier than most that the brain must be the
place of birth of implicit forms. They created works that empha
sized a few organs?the lips or the nose, for example?and left
it to the imagination of the beholder to complete the form.
Michaelangelo achieved much the same effect by leaving many
of his sculptures unfinished. Why he did so remains a question
of debate; my interpretation is that this was one solution to the
problem of representing the many facets of spiritual beauty and
divine love?it was too great a task even for the mighty
Michaelangelo. We know that he usually refused to execute
portraits, believing that he could not represent all the beauty

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 89
that his brain had formed a Concept of. Two exceptions are his
portraits of Andrea Quaratesi and of Tomasso de' Cavalieri, the
young nobleman who had overwhelmed him with his beauty and
had come to dominate his emotional life in his later years,
unleashing a furious creative energy of great brilliance. But the
difficulty of portraying physical beauty was nothing compared
to that of depicting spiritual beauty and divine love. As a devout
Catholic, Michaelangelo found that love in the life of Jesus, and
particularly in the last moments on the Cross and after the
descent from it, the subject of several of his sculptures. By
leaving many of his sculptures unfinished?for example, the
Kondanini Pi?ta and the San Matteo?Michaelangelo invites the
spectator to be imaginatively involved, and the spectator's view
will fit many of the Concepts, the stored representations, in his
brain. The forms that the unfinished work merely suggests be
come concretely realized only in the spectator's brain. There is
an ambiguity here too, and therefore a constancy about these
unfinished works, but the ambiguity is reached by a different
route. Perhaps the most definitive hint at what Michaelangelo
intended is found in his Rime or Sonnets, where, next to his
works, he best expounds his views on art and beauty. In one,
dedicated to Vittoria Colonna, Marchesa di Pescara, he wrote:

The greatest artists have no thought to show that

Which the marble in its superfluous shell does not contain
To break the marble spell is all that the hand
That serves the brain can do50

VI.

The Alexandrian Neo-Platonist Plotinus, with whose writings

Michaelangelo was surely acquainted, remarked, "The form is
in the sculptor long before it ever enters the stone," a biological
truth that enables the sculptor to fashion his work and the
spectator to appreciate it.51 But if the form is in the artist (and
the spectator), maybe neither need the forms in the outside
world at all. That was the real starting point for the work of the
Russian suprematist Kasimir Malevich, a neurobiologically in
teresting artist, who wrote that "art wants nothing further to do
with the objective world as such." The use of the word "further"

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 90 Semir Zeki
here gives biological credibility to Malevich's doctrine because,
as discussed above, the brain needs to be visually nourished at
critical periods after birth so as not to remain almost indefinitely
blind. So the nonobjective sensation and art of which Malevich
speaks is really the introspective art of a brain that is well
acquainted with the objective world; it has already selected all
the information necessary to be able to identify and categorize
objects. And true to its aims of being a search for essentials and
constants, we find that as art developed more and more in the
modern era, much of it became better and better tailored to the
physiology of the parallel processing-perceptual systems and the
visual areas that we have only recently discovered, and specifi
cally to the physiology of single cells in them. The physiology of
these areas is itself tailored to extract the essential information
in the visual environment?there exists an Einf?hlung, that un
translatable term signifying a link between the "pr?existent"
forms within the individual and the forms in the outside world
that are reflected back, the "art de peindre des ensembles nouveaux
emprunt?s non ? la r?alit? visuelle, mais ? celle que sugg?rent ?
l'artiste l'instinct et l'intuition," as Guillaume Apollinaire said
of cubism.52
Physiologically, the Einf?hlung is expressed in what I have
called the art of the receptive field and I shall give but two brief
examples of it here.53 The receptive field is one of the most
important concepts to emerge from sensory physiology in the
past fifty years. It refers to the part of the body (in the case of the
visual system, the part of the retina or its projection into the
visual field) that, when stimulated, results in a reaction from the
cell, specifically, an increase or decrease in its resting electrical
discharge rate. To be able to activate a cell in the visual brain,
one must not only stimulate in the correct place (i.e., stimulate
the receptive field) but also stimulate the receptive field with the
correct visual stimulus, because cells in the visual brain are
remarkably fussy about the kind of visual stimulus to which they
will respond. The art of the receptive field may thus be defined
as that art whose characteristic components resemble the charac
teristics of the receptive fields of cells in the visual brain and
which can therefore be used to activate such cells.

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 91
One group of cells, discovered by David Hubel and Torsten
Wiesel in 1959, will only respond to lines of particular orienta
tion, since the orientational preferences of different cells are
different and each responds increasingly more grudgingly as one
departs from the preferred orientation until the response disap
pears at the orthogonal orientation. Such cells are a prominent
feature of area VI and some other areas surrounding it, notably
V3, but they are also found in other areas. They are usually
considered to be the physiological "building blocks" of form
perception, though how one moves from such cells to the cre
ation of forms remains unknown. It is interesting that the line is
among the most prominent features of the "nonobjective" art of
Malevich and his successors. Lines are the predominant and
sometimes only feature in the paintings of artists as diverse as
Olga Rozanova, Barnett Newman, Robert Motherwell, Ellsworth
Kelly, Gene Davis, Robert Mangold, and Ad Reinhardt, to men
tion but a few. Together with the rectangle and the circle, they
were considered by Malevich to be the most elemental aspect of
the nonobjective world. Mondrian, too, came to emphasize lines
but reached that conclusion from an intellectually (though per
haps not physiologically) different route. Art, he believed, "shows
us that there are also constant truths concerning forms," and it
was the aim of objective art, as he saw it, to reduce all complex
forms to one or a few universal forms, the constant elements
that would be the constituent of all forms. Thus, as he said, to
"discover consciously or unconsciously the fundamental laws
hidden in reality" and "to create pure reality plastically it is
necessary to reduce natural forms to the constant elements."54
He sought, in other words, the Platonic Ideal for form (though
he did not describe it in these terms). This search led to the
vertical and horizontal lines, or so he believed. These "exist
everywhere and dominate everything." Moreover, the straight
line "is a stronger and more profound expression than the curve"
because "all curvature resolves into the straight, no place re
mains for the curved."55 He wrote, "Among the different forms,
we may consider those as being neutral which have neither the
complexity nor the particularities possessed by natural forms or
abstract forms in general."56

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 92 Semir Zeki
This emphasis on line in many of the more modern and ab
stract works of art, in all probability, is derived not from a
profound knowledge of geometry but simply from the experi
mentation of artists to reduce the complexity of forms into their
essentials or, to put it in neurological terms, to try and find out
what the essence of form as represented in the brain may be.
This is my interpretation, not that of artists, but I cannot see
that it is any less valid than other interpretations. Kahnwiler, the
art dealer, tells us, "C'est uniquement l'apparition, chez les
cubistes, de lignes droites ... qui a fait croire a une g?om?trie
dont il n'y a, en r?alit?, nulle trace. Ces droites . . . reflets de la
base m?me, de l'a priori, de toute perception visuelle humaine,
se retrouvent, en fait, dans toute oeuvre d'art plastique des que
le souci d'imitation a disparu."57 This is as explicit a statement
as any?coming from one who, if not an artist himself, was at
least well acquainted with artists and their work?that the artist
is trying to represent the essentials of form as constituted in his
visual perception, which I take to mean the brain. Gleizes and
Metzinger, both artists, emphasized the straight lines and the
relationship that they have to each other, as did Mondrian. They
wrote, "The diversity of the relations of line to line must be
indefinite; on this condition it incorporates quality, the incom
mensurable sum of the affinities perceived between that which
we discern and that which pre-exists within us."58 Once again, I
interpret "that which pre-exists within us" to mean that which
is in our brains. Although Gleizes and Metzinger are discussing
here the relations between lines, it is nevertheless lines that they
have chosen to emphasize. Equally interesting are the specula
tions of M?cislas Golberg, a man said to have had a powerful
influence on Matisse. In his book La Morale des Lignes he
emphasized lines, especially the vertical and the horizontal, and
dreamed of a return to geometry, "mais une g?om?trie mitig?e,
soumise elle-m?me ? des lois de simplification et d'unification,"
which he thought was important for "le d?pouillement de la
r?alit? dans sa forme la plus abstraite," which in turn was
essential for "la simplification et la modernisation du dessin."59
And although he attached subjective sentiments to the vertical
and the horizontal, it is nevertheless these that he thought of as
important in modernizing art. "And is this not already a very

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 93
appreciable contribution to artistic evolution and, above all, to
the intelligence of contemporary art where the line, presented
sometimes without the support of a traditional 'subject,' has to
be interpreted and understood by itself and for itself?"60
I do not mean to imply that it is uniquely the stimulation of
the orientation-selective cells in the brain that results in the
aesthetic experience produced by a Malevich or a Barnett Newman,
but only that the constituent elements of these works are a
powerful stimulus for these cells and, moreover, that a brain
deprived of such cells?either because of blindness during the
critical period after birth or through pathological reasons?will
not be able to appreciate these paintings at all. Given the impor
tance that lines have assumed in much of modern and abstract
painting, and given that lines constitute probably the most basic
visual stimulus with which to excite a very important category
of cell in the cortex, it is interesting to ask whether the relation
ship between the two is entirely fortuitous.
It is in kinetic art that we find one of the best examples of the
art of the receptive field, and its evolution powerfully shows
how an art form became better tailored to the physiology of a
specific visual area in the brain?area V5, in which visual mo
tion is emphasized.61 Kinetic art was born out of a dissatisfac
tion, ostensibly due to social and political reasons, with an art
form that seemed to exclude movement or the fourth dimension,
as Naum Gabo called it. The demand for its inclusion was
strongly featured in the Futurist Manifesto of Naum Gabo and
Antoine Pevsner and in Marinetti's Manifesto of Futurism. For
all the shrill demands, especially from the Italian artists, move
ment was usually represented statically, as in Giacomo Balla's
paintings or those of Umberto Buccioni. There are a few early
exceptions, such as Gabo's Kinetic Sculpture, but they are rare.
Marcel Duchamp, influenced by the chronophotography of Jules
Etienne Marais in France, began to produce paintings that sug
gested movement statically; of these the most famous is perhaps
Nude Descending the Staircase II. From about 1910 onwards,
motion was very much on Duchamp's mind although he did not
exploit it explicitly, perhaps because he did not know how to do
so or had not yet settled on the best way of doing so. Perhaps
Rickey is right in asserting that "Duchamp showed, by deferring

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 94 Semir Zeki
his work with movement for years and confining it to optical
phenomena, that his concern therein was dadaist and superfi
cial."62 At any rate, by 1913 he had produced his famous Bicycle
Wheel, the "Ready-Made," which he called a Mobile. Although
usually immobile when exhibited in an art gallery, it is com
monly thought of as a precursor to kinetic art, even though
Duchamp himself did not consider this or machines in general to
be artistic objects, referring to them as "non-art."63 Indeed, the
Bicycle Wheel was to him only one ready-made among many,
which included such interesting objects as urinals?"art without
an artist," he called it, a concept that later was to be exploited
so well commercially by Andy Warhol, who, it is said, showed
the world that anything could be famous for fifteen minutes. The
real incorporation of motion in Duchamp's hands came much
later, when he produced his Rotoreliefs in the 1920s.
Duchamp was not alone in trying to emphasize motion, but
the gulf between the idea and its implementation in works of art
was not much easier for other artists; it required some degree of
technical mastery, of getting at least parts of the work of art into
motion, which is perhaps one reason why its actual incorpora
tion into works of art was to take a relatively long time. The
surrealists, too, who desired a retreat from all that was rational
and predictable, found in motion the unpredictability for which
they had yearned. Picabia designed imaginary machines, such as
his Machine tournez vite and his Parade amoureuse, the latter
somewhat reminiscent of Duchamp's La mari?e and, like it,
lacking the real motion that it exalted. Until Calder invented his
mobiles, the generation of motion depended upon machines, and
machines did not seem beautiful or desirable works of art to
everyone, not even to the cynical Duchamp.
It was in fact Alexander Calder who best developed the art of
the mobile, popularized it, and planted it in the popular mind. In
many ways, the mobile was an ingenious invention. It was not
dependent upon any profound knowledge of motors and engi
neering, although Calder's first mobiles were power driven.
Mobiles, in other words, were relatively easy to execute. Motion
was the dominant element, and to aid the dominance Calder
decided to limit himself largely to the use of black and white, the
two most contrasting colors, as he called them. Red was to him

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 95
the color best opposed to these two, but all the secondary colors
"confused" the clarity of the mobiles.64
One of the specializations in the human visual brain is that for
visual motion. This specialization is centered on area V5, where
all cells are selectively responsive to motion and the great major
ity are also selective for the direction of motion, responding
vigorously when the stimulus moves in one direction but remain
ing silent or even being inhibited when it moves in the opposite
or "null" direction. These cells are indifferent to the color of the
stimulus and usually indifferent to the form as well; indeed,
most of them respond best when the stimulus is a spot that is a
fraction of the receptive field size. It is interesting to consider
here how the mobiles of Calder stimulate the cells of area V5.
Viewed from a distance, each element of the mobile is a sort of
spot, small or large, depending upon its size. Once it moves in
the appropriate direction within the receptive field of a cell in
V5, it will lead to a vigorous response from it. In a mobile, of
course, the different elements will move in different directions,
and each element will stimulate not one but many cells, each cell
(or group of cells) being specifically tuned to respond to motion
in the respective direction in which the element of the mobile is
moving. There are many other interesting features about our
perception of mobiles that I have discussed elsewhere,65 but the
important point to emphasize here is that in its development,
kinetic art, especially in the hands of Calder, resulted in works
that act as perfect stimuli for the cells of V5. Another important
feature that perhaps reinforces the view that I present here?that
artists try to learn something about the organization of the
visual brain, though with techniques unique to them?is found
in the general emphasis on movement and in a lack of emphasis
on color and form, mirroring so well the physiology of V5.
In giving the above two examples, it is worth emphasizing
that there is much about the perception of lines and of motion
that we still do not understand physiologically, and it is there
fore impossible to relate directly the experience of even one line
to what really happens in the brain. If viewed at a sufficiently
close distance, even a single vertical line, for example, may fall
on the receptive fields of many cells that are specific for the
vertical orientation; how the brain combines the responses of

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
96 Semir Zeki
these cells to indicate a continuous vertical line is a mystery that
neurology has not yet solved, and nor has it solved the question
of how it may differentiate one vertical line from other vertical
lines that are distinct from it and indeed differentiate the entire
tableau from what surrounds it. No less puzzling is the coher
ence in a work of kinetic art, where the brain can interpret
different elements that fall on different receptive fields as form
ing part of the same work. However, these unsolved neurologi
cal problems should not inhibit us from noting that what the
physiologists call the building blocks of form?the oriented lines?
are the same ones that artists, keen on representing the constant
elements of form, have used and that what physiologists con
sider to be the building blocks in the perception of motion?the
cells that respond to motion in a given direction?are the very
ones used by an artist such as Calder in his mobiles.

VII.

Jean-Paul Sartre was quite ecstatic about the work of Calder. He

wrote, "La sculpture sugg?re le mouvement, la pienture sugg?re
la profondeur o? la lumi?re. Calder ne sugg?re rien: il attrape de
vrais mouvements vivants et les fa?onne. Ses mobiles ne signifient
rien, ne renvoient ? rien qu'? eux m?mes: ils sont, voil? tout; ce
sont des absolus."66 This is not an uninteresting observation,
and one can draw at least a superficial parallel between his
absolutes and the absolutes of form that were such an obsession
of Mondrian and others. The search for these absolutes leads to
abstraction. Abstraction has been used to describe many differ
ent schools and movements; I use it here in its broadest sense, to
signify works in which neither the work itself nor its constituent
parts represent any recognizable objects in the visual world
(noniconic abstraction). It is obvious that in this context, ab
stract art differs radically from representational art. The ques
tion is whether there is a significant difference in the pattern of
brain activity when subjects look at abstract as opposed to
representational art?
A hint that there may be substantial differences can be found
in recent imaging experiments from this laboratory, which have
been inspired in part by the fauvist dream of liberating colors to

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 91
give them more expressive power. But from what can color be
"liberated"? It is not easy to liberate it from form for good
physiological reasons. The fauvists therefore settled on a differ
ent solution, which was to invest objects with colors that are not
usually associated with them, as Andr? Derain's View of Charing
Cross Bridge and other fauvist paintings testify. Unknown to
them, and only uncovered in our imaging experiments, they were
exploiting different neurological pathways in the visual brain
than the ones used in representational art, where objects are
vested in the "correct" colors.
Color is a biological signaling mechanism that exemplifies
very well the brain's quest for knowledge under continually
changing conditions. It is common knowledge that the basis of
color vision is found in light?which itself has no color, being
electromagnetic radiation?having many different wavelengths,
stretching from red (long wave) at one end to blue (short wave)
at the other, and the fact that different surfaces have different
efficiencies for reflecting light of different wavelengths. What
the brain seemingly does is to compare the efficiency of different
surfaces for reflecting light of the same wavebands and thus
make itself independent of the actual amount of light of any
given waveband reflected from a single surface, since the latter
changes continually depending upon the illuminant in which the
surface is viewed. If the brain assigned a color to a surface as a
function of the wavelength composition of the light reflected
from it?characterizing it as green when it reflects more green
(middle-wave) light and blue when it reflects more blue (short
wave) light, with the dominant wavelength constituting a sort of
code that the brain has to decipher?then the brain would be at
the mercy of any and every change in wavelength composition
reflected from the surface. Instead the brain has evolved an
ingenious mechanism, whose neural implementation remains
obscure, to take the ratio of light of a given wavelength reflected
from the center and the surround. While the precise amount of
light of a given wavelength reflected from a surface changes, the
ratio of light of that same waveband reflected from the surface
and from surrounding surfaces always remains the same. Color
is therefore a construction of the brain, an interpretation that it
gives to the reflective efficiency of different surfaces for the

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 98 Semir Zeki
different wavelengths of light, which is why James Clerk Max
well referred to color as "a mental science." But to be able to
take ratios, there must be a boundary between one surface and
the surrounding surface, and that boundary has a shape?hence
the impossibility (except in very rare pathological conditions) of
divorcing color, and thus liberating it, from shape. Color there
fore follows the logic of the brain's operations. Andr? Malraux
was right when he drew attention in Les Voix du Silence to
Cezanne's remark that "Il y a une logique color?e; le peintre ne
doit ob?issance qu' ? elle, jamais ? la logique du cerveau,"
describing it as "cette phrase maladroite [qui] nous r?v?le pourquoi,
sur l'essentiel de son art, tout peintre de g?nie est un muet,"
although I would have preferred it if Malraux had said "devrait
?tre muet" instead.67
It is obvious that at the ratio-taking, computational stage
there are no "wrong" colors. Making a square red is as good as
making it blue. Edwin Land's paradigm in studying color vision
consisted of an abstract multicolored scene with no recognizable
objects, rather like the paintings of Mondrian. When humans
view such a scene the increase in regional activity in their brain
occurs in area V4, the color center. But colors are not normally
viewed in this way; they are instead properties of surfaces and
objects. What happens in the brain when humans view colored
objects and scenes depends upon whether the objects are dressed
in the right or the wrong colors, but in either case it is different
from the activity produced by colors in the abstract, as in a
Mondrian. If the objects are dressed in normal colors a more
extensive part of the brain, including the frontal lobes, becomes
active in addition to V4. But if they are dressed in abnormal
colors, as in fauvist paintings, a different set of areas (in addi
tion to V4) becomes active.
These results are replete with neurological interest, but in the
present context they allow us to draw two interesting conclu
sions. The first is that abstract paintings in color do not need to
recruit those additional brain areas that are mobilized when we
view representational art in color. The second is that the fauvists
had unwittingly uncovered certain truths about the organization
of the visual brain about which they were and remain ignorant?
namely, that their art used pathways that are quite distinct from

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 99
the ones used by representational art that portrays objects in
normal colors.

VIII.

I have tried, using only a few examples, to explain that we have

learned enough about the visual brain in the past quarter of a
century to begin to study the biological foundations of aesthet
ics. Aesthetics, like all other human activities, must obey the
rules of the brain of whose activity it is a product, and it is my
conviction that no theory of aesthetics is likely to be complete,
let alone profound, unless it is based on an understanding of the
workings of the brain. There is, of course, much that has been
left unsaid in this brief essay?about topics such as portrait
painting, impressionist art, or op art?but these different ten
dencies can also be discussed within the overall context of a
search for knowledge. There are other questions that are diffi
cult to write about at present: why some artists are drawn to
paint in a particular genre, why some of us prefer certain schools
to others, the role of the imagination in producing works of art,
the relationship between artistic creativity and sexual impulses
(since they are both reproductive processes), the emotive power
of works of art, the role of culture and historical knowledge in
appreciating and interpreting works of art. But I have been
exploring a topic here that is new and have concerned myself
exclusively with the perceptive aspects. There is much that has
yet to be discovered and described.
The approach that I have adopted may seem distasteful to
some. Art, they might say, is an aesthetic experience whose basis
is opaque and indeed should remain so. It has derived much of
its value from the different way in which it arouses, satisfies,
and disturbs different individuals, and to profane physiologi
cally the secrets of fantasy in this way implies that what happens
in one brain is very similar to what happens in other brains
when we view works of art. There is substance to that argument.
But we should consider that what happens in different brains
when we view works of art is very similar, at least at an elemen
tary level, which is one reason why we can communicate about
art and through art without the need for the written or spoken

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 100 Semir Zeki
word. And no profound understanding of the workings of the
brain is likely to compromise our appreciation of art any more
than our understanding of how the visual brain functions is
likely to compromise the sense of vision. On the contrary, an
approach to the biological foundations of aesthetics is likely to
enhance the sense of beauty?the biological beauty of the brain.

ACKNOWLEDGMENTS

I gratefully acknowledge the help I received from the staff of the J. Paul Getty
Museum in California while I was a visiting museum scholar there at the kind
invitation of Mr. John Walsh, Director of the Museum. I am also much indebted
to Professor K. Bartels and to Andreas Bartels for their insightful comments, es
pecially concerning the Platonic doctrines.

ENDNOTES

^erstle Mack, La Vie de Paul C?zanne, quoted by Christopher Gray in Cubist

Aesthetic Theories (Baltimore, Md.: The Johns Hopkins Press, 1953).
2Naum Gabo, Of Divers Arts (New York: Pantheon Books, 1962).
3Paul Flechsig, Gehirnphysiologie und Willenstheorien, Fifth International Psy
chology Congress (1901), 73-89, trans. Gerhardt von Bonin in Some Papers
on the Cerebral Cortex (Springfield, 111.: Thomas, 1960).
4M. Vialet, "Consid?rations sur le centre visuel cortical ? propos de deux
nouveaux cas d'h?mianopsie suivis d'autopsie," Archives d'Ophtalmologie
14 (1894): 422-426.
5A. Monbrun, "Les affections des voies optiques r?trochiasmatiques et de
l'?corce visuelle," in Baillart et al., eds., Trait? d'Ophtalmologie, vol. 6 (Paris:
Masson, 1939).
6These are the terms of neurologists, not mine; they were current until the last
two decades.
7Semir Zeki, A Vision of the Brain (Oxford: Blackwell, 1993).
8Henri Matisse, Ecrits et propos sur l'art (Paris: Hermann, 1972), 365.
9Semir M. Zeki, "Functional Specialization in the Visual Cortex of the Rhesus
Monkey," Nature 274 (1978): 423-428.
10Ibid.

nMargaret S. Livingstone and David H. Hubel, "Anatomy and Physiology of a

Color System in the Primate Visual Cortex," Journal of Neuroscience 4
(1984): 309-356. Margaret S. Livingstone and David H. Hubel, "Connec

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 101
tions Between Layer 4B of Area 17 and the Thick Cytochrome Oxidase
Stripes of Area 18 in the Squirrel Monkey," Journal of Neuroscience 7
(1987): 3371-3377. S. Shipp and Semir Zeki, "Segregation of Pathways
Leading from Area V2 to Areas V4 and V5 of Macaque Monkey Visual Cor
tex," Nature 315 (1985): 32-325.
12Zeki, A Vision of the Brain.
13Konstantinos Moutoussis and Semir Zeki, "A Direct Demonstration of Percep
tual Asynchrony in Vision," Proceedings of the Royal Society of London, Se
ries B 264 (1997): 393-399.
14Semir Zeki, "Parallel Processing, Asynchronous Perception and a Distributed
System of Consciousness in Vision," The Neuroscientist (forthcoming).
15Semir Zeki, "A Century of Cerebral Achromatopsia," Brain 113 (1990):
1721-1777.
16Albert Gleizes and Jean Metzinger, Cubism (London: Fisher Unwin, 1913).
17Henri Matisse, "Notes d'un peintre," La Grande Revue LII (24): 731-745.
Reprinted in Jack D. Flam, ed., Matisse on Art (Oxford: Phaidon, 1978).
18Jacques Rivi?re, "Present Tendencies in Painting," Revue d'Europe et
d'Am?rique (March 1912): 384-406. Reprinted in Charles Harrison and Paul
Wood, eds., Art in Theory (Oxford: Blackwell, 1992).
19Herbert Read, The Philosophy of Art (London: Faber and Faber, 1964).
20This does not represent the view of all ancient Greeks; Aristotle, Plato's stu
dent, turned away from it.
21The example that Plato gives, that of a couch, is derived from the kind of fur
nishing used in the symposia frequented by Plato and his elite circles. The idea
is created by God, the craftsman (?T|jiio\)pYoQ makes a first example of it,
and artists subsequently represent different single views of the craftsman's
creation.
22Arthur Schopenhauer, The World As Will and Idea, 3d book, from Albert
Hofstader and Richard Kuhns, eds., Philosophies of Art and Beauty (Chi
cago, 111.: University of Chicago Press, 1964).
23John Constable, Syllabus of a Course of Lectures on the History of Landscape
Painting (London: Royal Institution of Great Britain, 1836).
24B. R?ssel, History of Western Philosophy (London: Allen and Unwin, 1946).
25G. W. F. Hegel, Aesthetics, vol. I, trans. T. M. Knox (Oxford: Clarendon Press,
1975).
26Ibid. Emphasis added.
27See, for example, Marius von Senden, Space and Sight (London: Methuen and
Co., 1932).
28David H. Hubel and Torsten N. Wiesel, "The Ferrier Lecture?Functional Ar
chitecture of Macaque Monkey Visual Cortex," Proceedings of the Royal
Society of London, Series B 198 (1977): 1-59.

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 102 Semir Zeki
29H. Harlow, "Love Created?Love Destroyed?Love Regained," in Mod?les
Animaux du Comportement Humain, no. 198 (Paris: Editions du Centre
National de la Recherche Scientifiques, 1972), 13-60.
30Semir Zeki and Balthus, La Qu?te de l'Essentiel (Paris: Les Belles Lettres,
1995).
31L. C. Perry, "Reminiscences of Claude Monet from 1889-1909," American
Magazine of Art XVIIII (1927), quoted by John Gage, Colour and Culture
(London: Thames and Hudson, 1993).
32N. K. Logothetis et al., "Shape Representation in the Inferior Temporal Cortex
of Monkeys," Current Biology 5 (1995): 552-563.
33Fritz Novotny, "Das Problem des Menschen C?zannes im Verhaeltnis zu
Seiner Kunst," Zeitschrift fur Aesthetic und Allegemeine Kunstwissenschaft
26 (1932): 278.
34Mack, La Vie de Paul C?zanne, quoted in Gray, Cubist Aesthetic Theories.
35Daniel Henry Kahnweiler, Juan Gris: Sa Vie, Son Oeuvre, Ses ?crits (Paris:
Gallimard, 1946), 326. Emphasis added.
36Ibid.
37Ibid.

38Rivi?re, "Present Tendencies in Painting," in Harrison and Wood, eds., Art in

Theory. Emphasis in original.
39Piet Mondrian, The New Art?The New Life: The Collected Writings of Piet
Mondrian (Boston, Mass.: G. K. Hall, 1986), 338-341.
40Kasimir Malevich, The Non-Objective World, trans. Howard Dearstyne (Chi
cago, 111.: P. Theobald, 1959), 102. Emphasis in original.
41Nilsen Laurvik, Is It Art? Post-Impressionism, Futurism, Cubism (New York:
The International Press, 1913).
42Marcel Proust, Vermeer de Delft (Paris: La Pl?ade, 1952), 128.
43Paul Claudel, L'oeil ?coute (Paris: Gallimard, 1946), 240.
44Claudel used the English terms since there is no good French equivalent.
45Here I disagree with Claudel, who says that the spectator is immediately in
vited in. This is true of some but not most of Vermeer's work; a notable ex
ception is Portrait of a Young Girl.
46Semir Zeki, in conversation with Balthus, Connaisance des Arts, 477.
47Schopenhauer, The World As Will and Idea.
48Edward Snow, A Study of Vermeer (Berkeley, Calif.: University of California
Press, 1979), 214.
49Gleizes and Metzinger, Cubism.
50I have used the translation by Symonds; other translations do not use the word
brain. The actual word used for brain in the original is intelleto. In Latin,
intellectus means perception or "a perceiving"?see R. J. Clements,

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms
 Art and the Brain 103
Michelangelo's Theory of Art (New York: New York University Press, 1961),
15?and Symonds has astutely, in my view, rendered it into "brain."
51Plotinus, Ennead V, Eighth Tractate: On the Intellectual Beauty, republished
in Hofstadter and Kuhns, eds., Philosophies of Art and Beauty.
52Guillaume Apollinnaire, Les Peintres Cubistes: M?ditations Esth?tiques (Paris:
Berg International, 1986). Apollinaire does not use the term Einf?hlung. The
notion of Einf?hlung in art was first elaborated by the German philosopher
Robert Vischer in a work entitled ?ber das optische Formgef?hl. Wilhelm
Worringer developed the notion further and applied it to abstract art in his
doctoral thesis at Berne University, published in 1908, entitled Abstraktion
und Einf?hlung, but Worringer sought other, nonneurbiological, explana
tions for the then developing abstract art. See D. Vallier, l'Art Abstrait (Paris:
Librarie G?n?rale Fran?aise, 1980).
53Semir Zeki, "The Woodhull Lecture: Visual Art and the Visual Brain," Pro
ceedings of the Royal Institute of Great Britain 68 (1997): 29-63. A more
detailed account is given in my forthcoming book, Behind the Seen.
54Mondrian, New Art?New Life, 338-341. Emphasis in original.
55Ibid., 75-81.
56Ibid.

57Kahnwiler, Juan Gris.

58Gleizes and Metzinger, Cubism.

59M?cislas Golberg, La Morale des Lignes, quoted in Ellen C. Oppler, Fauvism

Reexamined (New York: Garland Publishing, 1976), 413.
60Pierre Aubery, "M?cislas Golberg et l'art moderne," Gazette des Beaux Arts
66 (1965): 339-344.
61For a more detailed treatment of this subject, see Semir Zeki and M. Lamb,
"The Neurology of Kinetic Art," Brain 117 (1994): 607-636.
?2George W. Rickey, "The Morphology of Movement: A Study of Kinetic Art,"
Art Journal 22 (1963): 220-231.
63E. Lebovici, "Bouge, moeurs et ressuscite," Art studio 22 (1991): 6-21.
64Alexander Calder, extract from Julien Alvard and Roger van Gindertael, eds.,
T?moignages pour l'art abstrait (Paris: Editions Art d'aujourd'hui, 1952).
65Zeki and Lamb, "The Neurology of Kinetic Art."
66Jean-Paul Sartre, "Situations III," trans. Wade Baskin, in Essays in Aesthetics
(London: Peter Owen Ltd., 1964).
'Andr? Malraux, Les Voix du Silence (Paris: La Pl?iade, 1951), 344.

This content downloaded from 132.248.197.194 on Tue, 09 Aug 2016 18:18:00 UTC
All use subject to http://about.jstor.org/terms