|
return
to homepage return
to updates
CLIVE
BELL and
Formalism
![](bell2.jpg)
by
Miles Mathis
I have
mentioned Clive Bell before in my art papers, namely in a couple
of early papers
that discuss Cezanne. But I have not yet counter-critiqued Bell
with any rigor. Some will ask why I would do so now, given that
Bell is not exactly a household name in art criticism. I do
because I believe that Bell's influence is greater than is
commonly supposed. Much of the burgeoning stature of Cezanne in
the early part of the 20th century was due, one way or the other,
to Bell, and this growing influence of Cezanne and formalism led
directly to where we are now. In other words, the mistakes of
Bell have had a lasting effect on art criticism. These mistakes
are among the foremost mistakes of Modernism, and they have never
been pulled apart to the extent they must be. The mistakes have
acted as a hidden foundation beneath the other more obvious and
more recent mistakes, and it is this foundation that must be
addressed.
Bell was part of the famous Bloomsbury group
that flowered just before the first World War. He was married to
Virginia Woolf's sister, and so was positioned perfectly to write
for the magazines of the time. His first major book and the one I
will counter-critique here, Art,
was published in 1914 but compiled from articles Bell had written
in the years just before that, up to 1913. In this way, Bell
slightly predates his friend Roger Fry, who published Vision
and Design in 1920. Although Fry
published his “Essay in Aesthetics” before Bell really took
off, even Fry admits that Bell got to Cezanne before him, which
may be the deciding factor in all this. Cezanne is their crux, as
Whistler has been mine. In fact, I will argue here along the
lines of Whistler, who would have mocked any painting of Cezanne
and any output of either Fry or Bell as an “Essay on
Inaesthetics.”*
Although the main lines of either Fry's
or Bell's arguments appear at first glance to have a similarity
to Whistler's argument about form, and although the their
arguments have been conflated with his by art critics and artists
who came after, the differences are crucial. Let us see precisely
how. Whistler insisted in his Ten O'Clock Lecture that art was
not about morals or stories or descriptions; rather it was, or
should be, an arrangement of forms. This led, in pretty direct
channels, to Bell insisting that art was defined by “significant
form.” For him, it was not the subject matter that was primary,
or that caused “aesthetic pleasure,” it was the forms. It was
the shapes and colors.
You can see why Bell thought he
was following Whistler, and why many others have thought the
same. If you ignore the context, read in low light, and don't pay
proper attention, you can easily misunderstand Whistler in this
way. Even Swinburne misunderstood Whistler in this way, and it
cost them their friendship. However, I don't blame Whistler for
the misreading of any of these people, Bell, Swinburne, Fry, or
any of the rest. It is their fault for not listening closely
enough. Or perhaps, not being natural painters themselves, they
simply didn't have the ear to hear. Whistler admitted in court,
in his suit against Ruskin, that it “was like pouring notes
into the ears of a deaf man”, trying to make non-artists
understand what art was about. But I will try once more to pour,
and those with ears should try once more to open them.
When
Whistler tried to turn painting away from morals and stories, it
was a reaction to a glut of poor storytelling and hamhanded
moralizing by the painters of the first six decades of the 19th
century. He was therefore forced by circumstance to accentuate
the non-narrative aspects of painting. But we can see just by
looking at his work that he was not interested in jettisoning
subject matter from painting. His works rely on their subject
matter: it is the subjects that have the
forms. Venice was subject matter, and it
had form. His interesting people had form, their dresses had
color, the Thames had fog, and so on. In this way, his painting
was not an arrangement of abstract or theoretical form, it was an
arrangement of real things. It was a selection from life. Nor had
he any idea of jettisoning beauty from painting, since his
arrangements were chosen because they were beautiful. It wouldn't
have been enough for him just to paint a beautiful lady: that
wasn't artistic in itself. No, one must arrange
a beautiful lady in a beautiful composition. It was this
arrangement that was the art of it. That and
actually painting the arrangement. There
was arrangement in the very paint itself, in the way it was
applied.
The lady was already beautiful, and the artist
got no credit for that. It is what the artist did with this lady
that made him a good artist or a poor one. A poor artist could
make even the loveliest lady into a vulgar painting. A good
artist could make a dreary or common scene into a masterpiece.
But this did not make the dreary scene superior to the
lovely lady, as a general subject. Non-artists have misunderstood
such statements by artists to mean that lovely ladies should
never be painted again, or that the only scenes worthy of
painting were dreary or horrid ones. Nothing like that. The only
things artists want to forbid is bad paintings, and you can make
a bad painting of anything. You can ruin a glimpse of heaven, by
improper arrangement.
So, in highlighting the importance
of forms Whistler was in no way discounting subject matter,
beauty, or even storytelling. Whistler wasn't much interested in
storytelling himself, but for him even a narrative or historical
painting could have been successful, as long as it was also
successful as an arrangement of forms. The narrative or history
didn't disqualify a thing from being art, but a story arranged
poorly was like a beautiful woman in a gaudy or gawky
composition: all the art in the thing had been lost, and thereby
the beauty.
Bell never understood this. He took a
highlighting of form as an excuse to jettison non-form, and once
“significant form” became the defining quality of art for
him, all else in a painting could be dismissed. True, Bell didn't
immediately dismiss all other content, but his argument allowed
others to easily push his theory to its inglorious end. To see
what I mean, we only need to study Cezanne. For Bell, the fact
that Cezanne had highlighted the form and played down the rest
was a sort of purification of art. But Whistler would have seen
it differently, I assure you. For although Cezanne was good at
highlighting the form, he was not good at arranging it. His lines
and colors are awkward, unbalanced, and, in a word, vulgar.
Precisely because they lack grace and harmony, Whistler would say
they are inaesthetic. Aesthetics is
this grace and harmony. The forms cannot just be highlighted or
simplified, they have to be arranged aesthetically, which means
they must be arranged and painted
with consummate skill.
Neither Bell nor Fry were very
good painters, which is probably why they thought Cezanne was a
good painter. He was good because he was more like them. Here is
a portrait of Bell by Fry.
You see much of the clumsiness of Cezanne in Fry. Fry was
actually better at figures than Cezanne, but he still wasn't very
good. His line lacked the natural grace of the best painters, and
the same can be said of his color. I think it is entirely
possible that these critics who have championed Cezanne were not
so much championing Cezanne as they were championing themselves.
I suspect that the idea was buried somewhere deep within them
that if they could sell Cezanne to the world, they could
eventually sell themselves to the world. It was a lowering of
standards by roundabout means. And, as it turns out, this is what
eventually happened. The critics began as Fry and Bell did,
subtly turning the field away from grace and harmony by
championing Cezanne and others like him. A few decades later they
would give up this subtlety of argument and admit that they were
undermining art history on purpose. Clement Greenberg was the one
who led this phase of the collapse. And a decade or two later,
these critics had actually taken over art, becoming artists. We
see this first with Barnett Newman, a person with no artistic
ability at all, but great skill at manipulating opinion. After
Newman, this would become the standard. Art had been redefined as
art criticism, and artifacts were no longer created to cause
“aesthetic pleasure.” No, they had fallen far beneath the
original vulgarization of Bell. They were now created to cause a
critical
experience, or to say it another way, a delusion of artistic
relevance. That and to promote the critic as artist.
Yes,
Bell was certainly guilty of a horrible vulgarization of art, but
compared to where art ended up a century later, Bell actually
looks pretty tame now. He knew how to write, had a fairly keen
sense of humor for a critic, a full command of the language, and
said many things that were true. That immediately puts him in an
entirely different category than the current art critic, who
can't put two sensible sentences in a line. Let us look at what
Bell actually said, and see what was true and what was false.
On page 9, Bell says that “we have no way of
recognizing a work of art than our feeling for it.” True,
though Tolstoy said the same thing years earlier. At the bottom
of the same page, he says, “All systems of aesthetics must be
based on personal experience—that is to say, they must be
subjective.” True—and false. All aesthetics must start with
personal experience, as in the feeling one has for particular
works of art. But that does not make aesthetics subjective. Bell
pretends to rigor, but he extends a long misunderstanding of
objective and subjective. Bell implies that all things that are
based on personal experience are subjective, but that isn't so.
According to the original subjective/objective split, a category
could be based on personal experience and still be objective, if
you were experiencing things that were intrinsic to the object.
“Subjective” doesn't mean “based on personal experience.”
It means, “the quality experienced is not in the object, it is
a construct of the mind.” Bell has not proven or argued that
aesthetics is a construct of the mind, or that the object does
not itself have significant form, therefore he has not shown that
art is subjective. He has simply assumed it, on page 9. That is
what one would call a fundamental error of philosophy.
It
doesn't take a lot of esoteric quibbling to show that Bell is
wrong, it only requires showing that objects of art have
significant form in them already, before I experience that form.
In other words, we must ask whether my experience makes the form
significant, or whether it is the form that makes my experience
what it is. Since the form exists before my experience of it and
in the absence of my experience of it, the logical assumption is
that the form causes the experience. If art critics want to argue
the opposite, I would think the burden of proof is on them. How
can my experience of something make that something significant,
if it was not already significant, or potentially significant?
Logically it cannot, so all argument to the contrary is
illogical. I, the viewer, do not make a painting beautiful by
looking at it; it makes me experience beauty. The painting is
primary and the experience is secondary. The painting is cause
and the feeling is effect.
Do I make an apple red by
looking at it? No, something intrinsic to the apple causes my eye
to see red. The eye does not give red to the apple, the apple
gives red to the eye. The eye may translate what it receives,
yes, but it is a receiver. That is the point. The subject
receives and the object gives, not the reverse. We could get that
just from the definitions, without further study; but no matter
how closely we analyze, we come to the same conclusion. Just as
red must be given us by the object, so must any other quantity or
quality, no matter how complex.
Interestingly, Ruskin
understood this perfectly. He had no confusion regarding subject
and object, and no such confusion of aesthetics, so we may assume
that this bastardization of aesthetics was not universal, or not
English at any rate, until the early 20th century. In “Of the
Pathetic Fallacy,”** Ruskin said,
Now,
to get rid of all these ambiguities and troublesome words at
once, be it observed that the word ' Blue' does not mean the
sensation caused by a gentian (flower) on the human eye; but it
means the power of producing that sensation; and this power is
always there, in the thing, whether we are there to experience it
or not, and would remain there though there were not left a man
on the face of the earth. Precisely in the same way gunpowder has
a power of exploding. It will not explode if you put no match to
it. But it has always the power of so exploding, and is therefore
called an explosive compound, which it very positively and
assuredly is, whatever philosophy may say to the contrary.
As
with blue, so with aesthetic form. Paintings are clearly objects
with unchanging forms and colors, and as such they have
unchanging artistic qualities. They are what they are, no matter
what one critic feels or says, or does not feel or does not say,
or what a million such say or feel. Although theories of
aesthetics may be built on personal experiences of viewers,
paintings are not. A theory of painting is not a painting.
Furthermore, theories are neither subjective nor objective.
Experiences are either subjective or objective, but theories are
either true or false. I have shown that the theory that all
experiences are subjective is false, by definition. A dream is a
subjective experience. Looking at a painting is not. And I will
go on to show that many other theories are false. They are false
not because they do not conform to experiences, but because they
do not conform to the information we receive from objects. That
is to say, they do not conform to the facts.
So I have
already shown that the dissolution of art and aesthetics in the
early part of the 20th century was tied to a dissolution in
philosophy. These absurd theories came from critics unable to
make simple logical distinctions, as between a painting and a
theory of painting, or from critics unable to understand
straightforward definitions of words, as in the definitions of
objective and subjective. This is interesting because we find the
same thing in science. On my other website, I have shown a
breakdown in thought in the sciences during the same period. It
would appear that people could no longer think straight about
anything, for reasons yet to be shown. Many have said that the
first World War was the cause, but I think it is clear that the
war was an effect, not a cause. The breakdown of sense occured
before the war, and so could not be caused by it. I will not
pursue the real cause here, but I suggest we would be better
looking for it where Nietzsche looked for it in the 1860's and
70's: in the breakdown of hierarchies and the rise of
“democracy.” I put democracy in quotes there because I have
shown elsewhere that it was not democracy, strictly defined, that
was or is the problem. The problem is a perverted sense of
equality, by which those who have less skill demand equal
consideration. We see that clearly here in this problem, where
people who can't paint well demand equal consideration as artists
or art experts. Real qualifications are ignored, while equality,
equal opinion, and relativism are promoted. This can only promote
a general breakdown of sense, as those with less sense are given
greater and greater pulpits.
We reach another big
philosophical problem with Bell as soon as we reach page 12. Bell
tells us that art is defined by significant form, and that
significant form is form that moves us artistically. That is
circular, and has no real content. It is to define art as that
which has artistic form, which can tell us nothing we didn't
already know. It is like defining redness as that quality which a
red thing has. Of course most definitions are circular like that,
but it does not prevent us from being annoyed with them. A
dictionary is forced to say things like that, the writer of a
book on aesthetics is not. The writer could just as easily write
a book on something else, or keep quiet.
In fact, Bell
begins his book with this pair of sentences:
It
is improbable that more nonsense has been written about
aesthetics than anything else: the literature of the subject is
not large enough for that. It is certain, however, that about no
subject with which I am acquainted has so little been said that
is at all to the purpose.
Unfortunately,
Bell corrects this by simply adding to the literature of the
subject. Defining art as significant form is not to the purpose.
It is more nonsense. It is not false, but it is blather.
Bell
does try to refine this definition a bit, but his refinement is
mainly more blather as well. He says that painting may be
“descriptive”, in which case it has other, lesser, qualities
of art, but is not aesthetic. Aesthetics is not just “suggesting
emotion or conveying information,” it is “being an object of
emotion.” In other words, in descriptive paintings “it is not
their forms but the ideas or information suggested by their forms
that affect us.”
So art must cause emotion directly,
without any mediation by ideas or information. Well, we can see
what Bell intends by this, and he is partially correct. On one
level, this is precisely how art does work. On another level, we
still have a lot of confusion here. Just to start with the most
obvious problem, we have Bell calling art an object of emotion,
when he just told us that art is subjective. If art is an object
of emotion, and is causing us to feel things without any
mediation by ideas, then art can hardly be subjective. Bell has
not only used the word “object” here, he has also used the
word “cause.” The art object is the cause of the aesthetic
response. That is practically the definition of aesthetic
objectivism.
Another problem is Bell's method of
separating the aesthetic from the non-aesthetic, in that he makes
you think the separation can be or should be complete. For the
truth is that aesthetic forms are just non-aesthetic forms
arranged in a certain way, as I pointed out with Whistler.
Whistler is not demanding an arrangement of abstract forms, he is
demanding an arrangement of existing, natural forms. These
existing natural forms are normally descriptive, and they may
still be descriptive after they are arranged, in which case they
will be both descriptive and non-descriptive in the same
painting. So there is no real division into descriptive and
non-descriptive. The categories are not at all exclusive of one
another.
This is crucial, because this misunderstanding
is what led subsequent artists and critics to try to jettison the
descriptive from painting. They thought that by minimizing the
desciptive they could maximize the non-descriptive or aesthetic
part of the painting. See Clement Greenberg, who based his career
on this. But we have seen that this method fails. This
“minimalism” has not maximized aesthetics, it has minimized
it. You cannot make art more artistic by jettisoning all real
objects. As Whistler knew, you make art more artistic by a more
artistic arrangement of real objects.
Why is this so? Why
has it proved impossible to separate the descriptive and the
non-descriptive? Why is abstraction always a ghost, even as
regards emotion? It is because Bell's aesthetic emotion is a
subset of all human emotion, and all human emotion is tied to the
real world. What I mean is that aesthetic emotion is not emotion
in some sort of artistic vacuum, purified from all previous
emotion and knowledge. You feel artistic emotion because you have
felt previous emotion, and all or most of this previous emotion
was felt for real things. That is to say, you cannot feel things
for paintings if you have not felt things for real people or
objects. This is why you cannot separate the descriptive from the
non-descriptive in a work of art. You need both. To elicit an
emotional response, you must have an arrangement of familiar
objects. The unfamiliar can cause no emotion but fear or unease.
The unfamiliar is neither pleasant nor beautiful.
Bell is
correct that it is the arrangement that is the art, although we
knew that from Whistler. But you cannot have an arrangement of
nothing. Nor can you have an arrangement of abstractions. The
arrangement of abstractions can only elicit emotion insofar as it
mimics familiar objects or compositions. An object of aesthetic
emotion is not an abstract form, it is a familiar form arranged
in a pleasing or poignant manner. The human mind is simply not
set up to respond emotionally to the abstract and unfamiliar. It
is set up to respond to the real, which is familiar and thereby
not abstract.
I
will be asked why people respond to abstract paintings, then.
Well, they don't, not strongly. Yes, they may respond to familiar
colors or pleasing patterns, but these responses are generally
quite weak. And in the rare cases that people respond strongly to
abstract paintings, we find it is the idea in the abstraction
that speaks to them, which immediately undercuts Bell's argument.
When I have heard people talk about Rothko, Agnes Martin,
Gottlieb, and so on, I have been struck by the amount of talk
about ideas I hear. For the people who like these paintings, the
abstractions don't seem to be objects of emotion, in the way Bell
is describing. These abstract paintings seem, rather, to be
suggesting ideas
or emotions, which makes them descriptive, by Bell's own
definitions.
As I have said in another paper, an abstract
painting is actually less direct than a descriptive one, since it
has to be interpreted to a greater extent by the mind. The mind
already knows what to make of a descriptive painting, so it is
free to intuit the pleasure in the arrangement. But with an
abstract painting, the mind has to first create a meaning. It
does this, most often, by attaching some previous explanation or
interpretation to the abstraction. It almost needs to have read
about the abstraction before it sees it, to know what to make of
it. It is then unclear if the viewer is feeling something about
the painting or about the explanation of it. Is it the text that
causes the emotion or the painting?
On the other hand, if
it is simply the colors and patterns that are causing pleasure,
it seems to me a weak drink. I have felt a low pleasure from such
paintings, the same sort of pleasure I felt from making God's
eyes with my own chosen yarn in grade school, but I wouldn't
attempt to define all of art on that pleasure. I have felt much
greater things in front of great paintings, and those feelings
can't be explained by Bell's theory of significant form. Form was
only a part of it. Subject matter was always also a large part of
it. It was not one or the other, but a combination.
![](padding.jpg)
Bell
tells us that Frith's Paddington Station
is not a work of art, but of course that is already going way too
far. It may not be among the greatest works of art, but it is
certainly a work of art. Bell wants to dismiss it simply because
it contains so much description, but that is no reason to dismiss
it. We would be better to critique it on its lack of arrangement,
which is what Whistler would have done. It has plenty of
arrangement, but this arrangement, though done with skill, is not
done with consummate skill. It might have been done better, and
in fact such things have been done better. Bell should have said
that its aesthetic content might have been higher. If he had, I
could not have argued with him.
Bell tells us that in
future we will go to photography to get what Frith is giving us,
and this is one of the places where that idea came from. When we
hear the same sort of thing from Robert Hughes, this is where he
learned it. The falsehood is long-lived. For Bell is wrong once
again. Frith's painting hasn't been replaced in the modern world
by photography, and can't be, since although photography can
catalog events and can even provide some degree of arrangement,
it can't do what Frith has done. Frith has composed this entire
thing, so that Frith's intention and emotion colors the entire
canvas, from corner to corner. No photograph can have that sort
of content. Photographers have a very limited control of their
canvas, compared to painters. Nature herself composes a far
larger percentage of the photo than of the painting, and a large
part of any photo has come to us with no effort or planning by
the photographer. This fact cannot be overlooked. Everything
touched by the human hand has an element of art in it, good or
bad, so that any painting is more artistic than any photo by
definition. Photos are highly interesting and partially
aesthetic, but they can in no way compete with paintings as works
of art. As far as photography has tried to replace what Frith
gave us, we have nothing but a void.
![](fildedoctor.jpg)
Bell
next dismisses Luke Filde's painting The
Doctor, which is just a smaller painting
of the Frith type. Bell works up quite a hatred of this little
painting, based on nothing as far as I can make out. Bell sees in
it a “sense of complacency in our own pitifulness and
generosity,” but I see nothing of the sort. The painting has a
bit of Hollywood in it, but I don't see the complacency. Who is
being complacent? What is not being done that should be done?
Bell may be implying that they needed better hospitals for the
poor, but this child is not being neglected. A private doctor is
always better than any hospital, as we now know, and this child
is not dying alone in the streets.
Bell says that “art
is above morals,” thinking, perhaps, that this again put him
with Whistler, but it doesn't. Whistler said that art had nothing
really to do with morals, which is true, but that doesn't imply
that the subject of a painting cannot be an explicitly emotional
one. Art can be moral and sentimental at the same time it is
being art. Art doesn't need
to be moral or sentimental, that was Whistler's point. But the
presence of morals or sentiment is not enough to disqualify a
painting from art.
You can see how Bell is an overzealous
hausfrau, sweeping the dinner out with the leavings. In his rush
to define and purify, he has swept the greater part of history's
real artifacts out the door. If we start sweeping out the Friths
and Fildes for being sentimental and descriptive, we have to
sweep out Michelangelo and Raphael, Titian and Botticelli,
Rembrandt and Rodin, Chardin and Corot. Bell usually didn't go
this far, but his theory doesn't give us a place to stop in the
housecleaning. In fact, art criticism quickly went where Bell was
pushing it, and it did dismiss all these artists and most others
as being descriptive, sentimental, moralistic, and so on. All the
art of the past was swept into a pile and dismissed as academic,
Alexandrian, aristocratic, hierarchic, or some other manufactured
and twisted term of abuse.
On page 20, Bell dismisses the
Futurists as inaesthetic, and of course we agree with him there.
The Futurists were strictly political and admitted to caring
nothing for art, so it is not surprising to find politics utterly
swamping art. I would not claim that political content is
disallowed in art, but its presence is always a danger. Politics
always wants to usurp the art, to replace it totally. Politics in
the modern world is much more predatory than morals or sentiment,
since it is much more powerful and ubiquitous. What morals used
to be, politics now is.
Bell then promotes the
primitives, telling us that because primitive art has “no
accurate representation” it is therefore free from descriptive
qualities. That does not follow. Poor description is not thereby
non-description. Beyond that, primitive art is often or always
drenched in religious symbolism. Almost no primitive art was
created strictly as a combination of forms; it was the
representation of an idea. As such, it must have been the
suggestion of
emotion, in the words of Bell. Primitive art was usually in the
form of a totem, and a totem is the exact opposite of a pure
form. Totems are drenched in ideas and rules and morals, so it is
unclear why a Filde painting should get blasted for such content
and primitive art should be sold to us as pure.
Primitive
art proves my point, not Bell's. The simplicity of primitive art
meant that the artifact must be cuing the primitive viewer to
ideas and emotions by its familiarity.
In other words, the primitive viewer already knew what the totem
meant, so its forms didn't need to be precise. It acted only as a
suggestion of what was already known. In this way, primitive art
is actually the least objective and the least aesthetic, by
Bell's own criteria. The idea has already been drummed into the
viewer by previous education, and the art acts only as a
reminder. The arrangement of the forms can therefore be very
clumsy, since the forms only have to roughly match the forms
already in the heads of the viewers. No grace or harmony is
required for this.
We can see that Bell is completely
misdefining what it is that he gets from primitive art. I have no
doubt he is receiving some strong emotion, but I strongly doubt
his report of it. It cannot be his “aesthetic emotion” he is
receiving, so it must be something else. I would suggest he is
being cued to the same cultural memory the original viewer had,
and that it is his distance from this memory that is pleasant. It
is exceedingly pleasant not to be a primitive any more. It is
exceedingly pleasant to dine on ones own elevation. It is
exceedingly pleasant not to live in a tree anymore. Nostalgia may
also be a large part of it. It may be pleasant to remember living
in a tree, provided one does not get cold and wet from the
memory.
But even here, Bell says some very true and
clever things. He says,
Formal
significance loses itself in preoccupation with exact
representation and ostentatious cunning.
True.
Just as politics can and often does usurp art, an overconcern for
precision can as well. Then he says, “A perfectly represented
form can be significant, only it is fatal to sacrifice
significance to representation.” True again. If a simplified
form creates a stronger emotional response, there is no reason to
prefer a more complex form in its stead. However, this must
depend on the situation. Not all emotions or representations call
for the simplest form. Some degree of complexity is often a
requirement of the art at hand.
Bell proves he
misunderstands the limits of his own rules when he says, “But
if a representative form has value, it is as form, not as
representation. . . . The representative element . . . is always
irrelevant.” That is false. If a representative form has value,
it is because the artist understood just how much representation
was required for the art. Since the representation is required
for the art, it cannot be of secondary value or of no value.
Again, it is the combination of representation and of formal
arrangement that is the art. In such a case, right representation
is not a by-product, an add-on, or a accident. It is one of the
fundaments.
On page 26, Bell says, “The significance of
art is unrelated to the significance of life. In this world the
emotions of life find no place. It is a world with emotions of
its own.” No. As I have already said above, this cannot be
true. Bell is blowing at his romantic worst here. I normally come
off as the romantic and idealist against these critics, but I
never say such absurd things as Bell says here. It is simply
precious to claim that art and life can be separated emotionally,
and I have never been a preciosista. Art is special, but it isn't
that special. Art may be a subworld or a superworld, but it isn't
a parallel plane, with its own rules and its own emotions. It is
precisely because Bell can say such things that he can create the
theory he does. Only by believing that art was a world all its
own could you create a theory in which everything to do with life
should be jettisoned. First you state the case and then you try
to make it true. You state that art should be separate, then you
go about separating it. You state that art is not about
representation, then you ditch all representation. You state that
art is not about description, then you ditch all description. You
state that art is not about convention, then you ditch all
convention. You are left with what all otherworlds have always
been left with: nothing.
On page 39, the second page of
chapter 2, Bell begins to wax really loathsome. He says, “Before
the late noon of the Renaissance, art was almost extinct.” He
dismisses Rembrandt to Van Dyck as “tedious portraiture.” I
said above that Bell rarely does what Greenberg does, dismissing
all the greats of Western art, but here we seeing him doing it.
All of art history was only a building up and precursor to his
“Post-Impressionist revival.” While he is showing his
ignorance, he also shows his confusion. He says, “The art of
Poussin, Claude, El Greco, Chardin, Ingres, and Renoir, to name a
few, moves us as that of Giotto and Cezanne.” These are his few
exceptions, we are to understand, but the mix is mysterious. How
do Poussin and Ingres fit into his definitions, especially? With
them we have a hard-edged description, which would appear to
conflict with his previous statements. Are Poussin and Ingres
mainly about forms? If description is superfluous, then a large
part of Ingres must be superfluous. And we find Ingres right next
to Renoir, although Renoir came from the Delacroix side of the
Ingres/Delcroix split. I can make no sense of this list. Bell is
free to prefer them, of course, but he might give us some clue as
to how they fit his theory.
As for the Giotto-Cezanne
pairing, I have never understood it. We continue to get it to
this day, but the joining has never been explained to me. Giotto
is warm and round while Cezanne is cold and square. Giotto is red
while Cezanne is blue. Giotto is almost all figures while Cezanne
is almost no figures. Giotto is big while Cezanne is little.
Giotto is ambitious while Cezanne is not. The only similarity
would appear to be a certain clumsiness in forms, although I
would say Giotto has a naïve charm where Cezanne does not. The
figures of Chardin and Corot bring this naïve charm into later
centuries, but the figures of Cezanne are never charming. Giotto
was the most skilled of his time, whereas Cezanne was among the
least skilled of the famous artists of his time. Like Goya,
Giotto seems to have succeeded brilliantly despite his natural
limitations. Cezanne has not. Even at his best, Cezanne seems
pinched. You can almost read his frustration in the strokes of
paint.
This is not to say there is nothing to like in
Cezanne. As most agree, he was at his best when he limited
himself the most, as with his fruit. But he has been terribly
oversold by the critics, and sold in false terms and similes and
comparisons. He should never be compared to Giotto, and should
never be listed above those “tedious portraitists” like
Rembrandt or Van Dyck. Van Dyck reached pinnacles of grace and
harmony Cezanne never so much as imagined. No one ever arranged
with more skill than Van Dyck. No one. If we judge art simply as
an arrangement of forms and colors, as Bell is trying to do, we
must put Van Dyck at or near the top and Cezanne far below.
Bell says that “Post-Impressionism is accused of being
a negative and destructive creed. In art no creed is healthy that
is anything else.” Absurd and exclamatory, especially
considering that Bell was the most vocal and excitable
cheerleader for Post-Impressionism that any movement ever had or
is ever likely to have. The false praise heaped upon the
Post-Impressionists by Bell and Fry couldn't have been any
louder, longer, more positive, or more overwrought. As just one
example, Bell finishes this chapter by stating,
Tradition
ordered the painter to be photographer, arcobat, archaeologist
and litterateur: Post-Impressionism invites to become an artist.
And just before that he said,
“Post-Impressionism takes its place as part of one of those
huge slopes into which we can divide the history of art and the
spiritual history of mankind.” Good lord, the spiritual history
of mankind, with Cezanne as Christ no doubt. Even Cezanne must
have wondered what he did to deserve this. If he had lived long
enough to witness it, he must have been embarrassed by such
maunderings and noodlings. Bell does everything but get down on
his knees and weep, and we must assume he did that in between
chapters.
On page 50, Bell tempers his loathsomeness
somewhat by speculating that maybe “the artist feels for
material beauty what we feel for a work of art?” Just so,
Clive. You have discovered a nut. But even then, Bell keeps
talking, to his detriment. He speculates that the artist
discovers an object of emotion in material beauty. Yes, but
again, not only that. As above, we have a mixture of the
descriptive and the non-descriptive that is appealing to the
artist. Yes, the forms cause us pleasure, but the forms are never
pure and we don't want them to be pure. It is the mixture that is
most pleasant. We do not desire that life be pure: we are most in
love with the mixture as it is, and do not wish to change it.
That is to say, the material beauty of a woman is for us both
form and description and idea. It is not form separated from
desire or description, but all three. When we paint a woman, say,
we do not wish to sift the form from the rest, because the form
by itself is bare and barren. The form is only pleasant married
to the real. In this way, figuration is like music. To create
harmony, you must have two lines of music playing off one
another. Tempering, likewise. To create a tempered scale, you
must have two notes that dissolve and resolve. It is the
resolution or harmony that is most pleasant, and purification can
only destroy it. Just so with painting.
Bell speculates
further that significant form may be the “thing in itself” or
“ultimate reality”, but again, he is just being hysterical.
There is no thing in itself, if by that one means some deeper
entity hiding behind the descriptive form. As Nietzsche told us,
what we experience is all the reality we can get. Therefore our
experience is real. Our experience already tells us much about
the thing in itself, since our experience is caused by the thing
in itself. Our experience of the thing may not be complete,
because our senses cannot soak up all possible data. But that
does not make our experience false. Nor can Bell's significant
form be taking us beyond the sensible, since how else do we
experience form and color except through our senses? Supposing it
were possible to experience life beyond the senses, we could not
do that through artistic form.
Don't misunderstand me, I
am not saying that there is nothing
behind forms, as Sartre tried to tell us. I am just saying there
is no division. There is not our experience, which is false, and
a reality behind it, which is real. Our experience is real and
something, so there is no nihilism involved. Our experience is
already a direct perception of what is real, therefore we do not
need to postulate something behind this reality. If reality were,
as Sartre said, backed by nothing, we would experience nothing.
Since we experience something, we may already conclude that
reality is something. There is no horrifying void behind
experience, just as there is no deeper reality. Experience is
incomplete, yes, but it is neither shallow nor unreal.
I
am also not saying there is no mystery in art. There is very
much, both in its viewing and its creation. But in art we are not
seeing a deeper reality, since reality is reality. The artistic
arrangement may be heightening and sharpening our senses and our
appreciation, but it is not taking us beyond the senses. A visual
art could hardly do that. An art of significant forms could
hardly be taking us beyond forms.
Bell again tempers his
loathsomeness by speaking, on page 64, “of the absolute
necessity of artistic conventions.” As I said above, Bell
rarely goes as far as Clement Greenberg and his successors. Art
critics after Bell set out to destroy all artistic conventions,
and that told us they were doing it on purpose. But on this
issue, Bell holds back. He understands that art cannot be created
from mist and good intentions. It cannot be created with airy
theories and white canvases, with a bunch of talk and promises.
Bell admits, “The effort would be feeble and the result would
be feeble. That is the danger of aestheticism for the artist.”
Yes, though we could refine that a bit and say “that is the
danger of critical formalism for the artist.” That is the
danger of basing art on critical theories.
On page 67
Bells says,
It is so easy to be
lifelike, that an attempt to be nothing more will never bring
into play the highest emotional and intellectual powers of the
artist. Just as the aesthetic problem is too vague, the
representative problem is too simple.
I
am reminded of a
similar recent claim by the portrait painter Stuart Pearson
Wright, who told us that anyone could create realist images now.
The lie is alive and longstanding, you see. We see that so many
of the current slanders and lies go back before the first World
War, many of them beginning with Bell and Fry. Bell and Fry,
neither of whom could paint beautiful figures, found the
representative problem too easy. “I can't do it, but it is
nonetheless too easy.” And now, because anyone can create
awful, monstrous figures using cameras and projectors and so on,
Wright assures us that realism is easy. But painting figures was
never easy and it isn't easy now. People who do it well are
extremely rare. Good figure painting is so rare now that most
people don't know the difference between bad figuration and good.
No one alive now is as good as Van Dyck. No one alive now is even
as good as Sargent or Bouguereau. So how is it “too easy”?
Bell is partially correct when he says that an attempt to
be nothing more than lifelike is bound to fail, but the best
realists never limited themselves to being lifelike. Good realism
has never been only that. Only a few beginners and hacks limit
themselves to trying to be lifelike, so that Bell's statement is
misdirection. We do not judge art on what a few beginners and
hacks are doing. And besides, Bell has buried his little truth in
a big lie. Telling us that realism is easy is just a hamhanded
attack on skill. It is a poorly disguised bit of resentment. It
is the attempt to replace the skilled by the unskilled, by
telling us that skill is easy or universal and therefore unworthy
of our admiration.
I have only counter-critiqued Bell's
first three chapters, but I think I will save the rest for later.
Sorting through the messes that critics make is sort of like
slogging through a room full of tangled coat hangers, or trying
to unwind a mile long extension cord, tied into knots. It isn't
hard work, that is to say, but it is tiresome. And it is
debilitating to an ordered mind, having to exist among the pages
of a disordered one, even for an hour or two. I begin to get
fluttery and irritable, and I long to hide away in Bach for an
equal amount of time, to reorder the recent chaos. Or to climb
inside a Titian portrait and breathe deep the perfection.
*Whistler mocked Ruskin's Slade
professorship of 1869, and would have mocked even more viciously
the Slade professorship of Fry in 1933. **This is a must read,
and you can find it here.
If this paper was useful to you
in any way, please consider donating a dollar (or more) to the
SAVE THE ARTISTS FOUNDATION. This will allow me to continue
writing these "unpublishable" things. Don't be confused
by paying Melisa Smith--that is just one of my many noms de
plume. If you are a Paypal user, there is no fee; so it might
be worth your while to become one. Otherwise they will rob us 33
cents for each transaction.
|