In the standard histories of modern art, Fauvism is considered the first major twentieth-century movement. Around 1905, the story goes, following hard upon the innovations of the Post-impressionists, a group of young artists led by Henri Matisse began to revolutionize European painting and make it really modern by using brash color, bold brushwork, and loose drawing. The journalist Louis Vauxcelles, a somewhat skittish supporter of avant-garde painting, is supposed to have given the movement its name in his review of the 1905 Salon d’Automne, where he referred to these artists as fauves, or “wild beasts.”1

In fact, Vauxcelles used the word fauves twice in his Salon review, in two different ways. The first time, the word was used to characterize the anticipated reaction of the philistines and academics who would presumably pounce on Matisse’s painting like wild beasts. Vauxcelles wrote that Matisse was courageous, “because his entry—and he knows it—will have the fate of a Christian virgin delivered to the wild beasts [fauves] in the arena.” Later in the same review, the word fauves was used in the opposite way, to characterize the effect created by the paintings of Matisse and his friends. Writing about two rather traditional sculptures by Albert Marque, Vauxcelles noted that they were surprising to find “in the midst of the orgy of pure color: Donatello among the wild beasts [fauves].” Vauxcelles had made a similar remark a few days earlier at a preview of the Salon, when upon seeing Marque’s sculptures alongside the brightly colored paintings, he had said: “Tiens, Donatello au milieu des fauves!” People had been amused and Vauxcelles was not about to throw away a good line.

But contrary to common belief, the name did not catch on right away. It was not until almost two years later, in Vauxcelles’s review of the 1907 Salon des Indépendants, that the epithet fauve was used in a direct way to describe Matisse and his colleagues:

M. Matisse, chief fauve; M. Derain, sub-chief fauve; Messers Othon Friesz and Dufy, following fauves; M. Girieud, indecisive fauve….2

The term came into common use toward the end of the year, but by then the painters associated with it were working in a somewhat different manner, so in a sense the movement was not named until after it was over.

It is no mere coincidence that the origin of the word “cubism” is also generally (though mistakenly) attributed to Vauxcelles, and that it, too, came into general use after there had been a radical transformation in the style that it was supposed to describe.3 Set side by side, Fauvism and Cubism make an appealing package as the first two major twentieth-century painting styles, evoking as they do the Nietzschean polarity between the Dionysian and the Apollonian that was so attractive at the beginning of the century. But in fact, they are both imprecise stylistic terms which I believe tend to falsify the situation of French painting during the first decade of the century.

The word fauve, for example, tells us much more about early twentieth-century attitudes toward art that appeared to violate social and ethical norms than it does about the painting style that it is supposed to describe. Its continued use is a matter of habit and convenience, as a kind of shorthand for referring to the bright and vigorously brushed paintings that Matisse, Derain, Vlaminck, and a number of other artists made around 1905–1906. In fact, it is difficult to characterize these paintings as a group other than to say that they frequently use primary hues organized around a polarity of red and green, and that they usually seem to have been done directly from nature, following the impressionist practice of direct observation and on-the-spot improvisation. What is usually characterized as Fauve painting actually includes a fairly wide variety of styles, sometimes within the work of a single artist. The pictures involved range from thinly painted canvases that use jerky brushstroke and broken color to canvases that are more densely painted and composed largely in broad planes.

The painters classified as Fauve and those later called Cubists tended to overlap. Georges Braque, who started out as a Fauve, became one of the inventors of Cubism. Derain and Dufy also later became associated with Cubism. And as early as 1905, many of the underlying concerns of Matisse’s painting were similar to those later associated with the Cubists.

For Matisse, the most important of these concerns was dealing with the radical notions of form and space that had been set forth by Cézanne, and that had both troubled and inspired him since the turn of the century. Matisse was particularly interested in the way that Cézanne had redefined the pictorial field and made his brush marks act as abstract forces within that field, able to traverse and conflate the boundaries between objects and the spaces around them. This was not only a move away from descriptive painting, but also away from representation, toward abstraction. Matisse’s 1905–1906 paintings reveal a central concern with opening up the pictorial field to make it virtually independent of the representation that it carries, and with inventing a pictorial language that could be used almost independently of what was supposed to be depicted. This phase of Matisse’s painting was characterized by a marked ambivalence between conceiving of the canvas as a kind of window that presented a view and treating it as an opaque pictorial field that was more like a wall.



Although there has been a certain amount of disagreement among scholars about precisely which painters belong to the Fauve group, the one issue on which virtually everyone is agreed—and on which virtually everyone has been in agreement since 1905—is that Matisse was the leader of the group and the most important artist in it. It is thus all the more surprising that Matisse is so poorly represented in the exhibition of Fauve painting currently at the Metropolitan Museum of Art, which is one of the largest exhibitions of its kind ever mounted.

I first saw “The Fauve Landscape” in Los Angeles, where it was bigger and somewhat more diffusely arranged than the exhibition now on view in New York, and where there were proportionately even fewer Matisses. But despite the differences, the overall conception of the show has remained the same. The paintings are arranged primarily according to where they were painted—Paris, the suburbs of Paris, Saint-Tropez and environs, Collioure, Normandy, and so forth. (In fact, a certain amount of juggling has been done. For instance, a Vlaminck painted at Chatou is in the Paris gallery, a Derain view of Martigues is shown among Braque’s and Dufy’s Antwerp pictures, and the grouping in the final gallery is basically chronological.) The catalog essays are similarly arranged, and the show’s organizers evidently believe that geography had a crucial role in the development of Fauve painting. Moreover, if there are surprisingly few Matisses, there is a relative abundance of works by minor artists, such as Dufy, Friesz, and Manguin.

The presence of so many minor—indeed mediocre—paintings has a curious effect on the viewer. On the one hand it seems to evoke more fully than ever before a cross section of French painting at a particularly interesting moment. To this end, the curator, Judi Freeman, has done a superb job of seeking out obscure pictures and persuading their owners to lend them. (William Lieberman has added some stunning Matisses in New York.) But at the same time, the variable and sometimes low quality of the work in the show poses some serious questions about the way the history of this moment in French painting is conceived of and written. Is it better to try to give an idea of the full range of what was produced in a given style, or is it preferable to be more selective and to show only the strongest or most representative work? This exhibition makes it clear that there are no easy answers to this question, which in fact raises more questions. For example, is the strongest work necessarily the most representative?

The quantity and variety of pictures in this exhibition are fascinating (at least to specialists), but also fatiguing (even to specialists). They also have the curious effect of confusing rather than clarifying an important aspect of the history of French painting at the beginning of the century—if one understands the writing of history to involve a selective synthesis of the past rather than a somewhat indiscriminate, though comprehensive, presentation of it. (And if one is really going to try to be comprehensive, then why not include artists like Girieud and Puy in the exhibition as well as the catalog?)

Geography has been substituted for chronology, and in the catalog it has even been translated into an aspect of ideology. Fauve painting is seen here as part of certain social concerns, which seem to include patriotism, regionalism, tourism, and even colonialism, although the exact nature of those concerns, or how they are actually expressed in the paintings, is never made quite clear. The main point seems to be that Fauve painting can best be understood by studying it in relation to sites, maps, postcards, photographs of landscapes, and anecdotes about the painters’ work habits.

Among the most instructive comparisons in the exhibition are those between pictures of the same or similar motifs painted by different artists of unequal gifts, as in the pictures that Braque and Friesz painted virtually side by side at La Ciotat in 1907. The structural weaknesses and uncertain execution of the Friesz paintings make us appreciate Braque’s masterly treatment of these motifs all the more. Seeing such works together also makes us realize how subjective the vision behind such paintings was, even though they were painted directly from nature. And it is precisely this subjectivity that makes us question the geographical arrangement of the show. If locale is supposed to have been as important as the exhibition and the catalog would lead us to believe, then why do the painting styles seem to have so little direct relationship to specific sites? Style seems to be something that the artists brought along with them to the different places they painted, rather than something that they discovered on the spot, or necessarily felt had to be different in different places. The various manners in which Derain worked during this period, for example, seem to depend more on the artistic models he adopted—Seurat, van Gogh, or Gauguin—than on where he was painting.


The emphasis on place also leads to some basic misunderstandings about the nature of Fauve painting. In James Herbert’s essay in The Fauve Landscape, he characterizes Matisse’s and Derain’s paintings in the south as “picturesque” landscapes, which he defines as “quite literally a view of a section of the countryside that conforms to the compositional standards set by previous landscape paintings.”

But in fact, Fauve paintings are for the most part anti-picturesque—even though they frequently represent views that could be considered as picturesque or scenic. The central notion here is that of the “view.” If Matisse’s landscape paintings, for example, presented transparent views (like those on postcards), rather than images in which the view is modified and even undermined by the pictorial field, then one might accurately describe them as being picturesque. But in fact, Matisse paints his picturesque subjects in a very unpicturesque way. (This is also true of early Cubist landscapes, such as Braque’s 1908 L’Estaque paintings, which also include obviously picturesque devices—such as the use of framing trees in the foreground—but employ them in an aggressively antipicturesque fashion.) Such paintings, in which real light, space, and atmosphere have been suppressed along with a sense of passing time, are in fact almost parodies of picturesque painting.

The exhibition and its accompanying catalog seem to be based on three underlying assumptions: that there is such a thing as a Fauve style, that this style is best expressed in landscape painting, and that it lasted from about 1904 through 1908. But each of these assumptions is questionable. And in each instance, the artist who fits least well into the category of Fauve painting as it is presented here is the supposed leader of the group, Matisse. No wonder there are so few Matisses in the exhibition, and no wonder so many of the ones that are included are of figures or interiors rather than landscapes.

That there is no such thing as a single Fauve style seems self-evident, and in her introduction to the catalog Judi Freeman has some very good things to say about the use of Fauve as a noun (to designate the painters involved), and as an adjective (to classify the pictures). “Fauvist” and “Fauvism,” as she rightly points out, are best avoided, since the artists involved never organized a real movement. But although she acknowledges that each of these artists had his own style (or styles), she gives the impression that they all were working toward similar goals.

This, however, does not seem to be the case. Vlaminck’s paintings, for example, appear to be part of an enterprise that is very different from Derain’s, which in turn was quite different from Braque’s. In Vlaminck’s Fauve canvases, painting and drawing are usually treated as separate processes; one has the feeling that the composition was drawn first and then colored in. Very little real interaction takes place between color, design, and space, and as a result the use of each of these elements tends to be rigid. The different areas in his paintings are frequently separated by drawn contours, which create a kind of cloisonné effect, and you often feel that his “boldness” amounts to little more than going heavy on the reds. Derain comes across as a distinctly stronger painter, whose drawing, color, and spatial construction are much more subtle, and supple. His compositions have a more painterly structure, and his use of the picture space is decidedly more dynamic.

But most important is the profound difference between Matisse’s 1905–1906 paintings and those of all the other artists who are categorized as Fauves, including Derain. Derain’s paintings, though much less formulaic than Vlaminck’s, are considerably more schematic than Matisse’s. These differences are particularly apparent if one compares their very different renderings of similar views of Collioure, which are hung next to each other at the Metropolitan Museum.

The Matisse Landscape at Collioure is not only more freely painted—and less “finished” looking—than Derain’s Landscape at Collioure, but it also uses the unpainted white canvas in a different way, and the difference is crucial. In the Derain painting, the whiteness functions very much as it would in a Neo–impressionist painting, creating a background of effulgent light against which the rhythmically spaced strokes of color are set. In the Matisse painting, by contrast, the white spaces are much more dynamic and active. They seem to interact with the irregularly placed brushstrokes and function as abstract plastic forces that have almost as much presence as the marks of color. This fluid conception of space grows directly out of Matisse’s understanding of Cézanne, who also used the white spaces of the canvas in a strong, structural way.

Compared with the Derain, the Matisse landscape also seems to be remarkably free of a schematic structure. Its extraordinary fluidity introduces chance and arbitrariness into its composition. Even its colors are rather arbitrary, and only indirectly refer back to nature. Such a painting is much less descriptive than one would traditionally expect a landscape painting to be, and seems to deny rather than affirm a specific time and place.

In the exhibition and its catalog, landscape painting is held to be the key to Fauve painting. In a limited sense, I suppose it may be considered as such, more or less by default, simply because landscape imagery will allow for a great deal of representational distortion without particularly upsetting anyone. Color a tree bright red, or move a house or an outcropping of rocks a couple of yards, and no one will take you to task for it. Move someone’s nose a couple of centimeters, though, or color it vivid green, and you will be attacked. This is a lesson that Matisse learned at the 1905 Salon d’Automne, where the painting that really set people’s teeth on edge was not a landscape but the portrait of his wife known as The Woman with the Hat. (Even Leo Stein, who bought it, called it the “nastiest smear of paint I had ever seen.”)

No doubt the countryside itself was important to the Fauve painters. But landscape also provided a convenient subject from a strictly pictorial stand-point. Landscapes are full of irregular forms and ambiguous spatial relationships, and lend themselves more easily than do figures to being seen as part of an abstract pictorial field. It is also possible that some of the Fauve artists preferred landscape because it was more suited to their modest talents. But in any case, by concentrating exclusively on landscape this exhibition eliminates some of the best and most inventive paintings of the period and fails to deal with some of the most important issues those paintings raise.

Ms. Freeman suggests that “the burning question for the Fauves was what to paint and how.” It seems to me, however, that the “what” was not landscape, which they were already painting with some success, but the human figure, which gave most of them trouble. Gertrude Stein was not far off the mark when she wrote, “If you do not solve your painting problem in painting human beings you do not solve it at all.”4 And for French painting at the beginning of the century, the ultimate painting problem was how to represent the human figure in a nearly abstract way and still make it convincing. This, moreover, was a problem that only Matisse, of all the Fauve painters, really made a sustained attempt to solve. When Ms. Freeman says that landscape is the dominant motif in Fauve painting, but that later, in 1907–1908, “Derain and Matisse eventually adopted the figure as central to all nature-based imagery,” she misses a central point. Matisse had already more or less abandoned landscape painting per se by the end of 1905, after which most of his paintings are of people and of interiors and still lifes—subjects which emphasize discrete and solid forms in a clearly delimited space. Derain, who like Braque was much less comfortable painting the human figure at this time, would take a while longer before he concentrated on figurative imagery. Vlaminck was, and remained, as John Klein remarks, “a terrible painter of the human figure”—a judgment borne out by his terrible 1908 Bathers in the last gallery of the exhibition.

A similar progression from landscape to figure and still life subjects would occur shortly afterward in Braque’s and Picasso’s early “cubist” painting. Many of their initial ideas were worked out with the relatively open imagery of landscapes; but the most difficult work was accomplished in the figurative imagery and still lifes, where they too had to confront the problems posed by solid form set within a more clearly defined space.

The date assigned to the end of Fauve painting is equally questionable. According to Ms. Freeman, Fauve painting ends in 1908, which she finds to be a “far more accurate” date than 1907, which is favored “by many historians.” It seems to me, however, that if one can speak cogently of something called “Fauve” painting, as generally characterized by bright color and vigorous brushstroke, then it ends no later than the spring of 1907—around the same time that Vauxcelles’s name for it began to be used. (I refer, of course, to Fauve painting as a historical phenomenon; for if you look in the windows of certain art galleries on upper Madison Avenue, you will see that versions of Fauve painting are still being made today.)

For everyone but Matisse (who is such a special case that I am tempted to say that he shouldn’t be considered a “Fauve” painter at all), Fauve painting came to a rather abrupt end after the “discovery” of Cézanne that followed the important exhibitions of Cézanne’s work in the fall of 1906 and the spring of 1907. Even Matisse, who had already been productively struggling with Cézanne for years, was profoundly affected, as can be seen in his famous Blue Nude, painted at the beginning of 1907.

The influence of Cézanne is clearly apparent in the last gallery of “The Fauve Landscape” (where the most impressive painting is Matisse’s Bathers with a Turtle, a figure composition). Derain’s Cézanne-like landscapes of 1907 are seen here along with Braque’s Cézannesque Trees at L’Estaque of 1908, and Matisse’s austerely decorative View of Collioure, also of 1908—which embodies yet another, more indirect, response to Cézanne. Taken together, these paintings make us realize to what degree labels such as Fauvism and Cubism distort the art they are supposed to help to describe.

In 1907, Fauve painting had only recently been named, and what we now call Cubist painting had not yet been given a name. The pictures now subsumed by these categories were then perceived not as mutually exclusive but as different and complementary responses to the issues raised by the art of Cézanne.

Virtually all advanced painting at that time attempted, in one way or another, to reformulate the nature and meaning of what were then considered the limits of known pictorial construction. Seen from our own perspective, it seems that between 1905 and 1910 the paintings of several French artists had in common a deep concern with the autonomy of the picture as distinct from what the picture represented. The various ways in which this reformulation of pictorial space was attempted overlapped and involved some of the same artists, who were preoccupied with many of the same issues. Unfortunately our standard terminologies obscure the ways their different approaches coincided and merged with one another.

They also obscure crucial differences between the ways that different artists used apparently similar styles. For Matisse, whose painting during the Fauve period was actually quite varied, “fauvism” represented an attitude rather than a style, a willingness to take risks and a desire to explore what he later called “the essential principles that made human language.” For Derain and Vlaminck, “fauvism” was more of a formula, a new set of rules with which to break the old ones. Similar differences would later develop between Picasso and Braque and the so-called “Salon Cubists.”

This exhibition, which is both more and less than an exhibition of Fauve painting, may help us to reevaluate and reinterpret the nature of advanced painting during the first decade of our century by inviting us to rethink some of the basic categories into which it is still being forced to fit.

This Issue

April 25, 1991