PRINT October 1992

The Matisse System

One of the most striking things in any big retrospective of Henri Matisse is how diverse his art is and yet, within this diversity, how a fundamental core unifies more than half his production and the important half at that. Indeed, it is probably universally admitted by now that whatever the charms of the so-called “Nice period,” of 1917–30, the great Matisse is that of the years 1905–17 and then 1931 to his death; and I don’t think anyone would deny that the 1911 Atelier rouge (Red studio) has more in common with the 1948 Intérieur rouge (Red interior), for example, than either work has with any of the odalisque canvases of the 1920s.

But what exactly is this core (I call it the “Matisse system”)? How can we define it? The usual answer is that it originates in Fauvism, and can be characterized as a certain “liberation of color.” But this does not satisfy me. For one thing, various others among the Fauve painters of 1905 were much more “liberated” in their color than Matisse (think of some of the André Derain or Maurice de Vlaminck works of the time), yet none of them could sustain the chromatic tension. (Fauvism is a movement that died within a season.) Second, I would argue that the Matisse system owes very little to Fauvism, and even that it was invented in order to break away from the jejune outbursts of Fauve color.

In exploring this system we have at our disposal the extraordinary guidance of Matisse’s own statements. I consider him one of the century’s best writers on art, and certainly the best commentator, by far, on his own work; my hypotheses on his painting are entirely dependent on what he himself endlessly wrote and said.

To sum up an argument I have developed elsewhere at length, I would say that at the very beginning of 1906, Matisse discovered that “one cm² of any blue is not as blue as a square meter of the same blue,” or, again, that “the quantity of color was its quality”—an equation to which he would return in almost all his statements on color.1 The discovery that color relations are above all surface-quantity relations was determining for almost his entire career (the large exception being the years 1917-30). It is important to note the way this discovery was made: through drawing, or, more precisely, while he was working on three black-and-white woodcuts often labeled “primitivist” for the crudity of their line.

Also important, of course, is when the discovery was made—very soon after the famous Fauve Salon, as a definitive farewell to Paul Signac and to Divisionism, a method that Fauvism had pushed to the limit but not at all overthrown. But I don’t want to dwell here on the break with Divisionism, which would require a long historical inquiry into not only Matisse’s two ill-thought-out pointillist adventures (in 1898-99 and 1904-5), but also his relationship at the time with the art of Van Gogh, Gauguin, and Cézanne. I’ll just note the talismanic status of the 1905-6 painting Bonheur de vivre, the first painting in which Matisse tests how the value and saturation of a given color change according to the quantity of surface it covers. As such, this is the work that marks the end of Fauvism as bar as he is concerned, and the beginning of his “system.”

Much more important for me is the fact that Matisse’s discovery of the equation between quantity and quality in color was made through drawing. As with most inventions, this discovery followed from the elimination of certain key parameters: color “proper” had to be absent for Matisse to discover not only that black and white are colors, but also that what makes them act as colors (modulation through the quantity of surface allowed them) plays an essential role in the behavior of all colors. Matisse’s statements on his discovery abound throughout his life. He is especially eloquent when speaking of his ink drawings of the 1930s: “In spite of the absence of shadows or halftones expressed by hatching, I do not renounce the play of values or modulations. I modulate with variations in the weight of line, and above all with the areas it delimits on the white paper.”2 Often, as well, he laments not having obtained in his painting the quality of modulation through surface quantity that he has mastered in drawing. But it is particularly striking to see him returning again and again, from the 1908 “Notes of a Painter” on, to the graphic principle of surface division as a key to his color.

The numerous statements about the Danse of 1930-33, a work created for the Barnes Foundation in Philadelphia, are exemplary. The painting almost constitutes an experimental verification of the Matisse system. Having completed a first version of his “architectural painting,” as he called it, Matisse found out that he had worked with the wrong measurements; but he could not simply enlarge his composition. To maintain the original color effect within a larger overall surface, he had to modify the picture entirely. This chance experiment was a major cause of his return to his system after the parenthesis of the Nice period.

All this might seem of little consequence, but I think it is essential for an understanding of Matisse’s art and of his position in the history of painting. First, it is the quantity/quality equation that holds the Matisse system together. It governs Matisse’s allover conception of the surface; the relationship of his compositions to the frame; his unique sense of scale, and the total disappearance in his works of the rendition of light by a play on values and traditional modeling. Second, Matisse invented a new form of drawing that entirely abolished the traditional distinction between drawing and color. (He often said that his black-and-white drawings were colorful.) The concept is that of surface division, and it immediately affected not only his drawing but his painting (that is, in the canvases that partake of the system I’m trying to describe here).

For me, this invention represents a gigantic leap in the history of painting, an extremely serious blow against the dualistic tradition of Western thought (spirit and matter, idea and form, drawing and color). It was a blow that was only begun to be understood later on by Jackson Pollock and Barnett Newman. It represents, in short, a fundamental attack on the idealist division between conception and realization: if a color has no identity before it covers a certain quantity of surface, it cannot be thought up beforehand.

I’ll test my hypothesis with immense pleasure at the Museum of Modern Art show, and especially regarding a period in Matisse’s art, perhaps my preferred period, that remains largely under-studied—the late ’40s, immediately before he began to make paper cutouts exclusively. The two to three years during which he painted Intérieur au rideau égyptien (Interior with Egyptian curtain, 1948), Le fauteuil rocaille (The rocaille armchair, 1946), Le silence habité des maisons (The silence living in houses, 1946), and many other canvases of the same quality are for me the most accomplished of his career. This was also (and not by chance, I think) a time when his drawings, especially his brushed-ink drawings, were at their most monumental—at their least distinguishable, that is, from his paintings. I like the fact that it was long after he accomplished his “revolutionary” work (usually attributed to 1911-12) that Matisse gave us his full measure. This provides us with a fabulous warning against the historicist mode of judgment followed by many of us art historians and critics, according to which what’s first, chronologically speaking, should necessarily be deemed more important. In short, I can hardly wait for the show.

Yve-Alain Bois teaches the history of modern art at Harvard University. His most recent book is Painting as Model.


1. Henri Matisse, quoted in Louis Aragon, Henri Matisse: A Novel, trans. Jean Stuart, New York: Harcourt Brace Jovanovich, 1971, 2:308; and Matisse, “Statements to Tériade,” 1929-30, in Jack D. Flam, ed., Matisse on Art, New York: Phaidon Press, 1973, p. 59. See also Yve-Alain Bois, “Matisse and ‘Arche-drawing,’” Painting as Model, Cambridge, Mass., and London; The MIT Press (an October Book), 1990, pp. 3-63.

2. Matisse, “Notes of a Painter on his Drawing,” 1939, in Flam, p. 82.