NCAW Autumn 03 | Dario Gamboni on Gauguin's Genesis of a Picture

Paul Gauguin's Genesis of a Picture: A Painter's Manifesto and Self-Analysis
by Dario Gamboni

Fig. 1 Paul Gauguin, Mana'o tupapa'u, 1892. Oil on burlap mounted on canvas. Albright-Knox Art Gallery, Buffalo

Introduction
There is an element of contradiction in the art historian's approach to interpreting the creative process. While artistic creativity is seen as stemming from the psychology of the creator, there is a tendency to deny the artist possession of direct or rational access to its understanding. Both attitudes have their roots in romantic art theory. These attitudes imply, on the one hand, that the artist's psyche is the locus of creation, and, on the other, that intuition and the unconscious play the major roles. Such ideas remained prevalent in the twentieth century, as attested, for example, by the famous talk given in April 1957 by Marcel Duchamp under the title "The Creative Act," which defined the artist as a "mediumistic being."¹

This element of contradiction explains the frequently ambivalent treatment of artists' accounts of their own creation: although we recognize that they possess a privileged and even unique position, we tend to dismiss their explanations as "rationalizations" and to prefer independent material evidence of the genesis of their works, such as studies or traces of earlier stages. This preference can often be justified with solid arguments, but it also happens to satisfy the general tendency of modern critical thought to "discover" and to "unveil," to oppose the hidden, the latent, and the repressed to the explicit and the manifest. Thus, our suspicion itself must be suspect, and it is necessary to examine in what regard, and under what conditions, use can be made of the artists' accounts of the origins of their works.

Genesis of a Picture
A good case study of the problematics of using an artist's own writings to interpret his or her work is Paul Gauguin's written account of his Mana'o tupapa'u of 1892 (Albright-Knox Art Gallery, Buffalo; fig. 1). The painting is one of Gauguin's most famous works and it had a special importance for him, as demonstrated by the high price he asked for it, its numerous variations, and its inclusion in the Self-Portrait with a Hat painted in the winter of 1893–94 in Paris (Musée d'Orsay, Paris).² One can argue that the painting's fame rests in part on Gauguin's remarks about it. The title he chose for these remarks, Genèse d'un tableau (genesis of a picture), indicates that he intended them to possess a paradigmatic value. By extension, the painting itself can be seen as programmatic, and close examination of it supports this claim. Taken together, Mana'o tupapa'u and his commentary Genesis of a Picture represent Gauguin's main contribution to the collective reflection on the creative process.

Gauguin's explanation of the painting exists in five versions. He first discussed it in two letters written from Tahiti in December 1892 to his wife Mette and to his friend Daniel de Monfreid; then, in a more developed form and under the title Genèse d'un tableau—which I will use here for the whole series of commentaries—in his manuscript Cahier pour Aline in 1893; and, back in France in 1893–94, in the manuscript Noa Noa, written alone and finally revised with the help of the poet and critic Charles Morice.³ These versions display many differences but are also in general agreement. One can summarize their argument as follows. The painting began as "a study of a Polynesian nude" shown in a "rather daring position, quite naked on a bed." The painter, however, wanted to avoid any charge of indecency and to render the "Kanaka spirit, its character, its tradition." For this purpose, he chose certain accessories (such as the pareu on the bed) and certain colors (yellow and violet), as well as a theme that could justify the motif. Since allusions to sex had to be avoided, the reason for the woman's position must be fear, experienced in the night, and therefore fear of the Tupapa'u, the spirit of the dead.

Gauguin concluded his reconstruction with a descriptive and interpretive summary: "To recapitulate: Musical part—undulating horizontal lines—harmonies in orange and blue linked by yellows and violets, from which they derive. The light and greenish sparks. Literary part—the spirit of a living girl linked with the spirit of Death. Night and day."⁴ His analysis distinguishes between several levels or components of the work, especially the "musical" and the "literary" parts. It also takes the form of a plot in which the making of the painting unfolds like a story ("so then I do this…"), emphasizing the rational logic of the process and the extent of the painter's control. The latest version, in Noa noa, adds to this account an anecdotal origin: Gauguin told how returning once at night from Papeete to this house, he found his "vahine" Teha'amana lying on their bed in the darkness, terrified by the night and the spirits. I shall come back later to this story.

The reception of Gauguin's explanations of the painting has varied greatly. While many authors, including Robert Goldwater, have seen them as truthful and convincing, others like Robert Rey have been more critical.⁵ In his dissertation on Gauguin's paintings of the first voyage to Tahiti, published in 1977, Richard S. Field argued that Gauguin had manipulated his reconstruction of the creative process in order to "reveal something about art rather than supply a key to a single painting," and that by "wanting us to believe that the entire meaning evolved through the stimulus of the evolving forms," he was "actually trying to remove the artist entirely from the mortal sphere—to raise the artistic process to the level of divine intuition and guidance."⁶

Audience and Function
In examining Gauguin's intentions in explaining the origin of his painting, Field rightly raised the question of his audience, which may help us understand the functions and meanings at stake. For whom was Gauguin writing? The letter to Monfreid is supposedly meant "for [him] only," but the letter to his wife reveals that the artist intended her to share it with others. Both letters refer to the Tahitian paintings that Gauguin had sent home and that would be exhibited in Copenhagen and Brussels in 1893–94; the letters also contain a French translation of the works' titles, which the general public would only encounter in the original (and approximate) Tahitian.⁷ Gauguin wrote to Mette about the group of paintings: "Many of the pictures, of course, will be incomprehensible and you will have something to amuse you. To enable you to understand, I proceed to explain the most questionable and the ones I would keep or sell dear. [There follows the explanation about Mana'o tupapa'u.] Here endeth the little sermon, which will arm you against the critics when they bombard you with their malicious questions."⁸ By way of his wife (and to a lesser extent of his friend), Gauguin thus addressed the critics at large. His Genesis of a Picture must therefore be understood as an intervention in the collective process of interpretation and within the context of the interaction between artists and critics.

The last third of the nineteenth century saw the rise and the establishment of what Harrison and Cynthia White have called "the dealer-critic system," in which the consecration and diffusion of artists and artworks depended on the free market and the press rather than on the formerly dominant state-controlled institutions, such as, in France, the Académie des Beaux-Arts, the École des Beaux-Arts, and the jury of the Salon.⁹ Dealers, critics, and other mediators thereby gained an increasing power over artists who sometimes protested verbally or visually. In 1896, two years after the presentation of Mana'o tupapa'u in the annual exhibition of the private association La Libre Esthétique in Brussels, James Ensor thus depicted in the painting The Dangerous Cooks (private collection) two key members of the Belgian art world—the lawyer, collector, and critic Edmond Picard and the secretary of the association Octave Maus—preparing artists like fish or meat before serving them to the critics.

Gauguin himself preferred the raw to the cooked, but he was also keenly aware of the new situation and tried to make the best of it, as demonstrated by his calculated and at times manipulative relationships with writers and critics such as Albert Aurier, Octave Mirbeau or August Strindberg.¹⁰ However, he came to experience and express a deep frustration at finding that the artists' new dependence upon the professional commentators could be even greater than had been the one upon the official institutions. In a piece of "counter-criticism" entitled Racontars de rapin, written at the end of his life in 1903, he would describe the evolution he had witnessed in his lifetime as a passage from the "regime of the sword" to the "regime of the man of letters."¹¹

Conception of the Creative Process
Seen in this context, Genesis of a Picture represents an intervention in the market of interpretations and in the power competition between artists and mediators. By giving as an "explanation" of his picture an account of its "genesis," which he alone had witnessed, Gauguin clearly claimed the artist's privileged access to the content and meaning of his work. However, there is no reason to reduce the formulation and communication of this account to a strategic move only. It also permitted Gauguin to defend a certain conception of the creative process and to require from the spectator an approach to art that corresponds and does justice to it. In the various versions of Genèse d'un tableau, the subject matter of the painting, its "literary part," does not exist prior to the work but emerges progressively from a sort of dialectical pas de deux with the formal choices, and they lead together to a symbolical level that can be neither dissociated from the painting as such nor simply turned into words.

There are many antecedents to this conception, especially in the writings of Eugène Delacroix and Charles Baudelaire. Closer chronologically to Gauguin, James Abbott McNeill Whistler had argued in his lecture Ten O'Clock, published in 1885 and three years later in Stéphane Mallarmé's French translation, that "the unattached writer [who had] become the middleman in this matter of art [had] brought the most complete misunderstanding as to the aim of the picture" because "for him a picture is more or less a hieroglyph or symbol of story."¹² However, Whistler's anti-"literary" stance tended to reduce the "aim of the picture" to formal harmony whereas Gauguin's position is much more complex. Odilon Redon would come closer to it when, in 1898, in an exchange of letters with the critic André Mellerio, he would criticize the notion of concept préalable (preliminary conception) and define it as only a point of departure. Redon described the creative process only in generic terms, as a sort of serendipitous flânerie guided by fantasy, and he refused to give specific explanations on the ground that "nothing is done in art by will alone" and "all is done by a docile submission to the coming of the 'unconscious'."¹³ But Gauguin resorted to explanations in order to dismiss similarly the notion that a work of art is the re-translatable visualization of a preconceived idea. In 1899, he thus felt compelled to refute the opinion of the critic André Fontainas, who had written that without the title inscribed on the canvas, it would have been impossible to discover the "meaning of the allegory" in Where do we come from? What are we? Where are we going? (1897–98; Museum of Fine Arts, Boston). Gauguin responded that the painting was not an allegory but a "musical poem" and that the inscription did not reveal a preliminary program but was a reflection added after the end of the creative process, a "signature" rather than a title.¹⁴ As Whistler had done before him, Gauguin attributed Fontainas's mistake to the literary bias of writers, and wrote that he had attempted to "translate his dream in a suggestive decoration without resorting to literary means."¹⁵

Fig. 2 David Pierre Giottin Humbert de Superville, Allegory, 1801. Engraving. Printroom, University of Leiden

Conception of Aesthetic Communication
"Translation" means here expression, and the "literary means" are rejected in favor of artistic or (more specifically) painterly means proper. The term "suggestive" is also revealing: in a letter of August 1901 to Monfreid, Gauguin would write that "in painting one should look for suggestion rather than description, just as in music."¹⁶ The description of the creative process in Genesis of a Picture is related to a conception of aesthetic communication in which the expressive capacities of the formal means are exploited in conjunction with the subject matter and semantic associations of the objects depicted or "suggested."¹⁷ To his wife, Gauguin wrote, for example, that he had wanted to "explain" the woman's fear in Mana'o tupapa'u with "as little as possible of the antiquated literary means" and used instead the colors violet, blue, orange-yellow, and greenish yellow.¹⁸ In his article of 1892 on the "symbolist" painters, in which he hailed Gauguin for the second time as the "initiator" of the movement, Albert Aurier had written that since the aim of art was no more "the direct and immediate reproduction of an object, all the elements of the pictorial language, lines, planes, shadows, lights, colours [became] abstract elements that can be combined, attenuated, exaggerated, distorted, according to their own expressive mode, in order to achieve the overall goal of the work, that is the expression of an idea, a dream, a thought."¹⁹ Among the explorers of this "language," Aurier cited Leonardo da Vinci, Charles Henry, and the Dutch neoclassical artist Humbert de Superville. It may be significant that the latter's Allegory (fig. 2) has been credited—without evidence of a direct link—as a compositional and iconographical source for Mana'o tupapa'u, since his Unconditional Signs in Art (1827–32), a book known in France through its transmission by Charles Blanc's Grammaire des arts du dessin (1867), was a major inspiration for artists seeking to base communication on visual properties.²⁰

Communication implies reception. Gauguin's Genesis of a Picture constantly—albeit implicitly—points to the impact that a work was intended to have upon the spectator. It thus belongs to the tradition of what is called in German Wirkungsästhetik (aesthetics of the effect) and can be compared with "The Philosophy of Composition," an essay of 1846 in which Edgar Allan Poe claimed to explain the genesis of his famous poem The Raven (1845).²¹ Poe's account proposes an even more rigorous causality than Gauguin's and is quite explicit about its intention: "It is my design to render it manifest that no one point in [the poem's] composition is referable either to accident or intuition—that the work proceeded, step by step, to its completion with the precision and rigid consequence of a mathematical problem." Poe claims to let the public "take a peep behind the scenes" and compares the creative process to a mechanical one by speaking of its "wheels and pinions"; his explanation is directed against the notion that poets compose "by a species of fine frenzy—and ecstatic intuition." But he also lets sound (the equivalent of Gauguin's "musical part") play a leading role in the process and defines the poem's plot—its subject matter—as "a pretext for the continuous use of the one word 'nevermore'," the refrain determined by the search for a "key-note" expressing the sadness meant to produce poetical beauty. Poe already used the notion of "suggestion," explaining that the motif of the bust of Pallas (on which the raven perches) had been "suggested by the bird" and emphasizing the general need for "some amount of suggestiveness—some under-current, however indefinite, of meaning."

Fig. 3 Paul Gauguin, Nevermore, 1897. Oil on canvas. Courtauld Institute Galleries, London

The theoretical and programmatic ambition of Poe's explanation is also more explicit than in Gauguin's case and is already made clear by its title. Venerated by Baudelaire and Mallarmé, Poe had a deep impact upon the French symbolists and his "Philosophy of Composition" served as an antidote to the unlimited faith in inspiration. By transposing it in the field of painting with Genesis of a Picture, Gauguin was again pleading for the equal standing and independence of his art. That he did so in writing and with arguments partly borrowed from literature is only an apparent paradox since he could rely on genuine parallels in literary and art (as well as musical) theory. The relationship between "The Philosophy of Composition" and Genesis of a Picture also attests to the existence of an intertext of reflections upon the creative process, formulated by those directly involved in it and crossing the borders of the different arts or media. (The link between the two was already noted by Robert Rey in 1923.²² ) The Raven had been translated into French by Mallarmé in 1875 and published with drawings by Édouard Manet; Gauguin had paid them a triple homage in 1891 with his etched portrait of the French leader of symbolism, whose head is echoed on the right by the ghostly bird, alluding to both the poet and the earlier illustrations by the painter.²³ Gauguin's interest in Poe is further evidenced by a quotation included in the Cahier pour Aline.²⁴ As for Poe's relevance for Mana'o tupapa'u, it is made manifest in the painting Nevermore of 1897 (Courtauld Institute Galleries, London; fig. 3), which is to some extent a variation on the earlier painting and where the tupapa'u is replaced by the bird. This replacement is consistent with the comments on the raven made by Poe in "The Philosophy of Composition." When the bird flaps its wings against the shutter, the lover adopts "the half-fancy that it was the spirit of his mistress that knocked," and the raven's repetition of the word "nevermore" in answer to his questions makes him believe "in the prophetic or demoniac character of the bird": the raven functions, in effect, as the "spirit of the dead." It is thus logical to suppose that The Raven and "The Philosophy of Composition" played, in the genesis of Mana'o tupapa'u, a role acknowledged indirectly in Genesis of a Picture and in Nevermore.

Heuristic Value
In my account, Genesis of a Picture appears to be an intervention in the market of interpretations and a contribution to the rivalry between artists and writers. It also expresses a certain conception of the creative process and of aesthetic communication, and points indirectly to sources of the painting that it fails to mention explicitly. In other words, its origin, its composition, and the intentions associated with it are at least as complex as those of the picture itself. Does it nevertheless possess a heuristic value in relation with Mana'o tupapa'u? Any answer to this question must remain conjectural, to the extent that there exist no other traces of the composition (in the dynamic sense of process) of the painting except possible steps toward the tupapa'u figure and an undated drawing of the reclining nude that could support the idea that this was its starting point.²⁵ But I think the answer is yes, provided one accepts that this value is mediated by the logic of the writing itself and resides in specific elements and in the movement of the text rather than in what it purports to reveal.

Fig. 4 Paul Gauguin, The Sleeping Child (Portrait of Clovis Gauguin), 1884. Oil on canvas. Private Collection

One detail that Genesis of a Picture can illuminate is the flowers depicted in the background of the painting. In his letter to Mette, Gauguin wrote that since they were "not real, only imaginary," he made them look like sparks or phosphorescences, which frighten Kanakas who believe them to partake of the spirit of the dead.²⁶ The connection is made clearer in the Cahier pour Aline: "These flowers are tupapa'u flowers, phosphorescences, a sign that the ghost is thinking of you. Tahitian beliefs."²⁷ Despite this specific reference and the fact that they resemble the hotu flowers that do gleam in the night in Tahiti, similar forms appear in the background of a depiction of his sleeping son, painted by Gauguin eight years earlier (private collection; fig. 4). The plant and bird forms are here justified as wallpaper motifs but implicitly evoke the oneiric world of the sleeper—they are equally "not real, only imaginary." In the Cahier pour Aline, the reference to "Tahitian beliefs" mediates from the detail of the flowers to the picture as a whole: "The title Mana'o tupapa'u has two meanings, either she thinks of the ghost, or the ghost thinks of her."²⁸ The letter to Mette introduces the list of titles with their translations by observing that "this language is bizarre and gives several meanings"; regarding Mana'o tupapa'u, it notes that mana'o means "to think" and "to believe," and translates the title both as "thinks of, or believes in, the ghost" and as "ghost or spirit of the dead watches over her."²⁹

Here again, the subject appears inspired by Gauguin's exploration of the life and thought of the Maori,³⁰ while reflecting his longstanding interest in thought processes and visualization. In a letter of 1888 to Vincent van Gogh, Gauguin had explained that in The Vision of the Sermon (National Gallery of Scotland, Edinburgh) "the landscape and the wrestling [of Jacob with the angel] exist only in the imagination of the praying people as a result of the sermon."³¹ One can argue that in spite of its religious iconography, The Vision of the Sermon is less about "belief" than about imagination, a distinction made unnecessary in the Tahitian pictures by the double sense of mana'o, "penser croire." What is new in Mana'o tupapa'u, on the other hand, is the fact that the thought/belief relationship is reciprocal and reversible: either the woman imagines the spirit of the dead or the spirit of the dead imagines the woman—or "watches over her," as its position in profile beside the bed also suggests. The various ontological levels of the painting, expressed by distinct stylistic treatments and levels of abstraction, are thus brought into a dynamic relationship.

Fig. 5 Édouard Manet, Olympia, 1863. Oil on canvas. Musée d'Orsay, Paris. ©Photo RMN—Hervé Lewandowski

These "levels of unreality"—to use the phrase introduced by Sven Sandström—also distinguish Mana'o tupapa'u from another source that is not mentioned in Genesis of a Picture, Manet's Olympia (Musée d'Orsay, Paris; fig. 5), which Gauguin had copied in February 1891.³² Paul Cézanne had already revisited the famous painting in his Modern Olympia (1873; Musée d'Orsay, Paris), which introduced the (male) spectator into the picture and made the reclining prostitute appear to him like a vision rather than a reality. This spectator is not directly represented in Mana'o tupapa'u, but the anecdote told in Noa Noa suggests his indirect presence. Striking matches in the unlit room, the narrator discovers the terror-stricken Teha'amana looking at him without recognizing him. He himself is paralyzed by a "strange uncertainty" that he describes as double: "Did I know what she thought I was, in that instant? Perhaps she took me…for one of those legendary demons and specters, the Tupapaus that filled the sleepless nights of her people? Did I even know who she really was? The intense feeling that possessed her, under the physical and moral domination of her superstitions, made her so foreign to me, so different from all that I had been able to glimpse until then."³³ In this story, at the same moment when Gauguin experiences the irreducible otherness of his Tahitian lover, he also becomes part of her imaginary world as a tupapa'u, a spirit of the dead. The transformation is supported in the original language by the fact that the narrator, coming back from town, is indeed a revenant, a French expression for "ghost." Insofar as the revised version of Noa Noa can be considered relevant for the painting, one can therefore suppose that the painter/lover/spectator is partly present within the picture by way of the tupapa'u.

Creative Process and "Traumarbeit"
This indirect presence or re-presentation can be interpreted as the result of a transfer or displacement. The latter term would be used a few years later by Sigmund Freud for one of the operations making up what he called the "dream-work," the elaboration of the dream.³⁴ In fact, Gauguin's account of the genesis of Mana'o tupapa'u can be closely compared to Freud's analysis. The first impulse is clearly erotic, even if the expression, for reasons of propriety, is slightly euphemized. In the letter to his wife, Gauguin begins his account as follows: "I made a nude after a young woman. In this position, a trifle would make her indecent. But this is how I want her, the lines and the movement interest me."³⁵ Je la veux ainsi can easily be understood in sexual terms, as can the parallel explanation, given in Cahier pour Aline, that he had been "seduced by a form, a movement." In Noa Noa, the narrator confesses that he had "never seen [his lover] so beautiful, and above all, never found her beauty so moving"; the episode concludes with a "sweet and ardent night, a tropical night." But this erotic impulse is checked and diverted by an internalized censorship: since the picture must not be indecent, the reference to lovemaking is replaced by the fear of the spirits, the figure of the lover by that of the tupapa'u.

One can also find in Mana'o tupapa'u equivalents of the other operations of Freud's "dream-work." The result of "condensation" can be observed in details like the "flowers" or the ornamental motifs of the pareu, which suggest a grotesque head smiling beneath the woman's thighs. These ambiguous or potential images, included by Gauguin in many of his Tahitian paintings, introduce another ontological level in the painting. They relate to his notion of "suggestion" and to his interest in mental imagery and in the Maoris' "beliefs" or "superstitions." In his letter to Monfreid, Gauguin makes clear that the night of Mana'o tupapa'u also represents the woman's "thought" and, in Noa Noa, the narrator describes the semi-obscurity of the room as "peopled with dangerous apparitions and equivocal suggestions."³⁶ Finally, the choice of objects and the pseudo-logical development of a theme meant to justify the initial motif can be compared to the operations that Freud calls "taking account of figurability" and "secondary elaboration."

Fig. 6 Frontispiece in Marie-Jean-Léon d'Hervey de Saint-Denys, Les rêves et les moyens de les diriger (Paris: Amyot, 1867)

We have seen that Gauguin had attempted to visualize the oneiric activity as early as 1884. He could not have been aware of Freud's work at this date or, in all probability, when he painted Mana'o tupapa'u and wrote Genesis of a Picture, but he could know many earlier works on the dream, especially by French researchers like Marie-Jean-Léon d'Hervey de Saint-Denys and Alfred Maury.³⁷ On the frontispiece to his 1867 book (fig. 6), d'Hervey de Saint-Denys depicted, beneath a typically "indecent" oneiric episode, six representations of "hypnagogic hallucinations," that is, images seen in a half-sleep state. They bear a generic resemblance to the quasi-abstract motifs employed by Gauguin in his evocations of dreams, "visions" and apparitions, which may be explained by the artist's knowledge of this plate or, more probably, by a common basis in direct observation.

Comparison between Freud's analysis of the oneiric process and Gauguin's account of the creative process is not limited to the issue of the dream.³⁸ Both intend to reconstitute a dynamic process in reverse—an aim shared by Poe but made almost tautological in his case by the fact that he claimed to have composed The Raven "backwards." It is well known that Freud would progressively extend his analysis of the dream-work to all psychical productions rooted in the unconscious, including literary and artistic works. As for Gauguin, he used the term "dream" when referring to the elaboration of his works in ways that go beyond traditional clichés. Explaining his making of Where do we come from? What are we? Where are we going? to Fontainas, for instance, he wrote that he had been "painting and dreaming at the same time" and that the title came to him "when I awoke, my work completed."³⁹ One can even compare Freud's interest in an 1884 article by the linguist Karl Abel—according to whom the most ancient languages ignored the principle of contradiction and contained words possessing two opposite meanings—with Gauguin's fascination with the "bizarre" ambiguity of the Tahitian language, remarkably exploited in Mana'o tupapa'u.⁴⁰ The parallel between Gauguin's self-interrogation about his creative activity and Freud's auto-analysis and preparation of Die Traumdeutung is thus not merely chronological but can be related to common preoccupations, including with "archaic" or "primitive" processes.⁴¹ In a sense, Gauguin's attempt to uncover the "genesis" of his picture, like Poe's more aggressive exposure of the "wheels and pinions" of poetical composition, partakes of the "unveiling" tendency that I mentioned at the beginning, for which Freud's "psycho-analysis" would serve as a major model. It takes a risk shunned by Mallarmé, who in 1894 called it "impious" to "disassemble fiction and the literary mechanism in public," and by Redon, who told Mellerio in 1898 that the "birth" of works of art should be kept hidden.⁴² Yet the "mystery" that Mallarmé and Redon wanted to preserve was also important to Gauguin, which may explain why he was finally ambivalent about Genesis of a Picture. He finished his letter to his wife by saying that what he had just written was "very boring" but "necessary for you over there"—that is, to deal with critics—and he concluded, in Cahier pour Aline: "This genesis is written for those who always want to know the whys, the because. For the others, [the painting] is quite simply a study of a nude from the Pacific."⁴³

1. The Essential Writings of Marcel Duchamp, eds. Michel Sanouillet and Elmer Peterson (London: Thames & Hudson, 1975), p. 138. A first version of the present paper was presented at the 26th International Conference of Art History organized by the Instituto de Investigaciones Estéticas of the Universidad Autonoma de Mexico in Saltillo, Coahuila (10–14 November 2002).

2. See Richard Brettell et al., The Art of Paul Gauguin, exh. cat., National Gallery of Art, Washington, D.C. (Washington, D.C.: National Gallery of Art, 1988), pp. 279–82, no. 154.

3. Paul Gauguin, Lettres à sa femme et à ses amis, ed. Maurice Malingue (Paris: Grasset, 1946), letter CXXXIV of 8 December 1892, pp. 239–42; Lettres de Gauguin à Daniel de Monfreid (Paris: Falaize, 1950), letter VII, pp. 97–101; Cahier pour Aline, pp. 7v, 8, 8v, 8bis (MS in the Bibliothèque d'art et d'archéologie, Paris; facsimile ed. Victor Merlhès [Bordeaux and Paris: William Blake and Co., Société des amis de la Bibliothèque d'art et d'archéologie, 1984]), pp. 18–21; Noa Noa (MS in the Musée du Louvre, Paris), pp. 108–10 (Noa Noa, Paris: Crès, 1929, pp. 92-94). The first version (MS in the Getty Center for the History of Art and the Humanities, Santa Monica, California) is translated in Noa Noa: Gauguin's Tahiti, ed. Nicholas Wadley, trans. Jonathan Griffin (Oxford: Phaidon, 1985), pp. 37–38.

4. "Récapitulons–Partie musicale–Lignes horizontales ondulantes–accords d'orangé et de bleu reliés par des jaunes et des violets leurs dérivés. Eclairés par étincelles verdâtres–Partie littéraire–L'esprit d'une vivante lié à l'esprit des Morts. / La nuit et le jour." Translation from Brettell 1988, p. 281.

5. Robert Goldwater, "The Genesis of a Picture: Theme and Form in Modern Painting," Critique 1 (October 1946), pp. 5–12; Robert Rey, Gauguin (Paris: Rieder, 1923), pp. 38–40.

6. Richard S. Field, Paul Gauguin: The Paintings of the First Voyage to Tahiti (Ph.D. diss., Harvard University, 1963; New York and London: Garland, 1977), p. 115.

7. "Cette traduction est seulement pour toi afin que tu puisse [sic] la donner à ceux qui te la demanderont. Mais sur le catalogue je veux qu'on mette les titres comme ils sont sur le tableau. Cette langue est bizarre et donne plusieurs sens." Gauguin, Lettres à sa femme et à ses amis, p. 240.

8. Translation by Henry J. Stenning in Paul Gauguin, Letters to his Wife and Friends, ed. Maurice Malingue (Cleveland and New York: World Publishing Co., 1949), pp. 176–78.

9. See Harrison and Cynthia White, Canvases and Careers: Institutional Change in the French Painting World (Chicago and London: University of Chicago Press, 1993; 1st ed., 1965).

10. See Elise Eckermann, "En lutte contre une puissance formidable": Paul Gauguin im Spannungsfeld von Kunstkritik und Kunstmarkt (Ph.D. diss., Rheinische Friedrich-Wilhelms-Universität zu Bonn, 2001; Weimar: Verlag und Datenbank für Geisteswissenschaften, 2003).

11. Paul Gauguin, Racontars de Rapin, ed. Victor Merlhès (Taravao: Éditions Avant et Après, 1994). On the interdependence and rivalry between artists and writers at the end of the nineteenth century, see Dario Gamboni, La plume et le pinceau: Odilon Redon et la littérature (Paris: Minuit, 1989).

12. James Abbott McNeill Whistler, The Gentle Art of Making Enemies, 2d enl. ed. (London: Heinemann, 1892; reprint, New York: Dover Publications, 1967), p. 146.

13. "Rien ne se fait en art par la volonté seule. / Tout se fait par la soumission docile à la venue de l' 'inconscient.' " Lettres d'Odilon Redon 1878–1916 (Paris and Brussels: Librairie nationale d'art et d'histoire and G. van Oest, 1923, p. 34, 16 August 1898).

14. Paul Gauguin, Lettres à André Fontainas (Caen: L'Echoppe, 1994), pp. 47, 14–16; see also Gauguin, Lettres à sa femme et à ses amis, letters CLXXX and CLXXII, pp. 286–90, 292–95.

15. "J'ai essayé dans un décor suggestif de traduire mon rêve sans aucun recours a des moyens littéraires…." Gauguin, Lettres à André Fontainas, p. 16; Gauguin, Lettres à sa femme et à ses amis, p. 289. Décor could also be translated as "scene" or "scenery," but given the positive meaning of décoration for Gauguin, it is more probable that he wanted to characterize the whole painting.

16. "… il y a en somme en peinture plus à chercher la suggestion que la description, comme le fait d'ailleurs la musique." Lettres de Gauguin à Daniel de Monfreid, letter LXXVII, p. 182; in his letter to Monfreid about Mana'o tupapa'u, Gauguin wrote that he chose to use a chrome yellow for the sheet because "this colour suggests the night without explaining it" (ibid., p. 101).

17. This conception was already sketched out by Gauguin in his letter of 14 January 1885 to Emile Schuffenecker. Gauguin, Lettres à sa femme et à ses amis, letter XI, pp. 44–47.

18. Freely translated from "il me faut expliquer cet effroi avec le moins possible de moyens littéraires comme autrefois on le faisait." Ibid., p. 241.

19. "Dans l'art ainsi compris, la fin n'étant plus la reproduction directe et immédiate de l'objet, tous les éléments de la langue picturale, lignes, plans, ombres, lumières, couleurs, deviennent, on le comprendra, les éléments abstraits qui peuvent être combinés, atténués, exagérés, déformés, selon leur mode expressif propre, pour arriver au but général de l'œuvre: l'expression de telle idée, de tel rêve, de telle pensée." Albert Aurier, Textes critiques 1889–1892: Des impressionistes au symbolistes (Paris: École nationale supérieure des Beaux-Arts, 1995), p. 103; originally published in "Les Symbolistes," Revue encyclopédique, no. 32, 1 April 1892, pp. 475–87).

20. Marcel Giry, "Une source inédite d'un tableau de Gauguin," Bulletin de la Société d'Histoire de l'Art Français (1970; published 1972), pp. 181–87. Charles Blanc's treatise was republished in 1885 and the impact of his transmission of Humbert de Superville's ideas has been especially noted in the case of Seurat. See, for example, Seurat, exh. cat., Galeries nationales du Grand Palais, Paris; Metropolitan Museum of Art, New York (Paris: Éditions de la Réunion des musées nationaux), 1991, pp. 435–36.

21. Edgar Allan Poe, "The Philosophy of Composition," in Tales, Poems, Essays (London and Glasgow: Collins, 1952), pp. 503–13. On Wirkungsästhetik, see Wolfgang Drost, "Zwei Etappen auf dem Wege von der normativen Ästhetik zur Wirksästhetik und zur Theorie von der 'Poesie des Kunstwerks': Simon-Théodore Jouffroy und Charles Baudelaire," in Albrecht Leuteritz et al., eds., Festschrift für Georg Scheja zum 70. Geburtstag (Sigmaringen: Thorbecke, 1975), pp. 203–12.

22. Rey 1923, p. 39.

23. See Brettell 1988, no. 116. The poem was recited (in Mallarmé's translation) during the banquet organized on 23 March 1891 to honor Gauguin's departure for Tahiti.

24. "Notes d'Edgar Poe" (Cahier pour Aline, p. 1v).

25. For the tupapa'u, Field pointed to two drawings of a woman in the Carnet de Tahiti (pp. 55v and 56r). See facsimiles in Carnet de Tahiti (Taravao: Éditions Avant et Après, 2001), unpaginated. The charcoal drawing of the reclining nude (c. 1892, location unknown) is reproduced in John Rewald, Post-Impressionism: From Van Gogh to Gauguin (New York: Museum of Modern Art, 1956), p. 527, and in Noa Noa: Gauguin's Tahiti, 1985, p. 78. Gauguin wrote to Monfreid that the nude in Mana'o tupapa'u had not been painted from life.

26. "Il y a quelques fleurs dans le fond mais elles ne doivent pas être réelles étant imaginatives, je les fais ressembler à des étincelles. Pour le Canaque les phosphorescences de la nuit sont de l'esprit des morts ils y croient et en ont peur." Gauguin, Lettres à sa femme et à ses amis, pp. 241–42. The Cahier pour Aline also speaks of "electrical sparks."

27. "Ces fleurs sont des fleurs de Tupapau, des phosphorescences, signe que le revenant s'occupe de vous. Croyances tahitiennes."

28. "Le titre Manaò Tupapaú a deux sens, ou: elle pense au revenant, ou: le revenant pense à elle."

29. "Manao Penser croire / pense ou croit au revenant / tupapau / (Esprit des ou Revenant / veille sur Morts / elle." Gauguin, Lettres à sa femme et à ses amis, p. 240.

30. See Jehanne Teilhet-Fisk, Paradise Revised: An Interpretation of Gauguin's Polynesian Symbolism (Ph.D. diss., University of California at Los Angeles, 1975; Ann Arbor: University of Michigan Research Press, 1983), passim; and Stephen F. Eisenman, Gauguin's Skirt (London: Thames and Hudson, 1997), pp. 119–30. Teilhet-Fisk (1975, p. 73) considered Mana'o tupapa'u to be "the first work by Gauguin, or any other artist, to reveal a remarkable knowledge and understanding of the Tahitian belief in the supernatural world."

31. "Pour moi dans ce tableau le paysage et la lutte n'existent que dans l'imagination des gens en prière par suite du sermon." Victor Merlhès, ed., Correspondance de Paul Gauguin, Paris, 1984, p. 232, no. 165.

32. Sven Sandström, Levels of Unreality: Studies in Structure and Construction in Italian Mural Painting during the Renaissance (Uppsala: Almqvist & Wiksell, 1963); see Brettell 1988, no. 117.

33. "Savais-je ce qu'à ce moment j'étais pour elle? Si elle ne me prenait pas, avec son visage inquiet, pour quelqu'un des démons ou des spectres, des tupapau dont les légendes de sa race emplissent les nuits sans sommeil? Savais-je même qui elle était en vérité? L'intensité du sentiment qui la possédait, sous l'empire physique et moral de ses superstitions, faisait d'elle un être si étranger à moi, si différent de tout ce que j'avais pu entrevoir jusque-là." Gauguin, Noa Noa (Louvre MS), p. 93.

34. Sigmund Freud, Die Traumdeutung (Leipzig and Vienna: Franz Deuticke, 1900 [published 1899]).

35. "Je fis un nu de jeune fille. Dans cette position, un rien, et elle est indécente. Cependant je la veux ainsi, les lignes et le mouvement m'intéressent." Gauguin, Lettres à sa femme et à ses amis, p. 241. However, Gauguin still qualified Mana'o tupapa'u in the same letter as "raide," in the sense of risqué.

36. "Ces fleurs sont en même temps comme des phosphorescences dans la nuit (dans sa pensée)." Lettres de Gauguin à Daniel de Monfreid, p. 101: "dans ces demi-ténèbres à coup sûr peuplées d'apparitions dangereuses, de suggestions equivoques…." Noa Noa (Louvre MS), p. 93. On the notion of "potential images" and its relevance for Gauguin, see Dario Gamboni, Potential Images: Ambiguity and Indeterminacy in Modern Art (London: Reaktion), 2002, esp. pp. 18–20, 86–96. Field (1977, pp. 114–15, 118) suggested the possibility that the anecdote told in Noa Noa, compounded with Gauguin's "growing awareness that the Tahitian equated darkness with the past, was the actual formative basis of the painting."

37. Anonymous [Marie-Jean-Léon d'Hervey de Saint-Denys], Les rêves et les moyens de les diriger. Observations pratiques (Paris: Amyot, 1867); L.-F.-Alfred Maury, Le sommeil et les rêves. Études psychologiques sur ces phénomènes et les divers états qui s'y rattachent suivies de recherches sur le dévelopement de l'instinct et de l'intelligence dans leurs rapports avec le phénomène du sommeil (Paris: Didier, 1862; 1st ed. 1861). See Henry F. Ellenberger, The Discovery of the Unconscious: The History and Evolution of Dynamic Psychiatry (New York: Basic Books, 1970); Stefanie Heraeus, Traumvorstellung und Bildidee. Surreale Strategien in der französischen Graphik des 19. Jahrhunderts (Berlin: Reimer, 1998); Stefanie Heraeus, "Artists and the Dream in Nineteenth-Century Paris: Towards a Prehistory of Surrealism," History Workshop Journal 48 (autumn 1999), pp. 153–68.

38. Field (1977, p. 268, n. 27) compared Gauguin's preoccupations with Freud's in general terms and referred to Niels Sandblad's suggestion that "this would make a very fertile field for investigation."

39. "… en peignant et rêvant tout à la fois … / Au réveil, mon œuvre terminée, je me dis, je dis: D'où venons-nous? que sommes-nous? Où allons-nous?" Gauguin, Lettres à André Fontainas, pp. 15–16.

40. Sigmund Freud, "Über den Gegensinn der Urworte" (1910), in Studienausgabe, vol. IV: Psychologische Schriften (Frankfurt am Main: Fischer, 1970), pp. 227–34.

41. Kirk Varnedoe suggested that "Gauguin's description of his elaboration of the Spirit (or Mana'o tupapa'u), from initial sensory impression to symbolic art, follows a similar pattern and implicitly aligns his creative process with the primitive's construction from initial utterance to developed language" and pointed to evidence of Gauguin's acquaintance with linguistic theory. See Kirk Varnedoe, "Gauguin," in William Rubin, ed., "Primitivism" in 20th Century Art: Affinity of the Tribal and the Modern, exh. cat., Museum of Modern Art, New York; Detroit Institute of Arts; Dallas Museum of Art (New York: Museum of Modern Art, 1984), p. 200 and p. 208, n. 62.

42. Stéphane Mallarmé, "La musique et les lettres," (1894), Oeuvres complètes, ed. Henri Mondor and G. Jean-Aubry (Paris: Gallimard, 1945), p. 647 ("le démontage impie de la fiction et conséquemment du mécanisme littéraire"); Lettres d'Odilon Redon, p. 30, letter of 21 July 1898 ("Le point initial de mes ouvrages vous importe-t-il tant que ça? Ne vaudrait-il pas mieux le cacher un peu; est-il bien de regarder ainsi la naissance!").

43. "Ce que je t'écris là est très ennuyeux mais je crois que cela t'est nécessaire pour là-bas." Gauguin, Lettres à sa femme et à ses amis, p. 242; "Cette genèse est écrite pour ceux qui veulent toujours savoir les pourquoi, les parce que. / Si non, c'est tout simplement une étude de nu océanien."