“Yes, my dear poet, to conceive literature and for it to have a reason must lead to this “lofty symphony” that no one perhaps will ever write; but it has haunted even the least aware and its principal features mark, subtly or vulgarly, every written work. Music in the proper sense, which we must pillage, plagiarize, if our own, unspoken, is insufficient, suggests such a poem.”

—Mallarmé, letter to Paul Valéry, May 5, 1891

On Thursday, February 27, 1890, in the salon of Berthe Morisot in Paris, Stéphane Mallarmé gave a lecture on his recently deceased friend, the poet Villiers de l’Isle-Adam. “A man accustomed to dream comes here to speak of another, who is dead,” he began. In the first row sat Edgar Degas, an admirer of Mallarmé (whose masterly photograph of Mallarmé and Renoir was one of the most beautiful exhibits at the Musée d’Orsay in last year’s commemoration of the centenary of Mallarmé’s death). After a few minutes Degas left precipitously, holding his head in his hands, and crying, “I do not understand, I do not understand.”

Most essays on Mallarmé begin with the difficulty of understanding him, and this review will be no exception. That is, in fact, how Bertrand Marchal opens the introduction to his splendid new edition of Mallarmé’s complete works, of which the first volume of two has just been published. He quotes at once Mallarmé’s ferocious answer in 1896, in his article “Le Mystère dans les lettres,” to an attack on the difficult style of modern poetry by the young Marcel Proust:

I prefer, faced with aggression, to retort that my contemporaries do not know how to read—

Except in the newspaper; it certainly provides the advantage of not interrupting the chorus of preoccupations.

Proust, who was eventually to become an admirer, had written three years before that Mallarmé had not much talent but

he was a brilliant talker. How unfortunate that so gifted a man should become insane every time he takes up the pen.

To be considered mad by one’s contemporaries is common enough in the history of nineteenth-century art, but the obscuring fog lifts with time as the public learns new ways of reading, looking, and listening. In Mallarmé’s case, however, if the lapse of a century has brought some comprehension and greater sympathy, most of his finest work remains today still unapproachable, or at least largely incomprehensible, not only to the general public, but even to the university-educated reader—even, in fact, to a good part of that curious, tiny, and still-dwindling public passionately interested in verse.

Nevertheless, this seemingly unreadable author has acquired immense prestige and is accepted without question as one of the greatest of French poets. He is rivaled in the latter half of the nineteenth century only by Baudelaire, Rimbaud, Verlaine, Laforgue, and the elderly Victor Hugo. The new edition is the second attempt to present his work in the Pléiade library, which aims to make available the important classics of world literature. The amount of the actual poetry proper—that is, the serious pieces of verse—is so small as to be unique for a major poet. This first volume of more than 1,500 pages starts with the poems that Mallarmé himself collected, a section of only forty-two pages. To this are added seventeen pages of verse he did not collect, twenty pages of poems in prose, and the twenty pages of his extraordinary typographical experiment Un Coup de dés jamais n’abolira le hasard (“A throw of the dice will never abolish chance”).

This astonishingly modest body of work of less than one hundred pages must be eked out with 70 pages of verse written before Mallarmé was twenty years old, and 120 pages of versified amusements—in particular rhymed addresses on post cards with which Mallarmé liked to puzzle the post office. To this are added a selection of letters concerning poetry, what remains of a few unfinished works, almost a hundred pages of earlier versions of the published poems, and over 300 pages of transcriptions of manuscript notes and sketches, as well as a further 350 pages of explanatory notes and variant readings. The rest of the prose works are reserved for a second volume. However, the line between his poetry and prose is hard to draw. Many of the critical articles and the journalism have a poetic quality equal to the verse, and the prose can be as difficult as the poetry—often, in fact, considerably more difficult.

The less than fifty pages of serious verse, however, remain the center of Mallarmé’s achievement, along with Un Coup de dés. It is only in the presentation of this small nucleus that the new edition can be faulted, and the blame is largely the publisher’s parsimony. Printing a sonnet with eleven lines on one page, and with the last three lines to be discovered overleaf, is a disgusting procedure when dealing with any poet. In the case of Mallarmé, who spent a lifetime considering the aspect of a page and for whom the white spaces around a poem were as important as the printed characters, we must call this a betrayal. It is a particularly foolish economy if one considers that beginning each poem of Mallarmé properly on a new page would have added only about twenty pages to a volume of 1,610 pages; and the trivial increase in expense could have been made up easily by printing the prose poems in the second volume—which will be done anyway.


I can see that Mallarmé’s own preference for a full blank leaf recto and verso between each poem would make excessive demands on modern publishing, but it is not too much to ask that some thought be given to the design of a page, at least with the small number of major poems. And with Un Coup de dés, although the large format demanded by the poet for arranging his words with different typefaces on the page like a constellation was not possible within the format of the series, we could have been spared the thoughtless insertion of little numbers and letters indicating variant readings and explanatory notes in the back of the book, a form of vandalism more or less equivalent to printing the names of the characters over their heads in the reproduction of a group or family portrait by Titian.

The appearance of the poem on the page was essential for Mallarmé’s aesthetic: he felt that the poem must be read as an object detached from the life of the author, separated from the society and culture in which it arose, and even independent of the world in which it existed, as if it had its being outside history. His poems may be said to detach themselves partially from the literary tradition that made them possible—at least they appeal over the head of that tradition to manifest themselves as part of the language itself, but a language purified of the ordinary slipshod meanings it has in life, cleansed of the daily need simply to communicate. “To give a purer sense to the words of the tribe” is perhaps the most often quoted line of Mallarmé (“Donner un sens plus pur aux mots de la tribu,” from the sonnet on Edgar Poe; “To purify the dialect of the tribe” is T.S. Eliot’s inspired version in “Little Gidding” from the Four Quartets). We learn from the new edition that Mallarmé first wrote “To give too pure a sense” (“trop pur“). The original version is more provocative (and I am not sure that the change was not made principally because Mallarmé wanted to add an extra “u” sound to supplement “pur” and “tribu“).

It may seem strange that the supremacy, the privileged status, of a poet should remain absolutely unchallenged when access to his work is largely denied to the largest part of his potential audience; the hermetic character of his poetry seals it off from most readers. The endurance of his prestige rests on two factors: on the passion that his work arouses in those who are determined to read him and to make sense of the poems, and on the professional importance of his achievement for the poets that succeeded him. It is not only French poetry that was radically altered by his example (his influence was crucial for Paul Valéry, René Char, Paul Eluard, Paul Claudel, Yves Bonnefoy, and many others). He was decisive in the creation of the modernist tradition in German and American poetry: his example was basic for Stefan George and Rainer Maria Rilke and remained powerful with Paul Celan, and his use of language was the model for Walter Benjamin and, I think, an important influence on Martin Heidegger’s style; in America, his work was essential for the generations that included T.S. Eliot, Wallace Stevens, and Hart Crane, and it has lasted until John Ashbery. Across the numbers of his imitators, he influenced even those who never read him. He was a liberating force.

To understand the nature of that liberation, it may be useful to start with the hatred he aroused. With the exception of Baudelaire, no poet of the past two centuries has been so fiercely execrated, and the horror inspired by Baudelaire is, on the surface at least, more easily explained by the defiantly immoral subject matter of his work. With Mallarmé, the subject matter was not so obviously objectionable—at least the fiercest attacks were launched mostly by critics who had no idea what the subject matter of most of the poems was, largely because they had not been able to figure it out. It is difficult to credit the lunacy of some of these attacks. One journalist, Max Nordau, who wrote a book called Degeneration in 1894 in which he famously attacked almost all forms of modern art—Wagner, Strauss, Manet, and whatever—inspiring a vigorous reply by George Bernard Shaw, found Mallarmé particularly obnoxious: he accused him of fostering a deplorable taste for the Pre-Raphaelites, and (oddly enough for a poet who had lost his religious faith) of spreading Neo-Catholicism, and wrote:


Let us add as well, in order to be complete, that one notices in Mallarmé “the long pointed ears of a satyr.” R. Hartmann, Frigerio and Lombroso have, after Darwin (who first emphasized the simian character of this particularity), determined the atavistic and degenerative significance of the pavilions of the ear which are abnormally long and pointed, and proved that one meets them frequently above all with criminals and the insane.

In a lecture that Mallarmé gave at Oxford, “Music and Letters,” he commented on Nordau’s attacks on modern art in general with a certain mildness and his usual elliptical complexity:

This vulgarizer has observed a fact. Nature does not engender an immediate and complete genius, which would correspond to a type of man and would not be any one man in particular; but practically, occultly, touches with an innocent thumb a certain faculty and almost abolishes it in the one to whom she proposes a contrary munificence; here these are pious arts or maternal actions that ask for a clairvoyance not lacking in tenderness from critic and judge. Let us continue: What happens? Deriving force from his privation, the crippled chosen one [infirme élu] grows toward his plenary intentions, and he leaves after him, of course, like a numberless waste product, his brothers, case histories medically labeled or election bulletins once the vote is over. The error of the pamphleteer in question is to have treated everything like a waste product. Thus the subtle arcana of physiology and of destiny, ought not to stray into the hands, too coarse to handle them, of an excellent foreman or of an upright adjuster—who stops in mid-path and behold! with the addition of some instinctive foresight, he would have understood, in time, the poor and holy natural processes and he would not have written his book.

This anticipates the theory of overcompensation of the Austrian psychiatrist Alfred Adler: it is from his weakness, from what he lacks, what is incomplete within himself, that the original genius draws the force that gives such power to his remaining faculties. This passage also affirms the Romantic notion of the source of genius in infirmity, of originality as a form of sickness. It is here, too, that Mallarmé faces the inconvenient side of avant-garde style, the minor artists who can only imitate the weakness that inspired the innovations. Bad conservative art comes from a mechanical imitation of the conventional; the failure of so much modern art comes from an inability to distinguish the infirmity of the innovator from his strength. After Mallarmé, a considerable amount of bad poetry appeared that was as hermetic and as forbidding as his masterly creations. He was, as I have said, a liberating force, but he loosed on the world a good deal of deplorable verse while he made possible some of the greatest achievements of modern poetry, including his own.

The infirmity of Mallarmé that gave him his genius was his despair. In spite of his desire to keep poetry detached from the life of the poet, this despair was both personal and metaphysical. His loss of religious faith was traumatic. He had no secular faith in science or history to compensate for it; like the greatest of Italian nineteenth-century poets, Giacomo Leopardi, he was persuaded of the emptiness, the total insignificance of human life: the universe was void of meaning. The more personal trauma was his life as a high school English teacher in a small Rhône valley town, Tournon, that he found ugly and unpleasant. He cannot have been an inspiring teacher; the students were rowdy and threw chalk at him. It was at Tournon that he began his most famous poems, “Hérodiade” and “The Afternoon of a Faun.” On his transfer to Paris, his situation was more cheerful, made agreeable above all through his friends and disciples who were fascinated by his theories of language and of aesthetics. What he said about Villiers de l’Isle-Adam was equally true about himself: “His life—I find nothing that corresponds to this term: truly and in the ordinary sense, did he live?” His poems mostly circulated privately. The few that were published in reviews created at one and the same time a growing hostility and admiration.

Paul Valéry, whose encounter with Mallarmé’s poetry transformed his view of literature and of life, eloquently reflected on the hostility that his master provoked.

In this strange work, a sort of absolute, there dwelt a magic power. Merely by existing, it acted as a charm and as a sword. With one stroke it divided all the people from those human beings who knew how to read. Its appearance of an enigma instantly irritated the vital nervous center of literary understanding. It seemed immediately, infallibly, to strike at the most sensitive of cultivated consciousness, to overstimulate the center itself where there exists, and is contained, some prodigious charge of self-esteem, and where resides that which cannot suffer not to understand

What happened is that the moment one laid eyes on it, this work without peer hit on, and grappled with, the fundamental convention of ordinary language: You would not read me if you had not already understood me.1

It is true that the intolerant protest against difficult art was as philistine and as morally disgraceful in Mallarmé’s time as it is today, and it is also true that not being able to understand can be the excuse for a childish tantrum. I knew a five-year-old boy who bit a foreign baby sitter on the leg because he could not grasp what she was saying to him. However, Valéry’s account stops short of a satisfactory explanation. Few sane people experience an uncontrollable resentment at not being able to comprehend what is beyond their technical competence, like differential equations or the more complicated formulas of quantum theory. Nor is there any inevitable shame in an inability to share someone else’s taste or artistic preferences. The problem with Mallarmé, as with so many innovators, lay on a deeper level. Reading him, many critics sensed that they were on the verge of understanding, trembling on the edge of comprehension: worse, they had a suspicion that what they were reading was very good. One anonymous critic wrote in 1876, after quoting a large extract from the opening of “The Afternoon of a Faun”:

What is terrible in the case of Monsieur Mallarmé is that, in reading these absolutely incomprehensible verses, one senses a sort of music which is an incontestable witness to a poetic temperament and a refined nature.

Is Monsieur Mallarmé in good faith? Does he want to replace poetry which says something with a new poetry, like Manet in painting, which limits itself to giving an impression? One must believe it in the interest of Monsieur Mallarmé…2

We can only marvel at a critical intelligence that can find the ex-quisite beginning of “The Afternoon of a Faun” absolutely incomprehensible and yet recognize the kinship of the poet to the friend he so much admired and who painted his portrait.

Mallarmé himself always denied, surely with at least a touch of irony, that his poems were difficult. He once said to the poet Henri de Régnier, speaking about a friend: “He’s a charming boy, but why does he explain my verses? That would tend to make one believe they were obscure.”3 Nevertheless, he was partly in earnest: read properly, he felt, his poetry gave at least a satisfactory surface meaning at once. In “The Mystery in Literature,” his answer to Proust’s attack on obscurity, he claimed:

Every piece of writing, on the outside of its treasure, must—out of respect for those from whom, after all, it borrows the language, for a different purpose—present with the words a meaning, even if an unimportant one: there is an advantage to turning away the idler, who is charmed that nothing here concerns him at first sight.

The idler is the one who scans a poem as if he were reading a newspaper, without the passion and the good will that enables a reader to abandon himself completely to the work. The exterior meaning to be given the idler as a sop to protect the treasure inside was not, however, always successful in Mallarmé’s case in appeasing that casual reader’s incipient anxiety. Mallarmé’s ambition did not allow him too easy a surrender to the idler’s susceptibilities. All too often in his work, the exterior meaning appeared to be only a provocation, a way of implying that there was a hidden sense. As Valéry remarked, the reader of Mallarmé is required to do much of the work for himself.

A sonnet published in 1887, not, perhaps, one of my favorites but a fine and relatively simple example of the poet’s method at an extreme point, opens enigmatically:

Victorieusement fui le suicide beau
Tison de gloire, sang par écume, or, tempête!

[Victoriously fled the beautiful suicide
Burning brand of glory, blood through foam, gold, tempest!]

The word “sunset” is unspoken, but it is represented by image and metaphor. In a splendid blaze of color, the sun voluntarily annihilates itself each evening.4 The meaning comes only a little further to the surface with the second quatrain:

Quoi! De tout cet éclat pas même le lambeau
S’attarde, il est minuit…

[Well! of all that brilliance not even the shred
Lingers, it is midnight…]

This way of describing by suggestion has an importance for the poem that exceeds the idea of the sunset. Withholding the referential meaning concentrates attention initially upon the technique of representation: the poem refuses to allow the reader to substitute immediately the concept for the description. To understand we must return over and over again to the lines. Mallarmé fixes the attention of the reader where it properly belongs—on the words of the poem, the assonance, the rhythm, the juxtaposition of images, the emotional associations of burning brand, glory, blood, foam, suicide. He forces the reader to pause at each point in the attempt to understand the relations of one image to another in the strange collection, and this allows, as he himself put it, the different facets of the words to reflect off each other. Some wit once said that Mallarmé was a poet so difficult that only foreigners could understand him. There is this much truth in this: foreigners do not read paragraphs or even whole sentences; they read word by word. That is how these poems are to be read, not only to be understood, but even to be appreciated. And they cannot be understood without being appreciated.

This stands the classical way of reading poetry on its head. In general, first we take in the text visually, and we understand it almost as we take it in, and afterward we find it interesting or beautiful. With the modernist tra-dition of poetry, which is more or less founded by Mallarmé, understanding is temporarily postponed until we have savored and enjoyed the poetic art: image, metaphor, assonance, rhythm, rhyme, alliteration, and the connotations of the words that count for more in poetry than the referential meaning. All these aspects of language have traditionally played a role in poetry that displaces in importance what may be considered the message—that is, any meaning that can be translated into prose. Those who read paying little attention to these other aspects and seek only for the message have always and inevitably misunderstood the poetry and even failed to grasp the message. Temporarily blocking access to the message was only Mallarmé’s drastic way of avoiding misunderstanding. That is why he insisted he was not obscure, or even difficult.

In a letter to the English essayist Edmund Gosse he made this point emphatically. He quoted from Gosse’s article on his poetry a sentence which, he said, had a “diamond clairvoyance”: Gosse wrote, “His desire…was to use words in such harmonious combination as will induce in the reader a mood or a condition which is not mentioned in the text, but was nevertheless paramount in the poet’s mind at the moment of composition” (italics added by Mallarmé to Gosse’s text). Mallarmé commented gratefully:

It’s all there. I make Music, and I name by this not only that music that one can draw from the euphonious bringing together of words,—that is an initial condition that goes without saying: but also that music which goes beyond and is magically produced by certain dispositions of speech [parole], where this remains only in the state of a means of material communication with the reader like the keys of a piano. Truly, between the lines and above the gaze this takes place, in all its purity, without the intervention of gut strings or of pistons as in the orchestra, which is already industrial; but it’s the same thing as the orchestra, only literary and silent. Poets in all ages have never done anything else, and it’s amusing today, that is all, to become aware of it…. I only quibble with you over obscurity: no, my dear poet, except by clumsiness or awkwardness, I am not obscure the moment one reads me in order to seek what I have set forth above, or the manifestation of an art which makes use—let us say incidentally, I know the profound cause—of language: and I become obscure, of course! if one makes a mistake and thinks one is opening a newspaper.

The art of reading Mallarmé requires us to realize that the enigmatic surface of his poetry does not cover or hide a secret; and we cannot discard the surface once the treasure has been unearthed. The solution to the enigma is on the surface, which itself becomes the treasure as our experience of it grows.

Modernism in the other arts closely parallels the development effected by Mallarmé in poetry. Painting became harder to read at first glance. In a classical painting we recognize the objects portrayed at once, and then we admire the way they have been represented, the construction, the play of colors, the expressive lines. In a Cubist painting, on the other hand, it is often hard to identify what object is represented until we have admired the painted surface. Already in Manet, sections of the canvas can appear to us obtrusively first as a pattern of brush strokes before we can grasp what they represent—and it was in this aspect of the picture that Manet sometimes took the greatest pride.5 In much of James Joyce’s Ulysses it is often difficult to figure out the plot before we have enjoyed the author’s stylistic virtuosity; in the hospital sequence, for example, we can only realize what is going on if we appreciate the series of parodies that ranges through the entire history of English literature.

Music inspired Mallarmé, as it did so many of the Romantics, because the role of the message in it, of what Mallarmé called “material communication,” is so substantially reduced. Before modernism, however, the emotion or sentiment represented by the music was evident at once; after this initial response, music lovers could turn to a perception of the art with which this emotion was conveyed, the ingenuity of the representation and the abstract beauty of the melodic lines. With Schoenberg and Webern, however, and with Stravinsky (starting with The Rite of Spring), we must generally begin with a dispassionate understanding of the art and an appreciation of the technique in order to comprehend the emotional content; for those who succeed, the sentiment is perceived at last with an intensity rare in music that allows a more facile comprehension. (That is why some critics still find Schoenberg’s music to have no feeling, while his admirers find it, on the contrary, often embarrassingly overcharged with emotion.)

In those works of the modernist movement considered difficult or hermetic, in short, the content is partially withheld from us until we have understood the technique; the delayed perception then arrives with greater power. As Valéry remarked about his discovery of Mallarmé’s work:

He raised the condition of the reader, and with an admirable intelligence of real glory he chose in the world a small number of special amateurs who, once having tasted could no longer suffer impure poems, which could be understood at once and with-out resistance. Everything seemed naive and slack after they had read him.6

But Valéry’s invocation of a small elite is profoundly misleading and reflects his own intellectual aloofness; there is no reason to think that Mallarmé, Joyce, Webern, and other modernists would not have welcomed the admiration of thousands; but they wanted admiration on their own terms, and had to be satisfied with a grudging popularity, much of it only posthumous.

Mallarmé’s observation is profound: “Poets in all times have never done anything else, and it is amusing, today, that is all, to become aware of it.” It is now understood more clearly how much modernism has altered our view of the art of the past. Mallarmé was right: poets have always used all the different aspects of language that transcend the conveying of information, that have a value of their own independent of “material communication.” Some twentieth-century critics have since read Racine, Shakespeare, and other writers as if they were modernist poets, as if the surface of their work was enigmatic, and with the most skillful of these critics, it is evident that they were often justified in doing so. Poets have indeed written to some extent as Mallarmé did, and many readers of the past did, in fact, perceive this even if they were not fully aware of it. The modernist poet forced his readers, on pain of being lost in bewilderment, to become conscious of what has always been present in literature. Twentieth-century music has changed our understanding of Mozart and Beethoven. What we now hear was, of course, always there.

The extent to which the way to an understanding of content has necessarily been through a love and appreciation of the significance of technique, and a familiarity with it, can be shown by an article written in 1796 by the influential music critic Johann Friedrich Reichardt on Handel and Bach. They were both praised for their astonishing harmonic and contrapuntal mastery. But what Bach, in particular, lacked was a

deep feeling for expression…. If both these men had had more knowledge of man, of language, and of poetry, and if they had been bold enough to cast off all idle fashions and conventions, they would have become our highest ideals of art, and every great genius who today is not satisfied with equalling them would have had to overthrow our entire tonal system in order to clear a new field for himself.7

The sentiment in Bach’s music, which seems so intense to us today (and was already so understood by the generation that followed Reichardt), went unperceived by a critic who was dazzled by the technique but thought it excessive. Admission to the sentiment in Bach’s music required a familiarity with his style and a joyful acceptance of his complexity; like Mallarmé’s poetry, the elaborate and hermetic technique originally barred even the educated academic listener from the emotion implicit in almost every work, the physical passion of his lines. The reputation of Bach was made by the composers who studied him, from Mozart and Beethoven through Chopin until the present. The prestige of Mallarmé was created largely by the poets who were inspired by him.

Far from being merely a virtuoso play with the elements of language, as many of his contemporaries believed (including those who admired him as well as those who were frightened by him), his mature poetry has a tragic cast that has become more visible with time.8 His work is only in part a reaction to the deterioration of language in journalism and public life. It is fundamentally a despairing protest against ordinary life, made void by the irrelevance of all religious faith, by the corruption of general culture, by the gradual decline of the prestige of art except as a commodity. The greatest of his poems are about death, absence, empty rooms, unfulfilled ambitions. His most famous sonnet opens:

Le vierge, le vivace et le bel aujourd’hui
Va-t-il nous déchirer avec un coup d’aile ivre
Ce lac dur oublié que hante sous le givre
Le transparent glacier des vols qui n’ont pas fui!

[The virgin, ever returning and beautiful today
Will it tear us with a drunken blow of the wing
This hard forgotten lake haunted under the frost
By the transparent glacier of flights which have not flown!]

The “today” of that breathtaking first line is everyday life, and the vain hope is that it will break the winter ice in which a magnificent swan has been trapped, surrendering himself to the “resplendent ennui of sterile winter.” But as we read the poem, the second line implies at first that it is we who are rent by the eternally recurrent present, and the ambiguity makes us its victims.

Un cygne d’autrefois se souvient que c’est lui
Magnifique mais qui sans espoir se délivre
Pour n’avoir pas chanté la région où vivre
Quand du stérile hiver a resplendi l’ennui.

[A swan from the past remembers that it was he
Magnificent but without hope surrenders himself
For not having sung the region where life can be
When the ennui of sterile winter has become resplendent.]

The swan (the poet as we know from tradition) is trapped because in the past he has not sung of the region in which life can be. For Mallarmé that region was not the present, but the ideal world of poetry, which transforms language from a means of conveying a message into an object which is beautiful in itself and which exists only for itself. Poetry, however, cannot change the ever-returning present, and it does not, in fact, as the poet knew, create a world that can be lived in. The beautiful today whose existence was denied by the swan will not free it from the ice. In the last lines, the swan has become a phantom, immobilized in the cold dream of contempt (“Il s’immobilise au songe froid de mépris“). No one has ever expressed with greater power than Mallarmé the illusion of nineteenth-century aestheticism, and no image of the despair of ordinary daily life has ever surpassed the astonishing opening lines of this sonnet.

The poem that Mallarmé originally entitled “An allegorical sonnet of himself” is the description of an empty room. It ends with the image of the mirror facing the open window, in which the septet, the constellation of the Pleiades, the seven muses, has moved into sight at the open window. Mallarmé wanted to use only two rhyming sounds in this poem, one “or” (gold), the other “ix”: for the fourteen lines of a sonnet, one needs therefore eight and six rhymes. There are only five words in French that rhyme with “ix”: onyx, Phénix, Styx, nixe, and fixe. Mallarmé was forced, or was delighted, to invent a sixth—ptyx:

Sur les crédences, au salon vide: nul ptyx,
Aboli bibelot d’inanité sonore

[On the sideboards, in the empty salon: no ptyx,
Abolished trinket of sonorous inanity]

Mallarmé said he hoped that ptyx had no meaning in any language so that he could create one (it does, in fact, mean “fold” in Greek). To some it has seemed evident that a ptyx is a seashell, a common decorative object for a salon, and it is more than adequately described as sonorous inanity, since one can hold it to one’s ear and listen to the nonexistent sea. (In the original version of the poem, “abolished trinket” was “strange vessel” [insolite vaisseau]).

The ptyx is the allegorical figure of the poem itself. It is only an empty sound for it has no meaning in ordinary French; it is purely imaginative, but it comes into being by the logic of the poem’s structure; it is abolished (although it has never existed, it is no longer there), yet it takes on a meaning from the action of the poetry, by the will of the poet—who is also not there, because the room is empty and the master of the house is gone, but his symbolic image scintillates in the oblivion of the mirror.

Que, dans l’oubli fermé par le cadre, se fixe
De scintillations sitôt le septuor.

In the closed forgetfulness of the frame, is set
So soon the scintillating septet.]

The poetry of absence was not, in spite of all the intricate ways that these words reflect each other like distorting mirrors, a trivial game. It was an expression of Mallarmé’s despair. In the first stanza the void of the room is forcefully illuminated by this despair:

L’Angoisse, ce minuit, soutient, lampadophore,
Maint rêve vespéral…

[Anguish, this midnight, bears like a lampholder
Many a twilight dream…]

The faun, in “The Afternoon of a Faun,” renounces the fulfillment of sexual desire for its perpetuation by the poetic imagination. All of the late work of Mallarmé reenacts that renunciation, and the sonnet in “yx” is its emblem. Out of emptiness and absence he created a poetry of extraordinary intensity.

Nevertheless, he knew that, in one sense, art was a game of chance, although a deeply serious one. “Every thought emits a throw of the dice,” he wrote in the typographical poetic images of Un Coup de dés. The way the game must be played is compared to the movement of a constellation: “watching, doubting, rolling, glittering and meditating before stopping at some final point which consecrates it.” A throw of the dice will never abolish chance, he claimed:



How to read a poem, according to Mallarmé:


This practice—

Apply, according to the page, your ingenuousness to the white which inaugurates it, forgetful even of the title which would speak too loud: and, when, in a break (the smallest break, disseminated), chance, vanquished word by word, is lined up, the white returns inexorably, previously arbitrary, now positive, to conclude that nothing lies beyond and to authenticate the silence.

—From “Le Mystère dans les lettres”

This Issue

May 20, 1999