Mallarmé completes his oeuvre on chance.[i] But the word reappears continually in his writings. To begin with, it characterizes the structure of language: “the mere chance that persists in terms, despite the artifice of dipping them alternately into sense or sonority.”[ii] The statement has been compared to the Saussurean doctrine; Mallarmé’s “chance” appears, in fact, to announce that which the Course in General Linguistics will name the arbitrary sign.[iii] But the comment immediately following dispels such an impression: “Beside the opaque ombre [shade], ténèbres [shadows] is not very dark; what a disappointment, before the perversity that makes the timbres of jour [day] and nuit [night], contradictorily, dark in the first case, bright in the latter.”
Saussure having been the first to separate the signified, internal to language, from the thing signified, external to language, it would not be surprising to find that Mallarmé is indifferent to the distinction. He meditates exclusively on the relation between the phonic signifier, which he calls the sound, and the thing signified, which he calls the sense. Saussure’s arbitrary assumes the sound and the thing to be mutually independent; neither party promises anything to the other, nor expects anything of the other. As for the signifier and the signified, they are compared to the recto and verso of a sheet of paper; they know nothing of one another; thus, it would be as absurd to imagine that they could contradict each other as it would be to imagine that they could confirm each other. Mallarmé, on the other hand, assumes that the sonority of the term retains some property of the thing. The examples selected demonstrate that this expectation may be disappointed, but this disappointment, which is accompanied by regret, attests to a promise in which the subject was engaged. With regard to the things signified, the phonic form ténèbres (shadows) should be darker than ombre (shade); it is not; the phonic form jour (day) should be bright and the phonic form nuit (night) dark, but the reverse is the case.
Mallarmé admits two propositions that Saussure would not accept: first, that the qualities of the phonic form are of consequence; second, that these qualities should, ideally, correspond to the qualities of the thing signified. As a strict Galilean, Saussure eliminates qualities; even if one can concede that a phoneme has phonic qualities, those qualities are not, in themselves, functional in language; their role is strictly negative and oppositional: they contribute only to establishing differences between phonemes, because in language, there are only differences; in fact, they cease to be qualities and become distinctive features. Saussure announces to the speaking subject that the signifier is without qualities and that language is indifferent to the qualities of the thing signified. Mallarmé, on the other hand, evokes qualities: those of the signifier as well as those of the thing signified. Sound offers hope of sense; sense has certain expectations with regard to sound. On this condition only, one can affirm that the subject is, sometimes and even often, disappointed; “disappointment,” “perversity,” “contradictorily” — this vocabulary would have no relevance for Saussure. Mallarmé’s ‘chance,’ like Aristotle’s, imitates intentionality. Saussure’s arbitrary in no way resembles intentionality.
In English Words, one can recognize a catalogue of promises effectively kept. But Mallarmé never suggests that all English words, without exception, conform to the principles laid out. There is no doubt that he could have proposed examples illustrating the same type of linguistic malignity that he points out in French. Thus, chance first makes itself known in the disappointment that languages impose upon the subjects that speak them. Verse remedies this defect: “the versified line which from many expressions makes a total, new word, foreign to language, as if incantatory, achieves this isolation of speech, negating, in a sovereign sweep, the chance that persists in terms….”[iv] This fragment from “Crisis of Verse” is well known; nonetheless, it merits lingering over for a moment: “speech” (parole), here, is spoken aloud; this is indicated by the qualifier incantatory; it is isolated from language, as language makes no distinction between the written and the spoken. The language Mallarmé describes is literalized just as much as it is proffered. The language/speech (langue/parole) distinction is thus crucial in Mallarmé, but it does not correspond to Saussure’s distinction of the same name. Saussure opposes the individual dimension of speech to the collective dimension of language; this is not what Mallarmé stresses. Speech, for him, is fundamentally a matter of sonorous vibration: “…to transpose a fact of nature into its vibratory near-disappearance through the play of speech….”[v] In customary usage, this play reunites a sound with a “fact of nature,” day or night, for example; the relation between the qualities of the sound and the qualities of the fact of nature is governed by chance.
Such, at least, is the rule. Poetry constitutes an exception. In poetry, speech is revealed to be capable of fashioning language into an entirely new object: a word, made up of several expressions, in which the sonorous promise will be kept, though each expression taken in itself might fail to do so. Poetry does not interfere with signified things. In creating something new, it has only its sounds to work with. Working with the sonorities of words, combining them and opposing them, poetry can make it so that in a line, nuit becomes dark and jour bright, ténèbres darker than ombre.
In contrast, prose, left to its own devices, is incapable of such transmutations. Poetic prose constitutes an exception, but only because it makes use of fragments of verse: “the prose of any sumptuous writer, ornamental, set apart from the general carelessness, is as a broken line, playing with its timbres and even covert rhymes…”[vi] Broken or intact, the line enjoys comparable powers. Mallarmé does not name these sumptuous writers; he might be thinking of Villiers de l’Isle-Adam.
Does typical prose, non-poetic prose, accomplish nothing, then? On this point, Mallarmé explains himself by way of Victor Hugo, still in “Crisis of Verse.” “Hugo, in his mysterious task, drove [rabattit] all of prose — philosophy, oratory, history — into the realm of verse, and, seeing as he was verse personified, nearly confiscated, from anyone who thought, discussed or narrated, the right to speak.”[vii] Rabattre (to drive) belongs to the vocabulary of hunting; the Littré gives the definition “to force the prey into the area occupied by hunters.” The image is thus one of a hunt. Prose, previously, was dedicated to admirable enterprises: philosophy, oratory, history. It thought, discussed, narrated, privileging sense over sound. Hugo overturned the situation, luring all that is spoken into the domain of poetry. He drives, he confiscates.
This quasi-military landscape, differing little from that of a Napoleonic War, was made possible by a particular conception of verse: “The form called verse is simply, itself, literature; … verse appears as soon as diction accentuates itself, rhythm as soon as style.”[viii] The question concerns not just Hugo himself, but also his followers, who abstained from formulating the agenda explicitly. They were not aware of it; hence the expression “unconscious idea,” though it is “majestic” as well, with regard to what it enabled: the edification of a great empire of words; the annexation of everything said in language into the realm of verse.
Mallarmé’s statement has occasionally been taken in the opposite sense.[ix] That its exact implications be restored, a more complete citation is necessary: “Monument in the desert, surrounded by silence; in a crypt, the divinity of a majestic, unconscious idea, that the form called verse is simply, itself, literature; that verse appears as soon as diction accentuates itself, rhythm as soon as style.”[x] These are not the positions of Mallarmé, but those of Hugo. It is a case of free indirect discourse; Mallarmé says nothing for himself. He exposes a doctrine that he rejects. He pays homage to Hugo, but his homage does not amount to a declaration of adherence. All the less so because he is ambivalent; Mallarmé expresses a grievance: driving all forms of prose into the domain of verse, confiscating everyone’s “right to speak,” Hugo is responsible for the desert in which he stands.
Not only does Mallarmé keep his distance, he stands in opposition. Since Hugo’s death, verse has begun to liberate itself; “we have touched the line” (on a touché au vers), the announcement rings out in Oxford on March 1st, 1894; it summarizes the decade that has just passed. Prose must also recover the right to speak. In 1885, no one can think, discuss, or narrate anything in French without running up against the language forged by Hugo; philosophy, oratory, and history must free themselves from its empire. In its every branch, the empire must be opposed.[xi] An interpretation of Flaubert is discernible here: he rebelled; he was the vanguard.
For Flaubert is implicitly present: it seems difficult not to recognize, in the description of divinity, an echo of Salammbô. Hugo, monument in the desert, is compared to an archeological site; today, the statue in the crypt is as outmoded as are the gods of Carthage. It has not always been noted that the statement at Oxford, “We have touched the line,” takes up the last sentence of the novel: “Thus died the daughter of Hamilcar, for having touched the mantle of Tanith.”[xii] Hugo’s line bears the beauty, but also the vulnerability of this invisible object, the zaïmph: “…at the same time bluish as the night, yellow as the dawn, purple as the sun, multitudinous, diaphanous, sparkling, light. It was the mantle of the goddess, the sacred zaïmph that no one was to see.”[xiii] One might note that the word zaïmph in the passage was Flaubert’s invention, just as the word ptyx was originally Mallarmé’s. An allusion from “Crisis of Verse” captures our attention: “We are witnessing, as if to mark the end of the century …, far from the public square, a trembling of the veil in the temple with significant folds and, a little, its rending.”[xiv] Is this Flaubert or the Gospels? I lean toward Carthage.
It took the genius of Flaubert to bring the ancient beliefs into the present, to conquer the great divide separating them from us. It took the penetrating mind of Mallarmé to make the unconscious idea of Hugo’s oeuvre manifest. “Monument in the desert,” “… in a crypt, the divinity,” these are the expressions of an archeologist discovering a statue devoted to some forgotten form of worship. This is how a chief excavator would speak if, by chance, he stumbled upon a colossal representation of Tanith. Neither Flaubert nor the antiquities specialist believes in the gods of Carthage; Mallarmé is not a follower of Hugo, but an adversary.
He methodically analyzes the dogma of the cult to be combatted. Verse can subsume all of prose on the condition that one accept the following axioms: (1) “verse appears as soon as diction accentuates itself”; (2) this accentuation of diction makes audible the rhythm in prose; (3) in prose there is “rhythm as soon as style.” Hugo’s reasoning may be restored even more fully if its order is reversed: consider a type of prose that is governed by a particular style; this style is endowed with a powerful rhythm; to bring this rhythm out, speech must — and need only — accentuate itself; thus a line of verse takes shape: a simple putting into voice, and not “incantatory speech.” This vocalization allows the passage from prose into verse to take place simply and easily, but in the process, prose ceases to exist independent of verse. Hence the Hugolian doctrine that summarizes all the rest: “The form called verse is simply, itself, literature.” For prose to become literature or for it to become verse are one and the same.
But to become verse does not mean to be in verse form. Style will suffice, because it suffices to accentuate the rhythm produced by style to obtain poetry. Prose no longer exists, but in the end, neither does verse. The two merge in the great ocean that is literature; literature absorbs verse and verse is incarnated in the person of Victor Hugo. Between The Legend of the Ages and Les Misérables, the language remains identical to itself because it has reached its apex; the rhythm, be it born of versification or born of style, is of the same nature; in both cases, it emerges through simple reinforcement and accentuation. From that point on, literature constitutes a unified space that becomes confused with the Hugolian empire; within it, there is free movement between prose and verse.
The point bears stressing: Mallarmé reconstitutes this agenda for the sole purpose of distancing himself from it. Consequently, he reaches the affirmative theorems of “Crisis of Verse” through a systematic reversal; each article of faith in the Hugolian doctrine must be overturned; according to Mallarmé, (1) verse need not subsume prose; (2) verse form is not coextensive with literature, nor literature with verse; (3) one does not have verse as soon as diction is accentuated. Verse requires a supplementary component comprised of calculations, symmetries, plays of sonority and, running under it all, a design to create, by means of verse, this single word that language lacks, this word whose sounds correspond with its sense. A line sketches a rhythm, certainly, but that rhythm has nothing to do with style. Style may be necessary to prose, but it is foreign to verse. Style alone, even in the hands of the greatest prose writers, can never make jour bright and nuit dark; that privilege is reserved for the versified line, and, secondarily, for the broken line contained in poetic prose.
It is true that Mallarmé played with ambiguities. When he suggests, in Music and Letters, that “verse is everything when one writes,”[xv] the hurried reader thinks Hugo has resurfaced; but writes here is to be understood in a narrow sense: all writers, admired though they may be, do not write; a philosopher, a rhetorician, a historian — do not write. In the same text, the apparent equation “style, versification so long as there is cadence,” would seem to reassert the Hugolian faith; it does not, not in the least. A severe limitation has been imposed: style is not versification unless it has a cadence. Which would imply: there are styles with no cadence. Mallarmé does not name names; it falls to the reader to find examples.
By means of verse, poetry abolishes chance. “The Mystery in Letters” is explicit on this point. In response to an article by Proust, Mallarmé, exceptionally, explains his art;[xvi] he defends himself against the accusation of obscurity, but above all he lays out theses. He defines the poetic line (vers) as a line (ligne) bound by blanks, and as the site of a victory: “When chance aligned …, having been conquered, word by word, unfailingly the blank returns.”[xvii] The blank is that of the printed page; the victory is obtained by means of sonorities and sonorities alone.
We recall that Hugo had confiscated the right to speak from “anyone who thought, discussed, or narrated.”[xviii] In other words, verse alone was able to think, discuss, or narrate. We recall as well that Mallarmé systematically reverses the Hugolian doctrine; if we follow the logic to its conclusion, we reach the unstated proposition contained therein: verse does not think, does not discuss, does not narrate. Thus imperialism is disarmed; prose recovers its freedom and, in return, verse is left to its true destiny. The implicit proposition finds a strange echo in Chapter VI of The Interpretation of Dreams, in which Freud characterizes the dream as follows: “[The dream-work] does not think, calculate, or judge at all, but limits itself to the work of transformation.”[xix] Lacan takes up the formulation and extends it to the entire unconscious: knowledge “that does not think, calculate, or judge.”[xx] With Mallarmé, one discerns, on the horizon, the hypothesis of a verse that neither thinks, nor discusses, nor narrates, but accomplishes a task: it abolishes chance.
Verse has only one means at its disposal, but it is sovereign: it resounds. This is why Mallarmé engages music, another magician of sonorities. And opera. Having recalled Wagner’s “magnificent, instrumental polyphony,” Mallarmé wonders, “Does this mean that the traditional writer of verse, he who works with the humble and sacred artifice of speech alone, will attempt, by means of this sole resource, subtly elected, to compete? Yes, with something like an opera neither accompanied nor sung, but spoken.”[xxi] One might note the insistence upon speech and the spoken: verse must be declaimed aloud, in incantation. Mallarmé continues, “With two pages and their lines of verse, I provide a surrogate, adding the accompaniment of my whole self, for the world!”[xxii] One might say the two pages announce the Roll of the Dice, but that text is not in verse — a major difference, as we shall see. The phrase “the accompaniment of my whole self” sums up the implication of the body, summoned in its entirety into the declamation of poetry.
In short, we have arrived at what one might call Mallarmé’s fundamental theorem: Verse abolishes chance only when it is disjoined from thought.
And hence, a lemma: thought speaks in prose.
From which it follows that the prose of thought does not abolish chance. And the same applies for the prose of discourse or narration. Though the novel is not mentioned, its implication is clear. It must be added that the theorem and its lemma apply to Mallarmé himself. He explicitly includes philosophy amongst the prose genres; this inclusion is not self-evident. One would be justified in recognizing within it a trace of the importance Mallarmé accorded Hegel, or at least accorded his language, such as it was transmitted to him through the Vera translation. He had, on the basis of the translation, constructed a lexicon for his doctrinal writings. In so far as they depend upon philosophy, those writings are in “non-poetic” prose; they do not partake of the prose poem in any way. Because Mallarmé wrote prose poems as well, it is necessary to establish a difference in kind between them and the doctrinal texts, however perfect in form.
Human beings speak; they speak in languages; thus, they engage with the chance that governs the encounter between sound and sense. But they also engage with the chance that governs their own encounters. The internal law organizing the language and the external law determining the coexistence of speaking beings are exactly homogenous with one another. In fact, they form but one and the same law, because there is only one chance. Whatever the context from which it arises, chance designates, purely and simply, the law of multiplicity. Language, in Mallarmé’s account, is intrinsically multiple, not only because there are multiple languages, but also because each expression in each language is born of a collision of two elements foreign to each other: sound and sense. The multiplicity begins with the Two. But the same applies to speaking beings themselves. Considered simultaneously from a single point of view, they appear to form a crowd. The crowd is an unbounded entity, just as everyday language is, and each being, in the crowd, is a random collision of a body and a breath of life. One might mix the terminologies and describe languages as a crowd of languages, each language as a crowd of expressions, each expression as a crowd of two: the random co-existence of a sound and a sense. Reciprocally, with regard to speaking beings, the crowd, in the singular, functions in the same way that language, in the singular, does with regard to expressions.
This structural equivalence, or rather, identity, explains a turn of phrase articulated but once in Mallarmé. And it has often been overlooked. It appears in Bucolic, a prose piece that might be taken as an homage to the poet’s summers in Valvins:[xxiii] “The artist and the man of letters, who goes by the unique name of poet, has no business in a space devoted to the crowd or chance.”[xxiv] The substantives crowd and chance are treated as synonyms, freely interchangeable for one another. When Mallarmé speaks of chance, he speaks of the crowd and vice versa. A crowd of spoken words or a crowd of speaking beings, a chance encounter of sound and sense or a chance encounter of passers-by.
Verse abolishes chance at the core of language; should it not abolish chance in its form as a crowd? Here we come to the question of the Book. On the basis of the notes that Mallarmé left, it appears that, indeed, the Book was conceived as a means of organizing a multiplicity.[xxv] Mallarmé speculates on the number of participants and attempts to subject that number to the necessary constraint of calculation. An abolition of the crowd should result. One is tempted to take up Lenin’s phrase — Lenin, who, in 1901, defined the revolutionary newspaper as a “collective organizer.” Similarly, one might think of the Book as an organizer by means of which a group of speaking beings who are devoted to realizing it in spoken form isolates itself in the midst of the crowd. Just as the collaboration of drafters gives birth to the revolutionary newspaper, so do the reading sessions of the participants enable the functioning of the Book. In both cases, the organizer is collective, the better to organize the collectivity. In the end, I would dare to suppose that through the repeated action of the Book, the analogue to a revolutionary party would come into being. The analogy seems more fertile to me than the ceremonial of an atheist religion, too often evoked by commentators. After all, revolution undertakes to abolish the chance that, in society, slots someone or other into the class of the powerful or the class of the poor.
In any event, two abolitions may be distinguished; verse abolishes the chance in language, and the Book, in creating an organized coexistence, abolishes the chance of the crowd.[xxvi] Could the two abolitions be connected? Yes, if the Book is in verse, but no one can determine with certainty whether the Book as Mallarmé conceived it was to be in verse or not. Thus, two abolitions of chance are to be admitted as possible: one concerns the crowd, but we have no reliable trace of the procedures involved and some doubt whether it has ever been put into action; the other concerns language: it happens through verse; it has been enacted.
The operation is simple and clear. It is an operation of resistance — against Hugo, in poetry, against the crowd in society, against Hugo as the poet of the crowd. Is this the final word from Mallarmé? Obviously not. In “Crisis of Verse,” he announces the end of an era: “Our recent phase is, if not coming to a close, taking a break, or perhaps taking stock…”[xxvii] The move away from the versified line began with Hugo’s death in 1885; everything happened as if verse had waited for this moment to break apart. Now, “the whole language is escaping”: “the whole language,” not just verse.[xxviii]
And here one has to be wary. Mallarmé, as if for the fun of it, cast his statement in obscurity. In “Crisis of Verse,” he distinguishes not two periods — before and after Hugo’s death — but three. The period Mallarmé calls “our recent phase” ends at the moment he uses the phrase: 1895. It was preceded by a previous phase, precisely the one that Hugo dominated, despite the sporadic resistance of Baudelaire or Flaubert. It will be followed by a new phase inaugurated by the Roll of the Dice. This middle phase is one of open resistance, the war of the Titans against Olympus, one might say. It is not confined to the breakdown of verse; it includes Mallarmé’s own oeuvre. And Mallarmé does not participate in the break with verse; he keeps its distance. In truth, he works primarily with the versified line and adopts the very strictest rules of the sonnet. It is during this stage that, for fifteen years, he formulates axioms, theorems, and lemmas; it is during this stage that he promotes verse disconnected from thought. It would be difficult to overemphasize this point: all of the traits that, to our eyes, define Mallarmé’s poetry, belong to what he calls “our recent phase.”
Yet, “Crisis of Verse” characterizes that phase negatively. To illustrate the process by which the line is breaking apart, the lecturer selects an example he describes as “curious”: Henri de Régnier inserted, among regular alexandrines, lines of eleven, thirteen, or fourteen syllables; Jules Laforgue “initiated us to the particular charm of the line that’s slightly off.”[xxix] Régnier is compared to an instrumentalist who hesitates, respectfully, before the keys; Laforgue, on the other hand, takes part in a “mutiny.” Mallarmé comments, “Up to now, or in either of the cases just cited, there is nothing but reserve and abandon, brought on by exhaustion at the overuse of the national cadence; whose use, like that of the flag, should remain exceptional.”[xxx] Then comes the crucial clarification: “…willful infractions or knowing dissonances call upon our sensitivities, whereas scarcely fifteen years ago, the pedant, whom we have remained, was exasperated, as if confronted with some ignorant sacrilege!”[xxxi]
In a few lines, Mallarmé lays out three points: (a) he opposes his present position to his past position; previously, he would have rejected the games of Régnier and condemned Laforgue’s mutiny; today, he finds a subtle pleasure in them; (b) he dates his past position, situating it around 1880; (c) he characterizes it as pedantic. We are witnessing something we would not have expected from Mallarmé: a self-critique. And in the same moment, the historical analysis grows clearer. We knew that the “recent phase” presented two roads, the breakdown of the versified line on the one hand and its exaltation on the other, but now we learn that there had been a struggle between the two roads. Mallarmé reveals that the war fought against Hugo did not reside on an alliance; the assailants, on the contrary, were divided amongst themselves. From a chronological point of view, the operation of breaking down verse certainly began with Hugo’s death in 1885, but Mallarmé’s operation began earlier; his forces had been deployed from 1880 and doubtless before that.
What is he alluding to, more precisely? Jacques Roubaud’s studies enable us to find out.[xxxii] It is not just an issue of the versified line; it is also the privilege accorded to the sonnet and to the adherence, within the sonnet, to strict constraints that Baudelaire, for example, had ignored. Mallarmé converted to the doctrine formulated in 1872 by Théodore de Banville in his Short Treatise on French Poetry. In 1876, he publishes “The Tomb of Edgar Poe,” his first sonnet “à la Banville.” In 1887, he corrects his 1864 sonnet “The Chastened Clown” to bring it into conformity with Banville’s rules. And then in 1895, he calls it pedantic. This judgment applies, as we shall see, to the principles that presided over the most important of his poems. It does not imply a disavowal of the poems themselves; in 1894, Mallarmé is even preparing a new and expanded edition. It will appear after his death in 1899. However, the principles of composition are rejected. “The Roll of the Dice” does not just announce a new literary form; it also entails the abandonment of the forms in which Mallarmé previously wrote.
Some might readily use the term renegade to designate those who no do not consider themselves bound to think, after years of reflection, as they thought before having reflected. The author of “Crisis of Verse” expresses himself as a renegade with regard to the poetic principles that had been his. This decisive break, at once profound and serene, will give birth to the “Roll of the Dice.”
Between “Scribbled at the Theater,” which dates from 1893, and “Crisis of Verse,” times have changed. In 1893, the second phase is still underway, though coming to an end. In 1895, the third phase has just begun. Hence the differences with regard to verse that mark the two texts. In Mallarmé’s mind, the “Roll of the Dice” must attest to a new era, not only in the poem itself, but also in the preface he gave to the journal Cosmopolis.[xxxiii] As the final expression of the Mallarméan doctrine, it bears particular import. It is organized around two positions: that traditional verse be set aside, reserved for rare occasions; and that thought belong to Poetry — on which the text concludes: “Poetry, unique source.” Far from excluding one another, prose and versification come together to found a genre — or rather, Mallarmé hopes they will found one, comparable to the genre of the symphony in music.[xxxiv] “The endeavor takes part, unexpectedly, in some specific pursuits … of our time, free verse and the prose poem.”[xxxv] The conflict of the previous phase with regard to the breakdown of verse is thus brought to a close; “unexpectedly,” because nothing about the pedantic Mallarmé would have led us to expect such a gesture on his part. I do not rule out, in a reader of Hegel, a recollection of the classic ternary: following the thesis, incarnated by Hugo, a moment of antithesis had to ensue, placed beneath the sign of multiplied contradiction; finally, we arrive at the synthesis, also multiplied.
In short: Poetry is reached by means of prose, not verse. A more complete reversal of the Hugolian faith is unimaginable. Where prose was driven into the domain of verse, verse is now driven into that of prose. In order to speak in prose rather than verse, Poetry requires outside assistance: Music. “Influence, foreign, I know, that of Music heard in concert.”[xxxvi] Verse no longer offering the possibility of its linear perfection, the “perfect line,” Music proposes the organization of the page — the score and staves — as well as a sonorous play: the rise and fall of the voice, the greater or lesser intensity of delivery. While verse is addressed exclusively to the ear, Poetry, supported by the example of the musical score, adds sight: “so that people will open their eyes.”[xxxvii] The Page is substituted for the line as the minimal element of Poetry; but, the Page, in the critical terminology, was reserved for prose — more precisely, a prose in which style reigned. To praise a page of Chateaubriand was to grasp the essence of the great prose writer and master stylist.
The chronology of “Crisis of Verse” comes into focus in the preface to “A Roll of the Dice.” As long as Victor Hugo had to be fought, verse stood at the center, as did the sonnet. Free verse and the prose poem, though respected, fell to the margins. For verse, the stronghold of the Hugolian empire, to be re-conquered, it had to be disconnected from thought and restored exclusively to the play of sonorities. Only at this price did Poetry escape from the disaster its unlimited expansion promised to bring down upon it. Henceforth, the situation is reversed in its entirety. Poetry can and must think. Such is the task of the new genre inaugurated by Mallarmé. His concluding sentence merits careful reading: Mallarmé “maintains the worship” of the old verse; to it he allocates the empire of passion and reverie; but, he adds, the time has come for other pursuits: “one might treat, preferably (as in the following) such subjects of pure and complex imagination or intellect.”[xxxviii] This is precisely what Mallarmé says he has done: the parenthetical “as in the following” refers to the poem. Two words are significant here. The first is intellect; formerly, verse did not think, but Poetry, henceforth, makes use of prose and the greatest resource thereof: thought. The second important word is preferably: Poetry not only can think, it should.
Hence the pivotal sentence: “Never will a roll of the dice abolish chance.” It is incomprehensible unless put in relation to another sentence, the one that closes the “Roll of the Dice”: “Every Thought emits a Roll of the Dice.” But what, in turn, does this sentence mean? The answer becomes clearer if we recall the fundamental theorem and its lemma. By means of which the sentence might be translated: prose is but a succession of roll of the dice after roll of the dice. And indeed, the “Roll of the Dice” is in prose, the “Roll of the Dice” thinks, the “Roll of the Dice” is a roll of the dice.
The pivot sentence bears the form of a wise maxim; it articulates a thought. Truth be told, it articulates a tautology: a roll of the dice is a mechanism designed to generate chance; it would be contradictory to imagine that it could abolish it. Mallarmé isolates the weakest possible thought. Descartes sought the minimal thought in much the same way; he found it in the Cogito, which he himself considered a tired saying; hoc tritum, he says, “this worn-out phrase.” Mallarmé could hardly confer a more favorable judgment upon his own pivot sentence. Like the Cogito, it folds back on itself; in so far as it is a thought, it casts a roll of the dice; saying that a roll of the dice does not abolish chance, it says of itself that it is a roll of the dice and that it does not abolish chance. The words that compose it obey the law governing encounters of sound and sense; as in any prose sentence, they give rise to the principal effect of prose: thought as non-abolition of chance. But unlike all the other sentences, the pivot sentence states the effect it produces. In this respect, it could be called performative, just as, according to certain logicians, the Cogito is performative. Mallarmé’s sentence announces the Cogito of a poet who can say I because he has attained impersonality and achieved his own “elocutionary disappearance.”[xxxix] When combined with the terminal sentence “Every Thought emits a Roll of the Dice,” Mallarmé’s pivot sentence might be said, more precisely, to update his transformation of the Cogito: it thinks, thus it is a roll of the dice. One is inclined to recall Lacan’s aphorism: “I think where I am not,”[xl] except that the I has been erased.
The pivot sentence contains thirteen syllables. These thirteen syllables do not merely serve to establish that the sentence is in prose. They go to greater lengths than that; they articulate a stifled line in the way that prose writers had, until just recently, been trained to do to avoid alexandrines. The Autodidact in Nausea still subjects himself to this exercise. One who begins with six syllables, it was said, should follow up with seven or five. For if the first part of the sentence bears an even number of syllables, the second should not do so as well, nor should it be too obviously shorter — elimination of four, three, two and one — or longer —elimination of eight and nine. Moreover, it would take precious little to make the pivot sentence an alexandrine; a change in verb tense from future to historic past would have done it. Just as Boileau, in The Lectern, writes in the historic past tense “A reheated dinner never was worth anything” (Qu’un dîner réchauffé ne valut jamais rien), Mallarmé could have written “A roll of the dice never abolished chance” (Un coup de dés jamais n’abolit le hasard); a particular use of the historic past that recalls the gnomic aorist of Ancient Greek would have constructed a saying that could pass for a proverb. Beneath the thirteen syllables lurks the phantom of a disfigured line. Closer to Laforgue’s mutiny than to Régnier’s games, the pivot sentence rejects the versified line that presents itself. In the second virtual hemistich’s inclusion of an extra syllable, six is sacrificed at the altar of seven. The terminal sentence counts seven words. The seven letters of Anatole return, “sparse hallucination of agony,” the poem says.[xli] In the attentive ear, a plaintive silence resounds; no sentence will abolish the chance that brought the boy’s death, and each thought repeats the blow. The number has laid the cornerstone upon the grave.
“Nothing of the memorable crisis will have taken place.”[xlii] This crisis is none other than the crisis of verse. Mallarmé retracts the name that he himself had forged. We have touched the line: it is still news, but it is no longer a current event. Verse lost the battle, but Poetry won the war. Urgent concerns have changed; abolishing the chance in language no longer figures among them. What has happened that the theater of operations should make such an about-face? Quite simply the end of the war of liberation fought against Hugo. Poetry no longer needs to impose the rigorous asceticism of the versified line upon itself; it can and must explore the new avenues that afford it thought and imagination. Hugo having been definitively conquered, there is no longer any reason for Poetry to maintain the divisions, prohibitions and ambitions that combat required.
For its ambitions have also been extinguished. The preface does not just declare a victory; it also enacts a renunciation. Verse having been left to the past, the chance in language, in the new genre, takes up its usual course. And what of the chance of the crowd? Mallarmé had, at one point, imagined arithmetical procedures to combat it; if one trusts the notes on the Book, no concern was more pressing. If Mitsou Ronat and Quentin Meillassoux are indeed correct, the calculations were put into practice in the “Roll of the Dice”[xliii]; but there they sought merely to organize the page, without any aim of organizing a collectivity. Mallarmé had invoked Wagner and Bayreuth to illustrate what one could expect from an abolition of the crowd. And while Music is present in the preface, opera is not; the concert appears instead. And the concert, quite explicitly, is a happenstance gathering: dependent upon chance and the crowd.[xliv] So be it. The pivot sentence announces the end of a plan of action.
We are back at the political nihilism that I previously identified in my commentary on the swan sonnet.[xlv] Mallarmé himself consents to it; one need read him with but little attention. In “Bucolic,” we recall that crowd and chance are identified with one another; but the context is nonetheless important: “The artist and the man of letters, who goes by the unique name of poet, has no business in a space devoted to the crowd or chance […]. But perhaps it was necessary for him to have gone there and even to have held fast; in order to return, wiser and, anywhere, to bury as if useless, his precious tribute, with the certainty of [no/some] use (avec la certitude d’aucun emploi)…”[xlvi] Is the aucun negative or affirmative here? Is the certainty that no use will ever be found for the poem, or that one day a use will emerge? By chance, the language permits both readings. Mallarmé is careful not to offer any indication that would authorize a decisive one, as his prose does not abolish the chance in language. The Book, in any case, with its capital letter, is henceforth nowhere to be found. Nowhere to be found, “buried as if useless,” so what difference does it make whether it exists or not? To which the “Roll of the Dice” adds that, in fact, it does not exist. A page has been turned.
Combining “Bucolic” with the “Roll of the Dice,” we can draw three equations. Two are explicit: thought = roll of the dice; crowd = chance. The third is implicit: roll of the dice = chance. By means of synonymic substitution, we can retranslate the pivot sentence into a number of equivalent propositions. Some are overtly anti-political or even anti-messianic: never will a thought abolish the crowd; never will a crowd abolish the crowd; never will a chance occurrence abolish chance.
Other equations come to mind. It is merely chance that a speaking being is born, chance that he dies, chance that his name — the seven letters of Anatole — is what it is. And it is chance again that he should speak this language or that, chance that he should meet his fellow human beings, chance that he should take part in this particular world. Chance just like the chance correspondence of the sets of seven. And they are merely a crowd, the congregations that existed before to him. A crowd, the groups he organizes. A crowd, the interiority that he discovers in himself. A crowd, the starry sky that he contemplates outside. The speaking being is in every respect like a roll of the dice. A roll of the dice is nothing but another name for a speaking being. And just as a roll of the dice, because it both supposes and produces chance, is incapable of abolishing it, so is the speaking being incapable of abolishing the chance of which he is a product. It is not just the Revolution, not just the Book, but also Verse that cherished vain aspirations. Poetry alone speaks the truth: in the end, the only winner is chance.
[i] All Mallarmé citations are taken from the Pléiade edition edited by Bertrand Marchal, Œuvres complètes (Paris: Gallimard, 2003), henceforth abbreviated OC.
[ii] “le hasard demeuré aux termes malgré l’artifice de leur retrempe alternée en le sens et la sonorité.” (“Crise de Vers,” OC II, 213. The piece integrates a number of texts drafted at various intervals between 1886–1896). Citations refer to published English translations when available. I have silently modified these translations throughout.
[iii] See Gérard Genette, “Le jour, la nuit,” Langages, N° 12 (1968) 28–42.
[iv] “Le vers qui de plusieurs vocables refait un mot total, neuf, étranger à la langue et comme incantatoire, achève cet isolement de la parole : niant, d’un trait souverain, le hasard demeuré aux termes …” (“Crise de Vers,” OC II, 213).
[v] “… transposer un fait de nature en sa presque disparition vibratoire selon le jeu de la parole …” (“Crise de Vers,” OC II, 213).
[vi] “toute prose d’écrivain fastueux, soustraite [au] laisser-aller en usage, ornementale, vaut en tant qu’un vers rompu, jouant avec ses timbres et encore les rimes dissimulées …” (“La Musique et les Lettres,” OC II, 64). The text was published in 1894, on the basis of a paper delivered in French at Oxford, then at Cambridge.
[vii] “Hugo, dans sa tâche mystérieuse, rabattit toute la prose, philosophie, éloquence, histoire au vers, et, comme il était le vers personnellement, il confisqua chez qui pense, discourt ou narre, presque le droit à s’énoncer” (“Crise de Vers,” OC II, 205).
[viii] “La forme appelée vers est simplement elle-même la littérature ; … vers il y a sitôt que s’accentue la diction, rythme dès que style” (“Crise de Vers,” OC II, 205).
[ix] See, for example, Jacques Rancière, Politique de la sirène (Hachette/Poche: 2006).
[x] “Monument en ce désert, avec le silence loin ; dans une crypte, la divinité ainsi d’une majestueuse idée inconsciente, à savoir que la forme appelée vers est simplement elle-même la littérature ; que vers il y a sitôt que s’accentue la diction, rythme dès que style” (“Crise de Vers,” OC II, 205).
[xi] Cf. Daniel Delas, “On a touché au vers! Note sur la fonction manifestaire du poème en prose au XIXe siècle,” Littérature, N° 39 (1980) 54–60.
[xii] “Ainsi mourut la fille d’Hamilcar pour avoir touché au manteau de Tanit” dans Gustave Flaubert, Salammbô, Oeuvres complètes I (Paris: Gallimard, 1951) 994.
[xiii] “… tout à la fois bleuâtre comme la nuit, jaune comme l’aurore, pourpre comme le soleil, nombreux, diaphane, étincelant, léger. C’était là le manteau de la Déesse, le zaïmph saint que l’on ne pouvait voir” (776).
[xiv] “On assiste, comme finale d’un siècle …, hors de la place publique, à une inquiétude du voile dans le temple avec des plis significatifs et un peu sa déchirure” (“Crise de vers,” 205). In his commentary, Bertrand Marchal leans toward the Gospels. Mallarmé was not without a sardonic sense of humor; a double allusion would not be beyond him.
[xv] “La Musique et les Lettres,” OC II, 64.
[xvi] Marcel Proust’s article “Against Obscurity” (“Contre l’obscurité”) was published in July 1896 in the Revue blanche. “The Mystery in Letters” (“Le Mystère dans les lettres”), OC II, 229–234, was published in September of the same year, in the same journal.
[xvii] “Quand s’aligna … le hasard vaincu mot par mot, indéfectiblement le blanc revient” (“Le Mystère dans les lettres,” OC II, 229–234).
[xviii] “il confisqua chez qui pense, discourt ou narre, presque le droit à s’énoncer” (“Crise de Vers,” OC II, 205).
[xix] Sigmund Freud, La science des rêves (Paris: P.U.F., 1980) 432.
[xx] “qui ne pense pas, ni ne calcule, ni ne juge” in Jacques Lacan, “Télévision,” Autres écrits (Paris: Seuil, 2001) 518.
[xxi] “Va-t-il se faire que le traditionnel écrivain de vers, celui qui s’en tient aux artifices humbles et sacrés de la parole, tente, selon sa ressource unique subtilement élue, de rivaliser ! Oui, en tant qu’un opéra sans accompagnement ni chant, mais parlé.” (“Planches et feuillets,” Crayonné au théâtre, OC II, 195). The text dates from 1893.
[xxii] “Avec deux pages et leurs vers, je supplée, puis l’accompagnement de tout moi-même, au monde!”
[xxiii] Mallarmé’s country residence.
[xxiv] “L’artiste et lettré, qui se range sous l’unique vocable de poète, n’a lui, à faire dans un lieu adonné à la foule ou hasard.” “Bucolique,” Grands faits divers, OC II, 252. The text dates from 1895. The formulation of equivalence was written in on the proofs by Mallarmé.
[xxv] “Notes en vue du Livre,” OC I, 947–1060.
[xxvi] I am leaving aside the chance one might call editorial: “An order innate to the book of verse exists … everywhere, eliminating chance” (“Une ordonnance du livre de vers … partout, élimine le hasard,” “Crise de vers,” 211); “The haphazard collection, and there are indeed such; or chance… should never be anything but simulated” (“Le précaire recueil d’inspiration diverse, c’en est fait ; ou du hasard, qui ne doit … jamais qu’être simulé” (“Planches et feuillets,” 195). The distinction between “to eliminate” and “to abolish” is to be maintained.
[xxvii] “Notre phase, récente, sinon se ferme, prend arrêt ou peut-être conscience …” (“Crise de vers,” OC II, 204). The text dates from 1895.
[xxviii] “toute la langue s’évade” (“Crise de Vers,” OC II, 205).
[xxix] “nous initia au charme certain du vers faux.”
[xxx] “Jusqu’à présent, ou dans l’un et l’autre des modèles précités, rien, que réserve et abandon, à cause de la lassitude par abus de la cadence nationale ; dont l’emploi, ainsi que celui du drapeau, doit demeurer exceptionnel.”
[xxxi] “… des infractions volontaires ou de savantes dissonances en appellent à notre délicatesse, au lieu que se fût, il y a quinze ans à peine, le pédant, que nous demeurions, exaspéré, comme devant quelque sacrilège ignare !”
[xxxii] Jacques Roubaud, “Sur la forme du sonnet mallarméen.” Cahiers de l’Association internationale des études françaises, vol. 56, N° 1 (2004) 217–229.
[xxxiii] “Observation relative au poème Un coup de Dés jamais n’abolira le Hasard,” OC I, 391–392. In the Pléaide edition by B. Marchal, the facsimile reproduction of the 1897 version published in Cosmopolis is qualified as the “pre-original edition.” It should not be taken as a replacement for Mitsou Ronat’s edition, conceived on the basis of her own research and realized typographically by Tibor Papp (Paris: Change errant/D’atelier, 1980).
[xxxiv] “Observation,” 392.
[xxxv] “La tentative participe, avec imprévu, de poursuites particulières … à notre temps, le vers libre et le poème en prose”(“Observation,” OC I, 392).
[xxxvi] “Influence, je sais étrangère, celle de la Musique entendue au concert” (“Observation”).
[xxxvii] “pour que s’ouvrent des yeux” (“Observation,” 392).
[xxxviii] “ce serait le cas de traiter, de préférence (ainsi qu’il suit) tels sujets d’imagination pure et complexe ou intellect.”
[xxxix] “disparition élocutoire,” “Crise de Vers,” 211.
[xl] “Je pense où je ne suis pas,” Jacques Lacan, “L’instance de la lettre,” Écrits I (Paris: Seuil, 1999) 515.
[xli] “hallucination éparse d’agonie” (“Coup de Dés,” 383).
[xlii] “Rien de la mémorable crise n’aura eu lieu” (“Coup de Dés,” 384).
[xliii] Mitsou Ronat, “Le « Coup de Dés »: forme fixe?” Cahiers de l’Association internationale des études francaises, N° 32 (1980) 141-147; Quentin Meillassoux, Le Nombre et la sirène (Paris: Fayard, 2011).
[xliv] “Plaisir sacré,” Offices, OC II, 237. The text dates from 1893.
[xlv] Jean-Claude Milner, Mallarmé au Tombeau (Lagrasse: Verdier, 1999).
[xlvi] “L’artiste et lettré, qui se range sous l’unique vocable de poète, n’a lui, à faire dans un lieu adonné à la foule ou hasard; […] cependant, nécessaire d’y être venu et même d’avoir tenu bon ; pour s’en retourner, docte et, n’importe où, enfouir comme inutile, précieux son tribut, avec la certitude d’aucun emploi …” (“Bucolique,” Grands faits divers, OC II, 252).