- Social Sciences
- African Studies
- American Studies
- Asian Studies
- Communication Sciences
- Ethnic Studies
- European Studies
- Gender Studies
- Physical Sciences
- Life Sciences
- Animal Communications
- Cell Biology
- Evolutionary Biology
- Food Science and Technology
- Human Anatomy
Fans, Silks, and Ptyx: Mallarmé and Classical Chinese Poetry
by Richard Serrano
Fans, Silks, and Ptyx: Mallarmé and Classical Chinese Poetry
Updated: December 10th, 2012
Fans, Silks, and Ptyx: Mallarmk and Classical Chinese Poetry
HIS ARTICLE TAKES as its basic assumption that Mallarmk's later poems, which baffled his contemporaries and have pro- vided untold numbers of exegetes with gainful employment, func- tion like a certain kind of classical Chinese poetry.' Before turning
Although MallarinC could not read Chinese and never visited China, it would have been i~npossible for hirn to be unaware of the ~aponismrthat gripped Paris in the 1860s and accelerated in the 1870s and 1880s, and during this time Europeans seldom distinguished Japan from China. We know that Mallarm6 moved in circles of Japano~naniacs. Mallar~ni.'~
friend Philippe Burty, the noted art critic, was not only a prominent collector ofJapanese ol~jects, but was also "a founder-member of the secret SociPfC rlu J~ng- la^, whose members did rnuch to promote enthusiasm for things Japanese," although its name sounds Inore Chinese than Japanese (Millan 191). In 1874 Mallarmi. "met and became friendly with the art critic Theodore Duret and the engraver and printer Felix Bracque~nond, both of who~n, like Manet and hiinself, were keenly interested in Japanese art" (Millan 202). Mallarm6 also counted among his friends Monet and Whistler, both heavily influenced by Japa- nese art. Indeed, Mallarm6 had his own Japanese-decorated salon in his hoine at Valvins. The question of what Mallar~ne knew specifically about Chinese language, writing and literature is more difficult to answer. The 1860s saw two inajor transla- tions of classical Chinese poetry into French. Judith Gautier, the translator of one of the volumes, was a guest, as Mme. Catulle Mend&, at Mallar~nC's Avignon resi- dence 'just after the outbreak of the Franco-Prussian War of 1870" (Lloyd 55).We imagine that they had a great deal to say to each other, since M, and Mine. Mend& had just returned from a long visit with IVagner, a figure crucial to Mallar~nC. We also know that Mallarm6 read Gautier's novel about China, Llmgoninlpkial, describing it as "double~nent cher: une grande inerveille" in an 1869 letter to Henri Cazalis (Cl, 307). At the end of 1869 Mallarm6 wrote to several of his friends indicating a desire to obtain a doctorate in linguistics, which he saw "2la fois cornine une therapie contre l'impuissance IittCraire, et cornine le fondeinent scientifique de son oeuvre" (Marcha1 128). The reply of Eugene Lefebure, an Egyptologist and professor of Egyptology in Algiers, offers a tantaliling, detailed, and accurate description of hieroglyphic writing systeins, among which he includes the "clefs chinoises" (Cl, 316). He offers to send a book by L6on de Rosny, presumably LesEcrifuresfigurafiues et hiiroglyphiqzces der rl~firenfspeuples ancirns ef morlernrs, published in 1860. IVe know that Rosny's own expertise tended toward the Far East. He was considered "the leading French authority on Japan at this tiine" and inade an extensive contribution of "books, manuscripts, prints and photographs" to the Chinese display of the 1867 Exposition Universelle in Paris (Joh~lso~l
78, 87). Unfortunately, we do not have MallarinC's reply. The only direct reference Mallarm6 makes to China is in "Las de l'amer repos," in which he describes the scene painted on a porcelain cup by a Chinese artisan.
to Mallarm6 himself, let's look at the essential building block of classical Chinese poetry, the couplet. The following two lines (translated character by character in order to make their structure obvious) are from an eight-line poem by the minor Tang poet Du Xuilhe (A.D. 846-904) :
wind warm/warms bird(s) voice (s) shatter/broken
sun high/ascends flower (s) shadow(s) thicken/layered
(Tang1334) Du Xunhe arranges the elements of a late spring scene so as to reenact the experience for the reader. Thoroughly versed in the Chinese classics, and no doubt some sort of poet himself, Du Xunhe's iilteilded reader "intuitively" acknowledges the links among the elements at hand. As Mallarmi. himself tells us, ". . . les choses existent, nous n'avons pas $ les cri.er; nous n'avons qu'k en saisir les rapports. . ." (OC, 871).3 These relationships, suggested, but not forced, by Du Xunhe, hold the poem together, just as the relationships among actual birds and flowers and so on hold the world together. The wind is warm because the sun is high. The birds' voices shatter because the wind is warm. The birds are heard but not seen because they are in the thick shadows of the tree avoiding the sun. The poet thus recaptures a moment from which he has absented himself."
The movement of the reader through the poem, however, threat- ens the stasis of the scene. As has often been noted, characters in Chinese poetry maintain a certain ambiguity-it is not always imme- diately obvious what grammatical function should be assigned to a particular character in a given poem. Consequently, with the addi- tion of each character and, eventually, each line to the world of the poem, the reader is coiltiiluously forced to reevaluate how the poem should be read. The initial reading of the first two characters of this couplet would probably assume that the second character functions adjectivally, that is, "warm" describes "windn-the wind is warm. MThen we reach the second line, we find "high" in the same
'Translations from French or Chinese are my own unless otherwise noted. Since my argument insists on the fundainental ambiguity of Mallar~nC's language, I have left his poems untranslated.
"n this article I refer to three collections of Mallarme's works: the Pleiade edi- tion of the Oruvres conlfile'tes (OC); the 1983 Oeuvres comfilites: po6sirs (OCP); and his Co~~rspondancr
' I base my argument on Stephen Owen's explanation of this couplet (94-95).
position as "warm," reenforcing this reading because it, too, seems to be acljectival-the sun is high. But when we reach the end of the second line, we must read the final character either as "thick" or "growing thicker." If the sun is high, meaning overhead, the shad- ows would not be thickened or layered, but attenuated, as the sun in that position reduces shadows to their smallest possible area. We must then return to the word "high" and reinterpret it as a verbal form, "ascending." This revision in turn means that we cannot read the final character as "growing thicker," as the rising sun would thin rather than thicken the shadows. M'e must then read the final char- acter as adjectival rather than verbal-the shadows are thick rather than thickening. The couplet is no longer static, but active, a mo- ment on the verge of transformation.
These changes in our reading of the second line also force us to return to the first line. Knowing that Chinese couplets operate on principles of parallelism and antithesis, we realize that if the second character of line two is verbal, then the second character of line one is probably verbal as well. The wind is no longer warm, but warming, again transforming stasis into change. The birds are not silenced because the wind is warm, but because the wind is growing warmer, implying that the shattering of their voices is only just accomplished. Likewise, we assume that the final character of line one is adjectival and not verbal-their voices are shattered rather than in the process of shattering, just as the shadows of flowers in line two are thickly layered rather than growing thicker. In some sense, then, the addition of the second line reverses the situation of the first line. The situation of the birds is static (though threatened- once the sun is fully overhead, the birds will no longer have so much protection from the heat and therefore may leave the tree branches) while the wind is active-changing temperature itself, rather than merely changing the situation of the birds by its tem- perature.
In the apparent absence of the poet, the characters of Du Xunhe's couplet point to each other. The reader thus finds mean- ing in the spaces among the juxtapositions of the characters, the spaces in which the relationships among them are articulated. The meanings of the characters themselves shift according to the pat- tern of the other characters around them. Until acted upon by the reader, the couplet is merely ten characters, a succession of au- tonomous word-ob-jects without obvious connection to the others, since classical Chinese poetry largely avoids the sort of connective tactics (prepositions, subordinate clauses, etc.) we expect from poetry written in Western languages. This absence of connective words encourages the reader to re-read repeatedly, constantly re- placing tentative interpretations with new, better informed readings.
As is often noted, the difficulty of St6phane MallarmC's poetry, especially that of the 1880s and 1890s, stems in part from the elimi- nation of just such connective assistance to the reader, who must therefore assume a more active engagement with the language of the poem. Steven Winspur describes this strategy as MallarmC's "call to readers to make up the constative deficiencies of the written word with their own performative force," produciilg a poem that is "a making through words that constantly redefines everything in life" (6). According to Jacques Michon, "Mallarm6 seeks to destroy the surface effects or the linearity of language" in order to force the reader to pay closer attention to the words themselves (92). As Yves Bonnefoy describes it, "MallarmC, after having, in his own way, wiped the slate clean, thought he perceived that words could en- lighten us, reveal to us. . . the precise nature of the relationships between things" (14). And Guy Delfel calls MallarmC's poetics "the struggle between poetic orderand g.mn~,tnaticalor-der" (1 56).
TWO accessible and obvious examples of MallarmC's disordering of language are found among his sonnets, including one published in 1885, the first strophe of which is
I,e vicrge, le vivace et le be1 a~~jourd'hui Va-t-il nous dkchirer avec un coup d'aile ivre Ce lac dur oublie que hante sons le givre Le transparent glacier des vols qui n'ont pas fui! (OCP, 308)
As Malcolm Bowie points out, until we reach the final word of the first line, "aujourd'hui," a noun described by the adjective "bel," we do not realize that "vierge" and "vivace" are adjectives rather than adjectival nouns. "These phantom presences" of our first, provisional, reading "have already appeared on the stage" and per- haps never completely disappear from our consciousness. "In the same way," Bowie continues, "we must wait until the third line be- fore being told that 'nous' is the indirect and not the direct object of 'dCchirerH' (10-11). It is the hard, forgotten lake that is torn, not the "we" of the poem. In other words, even our (if this "nous" does, indeed, refer to us, the readers) relatioilship to the apparent action captured in the poem remains provisional and subject to revision, like the relationships among the various elements of the late spring scene in Du Xunhe's couplet.
Seven years later Mallarm6 took this evasion of grammatical specificity to an extreme in the sonnet "A la nue accablante tu," so much so, in fact, that Tolstoy singles this poem out for condemna- tion as uiliiltelligible in What i~ Art? (OCP, 396). Throughout the first strophe, the reader searches in vain for the verb that should accompany "tu," which we at first illcorrectly identify as the secoild person informal pronoun:
A la nue accablante tu Basse de basalte et de laves A ~n&~ne
les Cchos esclaves
Par une trompe sans vertu
Quel skpulcral naufrage (tu
Le sais, kcume, ~nais y baves)
Suprsme une entre les Cpaves
Xbolit le m2t dCvPtu. (OCP, 394)
It is not merely a desire to find an interlocutor in the poem that drives us to find a properly coiljugated verb form to accompany "tu" as you, but also the uilderstaildiilg that "tun is obviously not another adjective attached to "nue," since French grammar and its insis- tence on agreement between noun and verb would force Mallarmi. to turn "tun into "tue." However baffling Mallarmi's poetry may at first seem, we are always aware that he is intent on deforming sen- tence structure, not on creating grammatically incorrect sentences. Sure enough, when we reach the first line of the second strophe, "Quel si.pulcra1 naufrage (tu," we discover the phrase to which the adjective "tu" (past participle of "taire," or "silenced") refers. As if to mock us, Mallarmi. tosses in another "tun four lines down, this time the pronoun we thought we had eilcouiltered in our first reading: "tu/Le sais, icume ..." As Lloyd Austin poiilts out, this is hardly the sole grammatical ambiguity of the poem. "Is 'basse' a noun or an adjective?" he asks (196). Does "A" have the same function in the first line as it does in the second, we might add? Is it ''inrobbed of its accent by capitalization, or is it the third person conjugation of "avoir"?
In addition, the near-total absence of punctuation-the parentheses are ironic, containing within them the only conventionally grammatical and punctuated part of the poem-reinforces the first impression that this is nothing more than a series of "apparently unrelated words" (Austin 196). This is the same feeling of interpre- tational vertigo that grips the untrained M'estern reader when first confronted by a Chinese poem simply translated word for word. Ma1larmi.s poems thus make similar demands upon the reader as do Chinese poems. Certainly, no one could accuse Mallarmi of calling attention to himself in "A la ilue accablante tu," especially since the reader's first interpretative energies are directed to finding out what the illusory tu and not je of the poem is doing. In this case, "tu" est u7z autr'e.
It is a commonplace of Mallarmi. studies that in the late 1860s the poet underwent a profouild crisis of both body and spirit that trans- formed his poetry, so that by 1892 he could write poems such as "A la nue accablante tu ..." It remains unclear exactly what caused this transformation, but we have plenty of textual evidence to demon- strate what this transformation accomplished. How did Mallarmi. come to write poems like "A la nue accablante tu ..." that reveal con- cerns, strategies and affinities that we also recognize in traditional Chinese poetry?
The famous "Sonnet en -yx" ("Ses purs ongles en dCdiant leur onyx. . ."), provides an excelleilt example of the changes Mallarmi. wrought upon his poetry. M7e have two extant versions, one dating from 1868, perhaps mid-crisis, and one from 1887. I present the later version of the poem first, saving the earlier version for later comparison:
Ses purs ongles tres haut dediant leur onyx,
L'Xngoisse, ce minuit, soutient, la~npadophore,
Maint reve vesperal brfilC par le PhCnix
Que ne recueille pas de cineraire a~nphore
Sur les credences, au salon vide: nu1 ptyx,
Xboli bibelot d'inanite sonore,
(Car le Maitre est all6 puiser des pleurs an Styx
Xvec ce seul objet dont le NCant s'lionore).
Mais proche la croisee au nord vacante, un or Xgonise selon peut-@tre le dCcor Des licornes ruant du feu contre une nixe
Elle defunte nue en le ~niroir, encor Que, dans l'ouhli ferrne par le cadre, se flxe De scintillations sitBt le septuor. (OCP, 356)
Exegetes of the poem, particularly the later version, tend to divide themselves into two camps: those who see the poem as an example of MallarmC's belief that poetry is totally removed from the "real" world and those who see the poem as evidence of MallarmC's indis- putable referentiality not only to the "real" world, but also to his world, bourgeois Paris at the end of the ilineteeilth century. Realworlders, at their most ludicrous, insist that the sonnet de- scribes a scene of some sort, such as "A corpse . . . resting at night with the presence of no human soul to disturb it" (Myrick 127). Realworlders are also most anxious to identify the object to which "ptyx" refers: a seashell (usually a conch-noted as useful for listen- ing to the sound of the ocean), the "anfractuosity of a mountain," a fold, a leaf in the bud, and so on (Noulet 184;Nelson 52; La CharitC
179)). One critic even suggests that if MallarmC thought that the word was without meaning, he should have known better (Kromer 567). The irreferentialists, for their part, declare that the "ptyx" is nothing but "pure tautology" and "residue" or a "completely empty signifier" (Burt 73; Florence 72). It might even be nothing more than "a misprint that fills out a difficult rhyme scheme" (Burt 72).
The first Chinese poem I ever learned, Feng qiao ye Do, by the rela- tively obscure ninth-century poet Zhang Ji, still inspires a similar argument eleven centuries after it was ~vritten-Mallarmi. studies have ten centuries of catching up to do. I present a reasonable para- phrase that attempts to maintain the structure and ambiguities of the original:
Night Mooring at Maple Bridge
Moon sets, birds cry-frost fills the sky,
River maples, fishermen's fires face troubled sleep.
Beyond the walls of Gusu-Cold Mountain Temple,
The midnight bell sounds to the traveler's boat. (Cheng 298) The poem is usually understood to be an expression of the poet's loi~eliiless while travelling, so troubled that he cannot sleep. But when the temple bell rings he is reminded of his conilectioil to the seemingly endless tradition of Chinese pilgrimage aid (therefore) poetry. At first glance the poem seems completely straightforward. However, inhabitants of Suzhou (the modern name for Gusu) still argue, not about the ultimate significance of the poem, which is hardly disputed, but about its referentiality. The first four charac- ters of each of the first two lines may not be mere images suggesting the landscape, but names of actual places. In other words, "moon sets" is "hloon Sets," "birds cry" is "Birds Cry," "river maples" "River Maples," and "fishermen's fires" "Fishermen's Fires." Indeed, it was still possible in 1988 (and probably still today, despite rapid urban- ization of the countryside in Jiangsu Pro.irince) to follow a branch of the Grand Canal out of the city proper, noting geographical refer- ences on the way to Cold hlountain Temple. These Chinese realworlders insist that the reader can literally retrace the journey of the poet, pointing out that there is little logic in fishermen's fires and birds crying at midnight. Furthermore, they point out, we do not read "Cold hlountain Temple" as a reference to a cold moun- tain nor "Maple Bridge" as a maple bridge. The irreferentialists, on the other hand, insist on the poet's orga- nizing vision. He does not blindly list geographic points of interest as if writing some guide for other would-be Chinese pilgrims. The
crying birds and fishermen's fires function to keep the poet awake, or at least signal that he remains so. Their very strangeness empha- sizes his feelings of being itranger. The structure of the couplets also encourages such a reading. In the first couplet, a normal image, the setting of the moon, is followed by an unsettling one, crying birds. The second couplet follows a similar pattern: the maples on the river bank are followed by fishermen's fires.' The irreferentialists seem to insist on the meaning of the poem forming from a mzrage znterne des mots, although they would not suggest that the poem is either abstract or ~rholly invented.
Such a dispute might strike Western readers as a typhoon in a teapot, but the dispute among Mallarmkists is equally futile. Com- promise among the Chinese disputants seems not impossible- there is no reason that geographic place names cannot still main- tain the evocati.ire power of the images that inspired them. It may be that the place names in fact were borrowed from the poem rather than the other way around-the real world taking on the aspects of a poetic world that had surpassed it in importance. In any case, Zhang Ji not only selected the place names, if such they are, but also determined how to order them in his poem. The incongruity of birds crying and fishermen burning fires at midnight is intentional regardless of whether Zhang Ji invented them, recorded his obser- vation of them, or merely saw them as names on a map or signpost. Similarly, Mallarmk's ptjx remains odd and arbitrary whether he made it up, saw it on his mantel, or found it in a dictionary of En- glish botanical terminology.
Returning to Mallarmk's poem as a whole, we can apply what we have learned from our comparison of the poetics of "A la nue accablante tu ..." to that of the ninth-century Chinese poets Du Xunhe and Zhang Ji. The problem of the ptjx is far more easily re- sol.ired than the puzzle of the second line in which Mallarmk de- stroys what we expect will be a "continuous syntactic sequence" by dividing the line into four syntactical units separated by commas that leave the connections among the words unclear (Bowie 7).Not until we reach the fifth line can we assign grammatical labels to the words in question and thereby revise our reading, as Austin sug- gested we do in reading "A la nue accablante nue tu ..." Is "ce minuit" a nominal phrase supposed to substitute for and equal in force "L'Angoisse"? Or is it an adverbial of time telling us that An- guish is doing somkthing at midnight? Is "lampadophore" an adjec- tival phrase referring back to "L'Angoisse" or "ce minuit" or both? Or is "lampadophore" the direct object of the verb "soutient"? Run-
' This is also a perfect example of the antithesis possible within parallelism. The irnages in the first line, birds and moon, are related to the sky; the irnages in the second are at ground level. The structure of the poem forces the reader to follo~v the organizing gaze of the poet without ever saying "I."
ning counter to the perfectly alexandrine first line, the second line is chopped into four pieces that somehow seem to stand as separate entities waiting for the reader to supply the relationships among them. Bowie feels constrained to refer to them as "four ideas" (7), almost as if they somehow stand outside language, much as Orientalists often consider characters to be somehow directly re- lated to the "ideas" they represent, rather than mediated through language. Echoing the necessary strategies of reading and re-envi- sioning "A la nue accablante tu ..." and Du Xunhe's couplet, we pro- pose provisional readings of the second line of "Ses purs ongles ..." that never completely disappear, even after we revise them to ac- count for grammatical structures later in the poem. According to Bowie, "these subsidiary metonymic connections within the line" encourage us to "see anguish as a kind of midnight and midnight as a source of illumination," even if we later determine that this read- ing was misguided given the context of the rest of the poem (Bowie
8). Mallarmi: attempted to explain this poem to his friend Henri Cazalis, insisting that the meaning, "s'il en a un . . . est evoqui. par un mirage interne des mots memes" (Cl, 178) ("if there is one. . . is evoked by an internal mirage of words themselves"). While Mallarmi. was referring to the 1868 (and not the 1887) version of the sonnet, which we will turn to shortly, it is useful here to attempt to pinpoint what Mallarmi. might have meant by the word "mirage." Marion Sugano seems to consider a mirage to be a mere fantasy without reference to the "real" world because she accuses recent exegetes of taking Mallarmi:'~ letter as an invitation to explore "the nonreferential aspects of the text" (29)." An Orientalist under- stands "mirage" differently, however: as a vision of astonishing veri- similitude that replaces a void and is often the result of a great de- sire to see something that is not there, as a thirsty man in the desert sees water, or even a lake dotted with sailboats. In other words, a mirage is a vision not of something that cannot be there, but that is not there. In a certain sense, "the nineteenth-century bourgeoisie turned the home into a palace of illusions," a mirage given form, "which encouraged total dissociation from the world immediately outside" (Forty 101). Mallarmi. pro.irides a mirage of words that all but take on the characteristics of objects because the poet dissoci- ates them from their conventional syntactical functions, even as he dissociates his "salon vide" from the fantasy of bourgeois interior decoration. In the hands of a gifted poet, the mirage can overpower the reader to the point of replacing some reality, just as ZhangJi's
" F. C. St. Aubvn quotes without attribution a "recent critic" who calls Mallarini.'~ poetry "a vacuum," where "we breathe pure nothingness thinly disguised in ghostly veils and rnagic arabesques," which strikes me as a re-orientaliration of Mallarine's "mirage interne des mots" (St. Aubyn 7).
midnight mirage at Cold Mountain Temple forces itself upon con- temporary Suzhou over a milennium after he composed it.
Indeed, Sugano is correct to insist that "what we do not have is a mere description or word imitation of a room whose objects can be identified as present or absent" (22). This is not to say, however, that we should forget what a salon of 1887 bourgeois France looked like. As practically any photograph or painting of the time will indi- cate, to one of Mallarmt's contemporaries the idea of an empty sa- lonwas nearly incomprehensible. In addition to carpets, overstuffed furniture, and wall-hangings, the "bibelot was . . . found every- where," resulting in "ubiquity and clutter that turned into bric-a- brac" (Saisselin 70). Even Edmond de Goncourt, who with his brother Jules was the self-declared arbiter of taste and fashion of Mallarmt's generation, turned up his nose at the "the period's ma- nia for bibelots and referred to it as bricobracomania, as if it were a kind of disease" (Saisselin xiv). Mallarmt's salon vide is not a repre- sentation of the "comfortable clutter" of the "middle-class domestic interior," but, if representational at all, is relentlessly ironic in its representation (Bowie 9-1 0).
The only object actually in this salon, as far as we can tell, is a mirror in a gilt frame, and its importance is not that it exists in the room, but that it reflects what can be seen through the window. The "cinkraire amphore" is not there to hold the ashes of dreams burned by the Phoenix. The ptyx is doubly not there, both "nul" and, presumably, "Aboli" as well. Not even the "Maitre" is home, having gone off with this object with which the "Nkant" decorates itself. As with "A la nue accablante tu," it is part of the ironic force of the "Sonnet en -yx" that the only phrase with an easily accessible grammatical construction is separated from the rest of the poem by parentheses. The only part of the poem grammatically unambigu- ous insists on the absence of the Master. Perhaps even more ironic is that the most fully visualized image in the poem, the most corpo- really realized is that of "L'Angoisse," an abstraction. What is most there in this poem cannot, by definition, be there or anywhere. Like the first tentative readings we continually revise as we work our way through the grammatical tangles of the poem, the absent ob-jects remain ghostly presences as we attempt to visualize the poem. The salon may be empty, but it is full of words, made of words, a mirage of words.
Before turning to the ptjx and its place in the 1887 poem, and then to its place in the 1868 poem, let's take a closer look at the relationship of words to things in Mallarmk's poetry. Although once ignored in favor of his "serious" poetry, Mallarmt's Vers de circonstance are now discussed at length by a number of articles and books. Sugano describes these brief poems as "verse inscribed on what otherwise, once used, would have been destined for the rub- bish pile: (emptied) jugs of Calvados, pebbles gathered at the beach, envelopes, or Easter egg shells," but declares the "Eventails" exempt from this category (183). Indeed, some of the eventails, poems inscribed on fans, were considered serious poetry even before discourse studies allowed literary critics to extend their interest to such verse, and Mallarm6 himself seems to have envisioned these fans differently from other objects.
But, what is the purpose of inscribing poetry upon things? F.C. St. Aubyn tells us that the '"E.i~entails"' demonstrate very well how Mallarm6 transformed the object into poetry" (81). Jean-Pierre Richard describes the material object receiving the inscription as "ballast" for the poem, providing it with material weight (345). The stone on which he writes a few lines monumentalizes what is other- wise merely light verse. Sugano, on the other hand, interprets Mallarmk's intent as precisely the opposite: writing on publicly circulated objects is a type of "graffiti" which "desolemnizes," dernonumentalizes literature, stripping it of all weight (196). Vincent Kaufmann notes that the gift "appears this way almost as if canceled by the poem, which literally doubles and replaces it, mak- ing the given object the mirror (and the guarantee) of the syn~bolic gesture engaged by the written," so that the poem neither provides nor deprives (35). To graft Sugano onto Kaufmann, the graffiti writes the object into being, replaces the object upon which it is written, and, once textualized, replaces the gesture of writing on the object as well. And this is literalljl what happens. As Charles Minahen points out, the ". . . gift itself will eventually perish (or disappear into some museum never to be seen except by the few curious enough to seek it out or who happen to stumble upon it)" (32).All that is left are the words we read in the Oeuvres cornplites- Richard's ballast, Sugano's surface below the graffiti, and Iblrfmann's mirror (Mallarmk's objects disappear yet again in these analogies of critical discourse) are gone, if that is what the object supplied. And, alas, much of the pleasure is gone as well. Page after page of clever quatrains go dull once dissociated from both their object and the gesture. Unfortunately, not only the "thing," which Mallarm6 wished to evoke without naming, as he told Jules Huret "during the famous Enquite sur l'ivolution littiraire of 1891," but also "the effect it produces" is largely lost (Austin 67-68).
Whereas rocks and empty eggshells are no longer able to main- tain a place for themselves in our imaginations, the inentails (and ptyx, as we shall see) remain vivid enough to secure a place from which they are not easily dislodged. Are some of Mallarmk's iventnils able to transform themselves into evocations of the gestures we associate with them? Let's look at the "Eventail de Madame Mallarm6":
Avec cornrne pour langagr
Kien qu'un batternent aux cieux
Le f~rtur vers se degage
Du logis tres prkcieux
Aile tout bas la courri2re
Cet eventail si c'est lui
Le m@me par qui derriere
Toi quelque rniroir a lui
Lirnpide (oil \,a redescendre
Pourchasste en chaque grain
Un peu d'invisible cendre
Seule ii me rendre chagrin)
To~bjours tel il apparaisse
Entre tes inains sans paresse. (OCP, 388)
Deryk Mossop quotes and gently dismisses St. Aubyn: "The move- ment of the fan has dislodged some ashes from Mallarmi.'~ eternal cigarette. That these ashes will settle on the mirror disturbs the poet because it means work for the wife. The sonnet thus becomes a slight but charming tribute to the dutiful wife" (Barbier 114). He also quotes A.R. Chisolm's reading of the cenclre as "the dust of the years" which will fall upon the hair of his wife when it begins to turn gray (Barbier 114). Mossop himself suggests that the poem may be some sort of "tombeau" for Mme. Mallarmi. (Barbier 114). Poor Mme. Mallarmi: goes from harried to old to dead!
"Aile" seems to be the key grammatically ambiguous word of this poem. It also appears in thirteen of the other eighteen iuentazls. It is almost always a noun:
Avec cette nile ouverte. . . (OCP, 552,111)
J'a marque cette aile d'un vers. (OCP, 553, VII)
Ce peu d'aile. . . (OCP, 553. IX)
Avec la briae de cette aile. . . (OCP, 554. XI)
Garder inon aile dans ta main. (OCP, 296, "E\~entail de Mademoiselle Mallarme")
Or the aileis not only a noun, but also the addressee of Mallarmi.'~ poem:
Xile nncienne, donne-rnoi (OCP, 552,111
Palpite,/Aile,/rnais n'arrete. . . (OCP, 554, XV)
Me, inieux que sa main, nbrite . . . (OC,P, 555, XVI)
But the aile of the inentail of the harried, gray-haired or dead Mme. Mallarmi. is not easily categorized. Li.on Cellier makes aile a verb, but even then feels compelled to ask, "Is it in the imperative, the present indicative or the subjunctif present tense?" (Barbier 130).
In other words, in "Eventail de Madame Mallarmi:" Mallarmi: trans- forms the most important metaphor linking the fan to a bird-the wing-from a noun to flight itself-a flight from tense and mode. Nevertheless, our first reading of aile as wing remains a ghost in the background as we try to determine whether aile is a command, an objective statement, or a wish. In a sense, the fan disappears even before the inentail disappears into a museum or is lost or destroyed by the passage of time. Mallarmi: strips it of even metaphorical thingness, removes it from even the context of modal certainty. Having been stripped of being, tense and mode, the inentail becomes a timeless, motiveless gesture.
Emile Noulet, quoting A. Reyes, the Spanish translator of Mallarmi., says that the inentails do not even seem to be made of words (152). But certainly Mallarmi: was aware of fans as things. Putting pen to paper fan, he could hardly be unaware that the thing existed. Indeed, we have plenty of evidence to indicate that Mallarmi. knew where fans came from-the eighteenth century:
Toutef'ois, rien ne vaudra jainais un &\,entail, riche tant qu'on voudra par sa rnonture, ou meme trks-simple, rnais presentant, avant tout, une valeur idkale. Laquelle? celle d'une peinture: ancienne, de l'kcole de Boucher, de Watteau. . . (OC, 714)
(All the same, nothing is ever worth as much as a fan, as rich as one wants due to its handle, or even very simple, but presenting above all an ideal value. And what is it? That of painting: ancient, of the school of Boucher and Mratteau. . . )
This entry from the August 1, 1874 edition of La Dernidre modewas written many years before any of the inentails, as far as we know. In fact, Mallarmi.'~ linking of the inentail to Boucher and Watteau ex- tends back to the beginning of his literary career, to his first pub- lished poem, "Placet" (1862), which refers to Boucher in its final strophe:
Nornme~-nous et Boucher sur ~111rose eventail
Me peindra, fliite en mains, endorrnant ce bercail,
Duchesse, nornmez-nous berger de vos sourires. (OCP, 110)
Here Mallarmi. sees himself as afaux shepherd painted on the back of a pink fan by Boucher himself. In the 1860s and 1870s a fan is still, even for Mallarmi:, a surface demanding illustration or Sugano's graffiti; the ob-ject has not yet become the inentail, a dy- namic interplay of word, object, and gesture.
In the 1860s, thanks to the Goncourt brothers, collectors and critics Philippe Burty and Arskne Houssaye (to whom Mallarmi: dedicated "Placet"), and others, Boucher, Watteau, and other eighteenth-century artists were rehabilitated. This was due in part, at least initially, to a nostalgia for the ancien rigxme encouraged by the supporters of the Second Empire. In keeping with the rapid bibelotisation of bourgeois culture, these same artists came to be appreciated not so much for their paintings, but for their decorative arts, which were more easily reproduced and mass-produced for bricabracomaniacs. In the eighteenth century, as "the fa.irourite artist of Madame de Pompadour," Boucher decorated "her successive residences and design[ed] everything from stage-settings and costumes to snuff-boxes and Donbonniires; he created cartoons for Beauvais and Gobelins tapestries, invented motifs for Sevres porcelain, fans, and a host of petits objets" (Souffrin-Le Breton 295). He was therefore an ideal candidate for resuscitation in the late nineteenth century.
By Mallarmk's day, Boucher was known not only for his pink- cheeked nymphs and shepherds found on "snuffboxes, watch-cases, scent bottles" and other bibelots, but by the rnass produced engrav- ings inspired by him, nearly all of which were ckinoiseries and so described (Souffrin-Le Breton 309). Perhaps the most influential of these was the "Gravure cle M. 12/larsault, cl'apris Roucker," which appeared in L'Artiste in May, 1859, and by 1864 similar ckinoiseries d'apris Boucher had appeared in the magazine on five more occa- sions. Boucher became so inextricably linked with ckinoiserie that in the ''Fete d'hiver" of the Illuminations (probably written in the early 1870s), Rimbaud refers to the "Chinoises de Boucher" (Souffrin-Le Breton 291). The critic and collector Arsene Houssaye, who suffered a great nostalgia for the eighteenth century wrote that "Boucher always remained somewhat of a Chinese" (328). We can- not say for certain when Mallarmi: became fully aware of the Oriental provenance of the fan, but by the time of his essay "Quant au livre" (1895) he had situated the fan in the "Far East" (OC, 374).
Indeed, Mallarmk's reference to the fan in "Quant au livre" re- veals how his consciousness of its history allowed him to transform the mere object into the poetic form 6ventail.
Ce que pour I'extrEine-orient, I'Espagne et de delicieux illettres, l'erentail i la difference pres que cette autre aile de papier plus vire: infinirnent et sornrnaire en son deploieinent, cache le site pollr rapporter contre les lkvres une rnuette fleur peinte coinrne le rnot intact et nu1 de la songerie par les batternents approche (OC, 374).
(So that for the Far East, Spain and some delicio~~s
illiterates, the fan, unlike this other wing of livelier paper, infinitely and hasty in its deployment, hides the site to bring to the lips a silent flower painted like the whole and empty world of the reverie brought near by wing flutterings.)
Just before this passage, Mallarme introduces the image of the "acheteuse prompte 2choisir une brochure" ("customer pronlptly choosing a brochure"), not so much to read but "afin de la placer entre ses yeux et la mer" (OC, 374) ("in order to place it between her eyes and the sea"). He implicitly compares the booklet flutter- ing before the lips of the seaside woman (deliciously illetfrie,we as- surne) to the fan of someone, presumably a woman, of the Far East and Spain. The silent flower painted on one side of the fan ap- proaches the lips like the "mot intact et nu1"-by the beating of the fan, the booklet's, metaphorical wings. Mallarmi. not only links the fan to a gesture he also associates with the act of casual reading (or the casual avoidance of reading), but also traces this gesture back through the history of the fan to the Far East. Early in the sixteenth century, the folding fan had made its way to Spain, thanks to Portugese merchant fleets trading with China and Japan.
Spain was the land of the fan par excellence. As G. Wooliscroft Rhead notes in his 1910 history of the fan, the "Spanish fan . . . is perpetually in motion, portraying the feelings and thoughts that are passing through the mind of its owner" (135). Rhead also re- peats a popular legend explaining the origins of the fan in China.
. . . at the Feast of Lanterns, where, on an occasion when the heat became particu- larly oppressive, the beautiful daughter of a mandarin took off her mask, and agi- tated it so as to fan the air into a gentle bree7e; the rest of the fair revellers were so rnuch struck with the grace of the motion that they one and all let fall their masks and followed thc example of the mandarin's daughter. (Rhead 46)
Perhaps Mallarmi. hopes that the act of casual reading/non-read- ing will be regarded as equally graceful and attractive to the pass- erby as the agitating of a mask-made-fan is to revellers of ancient China! Certainly, as editor, writer, and fictive, letter-writing abonnie of La Dernidre mode, he was not unaware of the value of ornamenta- tion to the Sin-de-siicle French woman. But what strikes us most about this passage from Rhead is that the spectators are impressed not by the fan itself nor by the mandarin's lovely daughter, but by the graceful gesture of fanning. Indeed, the iventails of the 1880s and 1890s aim to be nothing more than this gesture. Mallarmi. ef- faces the scenes from Boucher, the ckinoiserie which, in his own poetry of the 1860s, he would have portray himself. The e'ventail, at least in the form it takes in the "E.i7entail de Madame Mallarmi." and "Eventail de Mademoiselle Mallarmi.," strips the fan of chinoiserie and, by returning to it the originating gesture, makes the object again alien to the French, perhaps for the first time in centuries.
It seems that the French of the ancien re'gime re-orientalized the fan, which they had imported from the Spanish and Portugese. We turn again to Rhead's history, where he explains that to French fans were applied
the beautiful varnishes and reliefs of China and Japan ... The decoration consists of either a single subject covering the whole field of the fan, or a system of one, three or many cartouches, occasionally as Inany as twenty miniatures, enclosed in an orna- mental setting, made up of a curious mixture of Chinese diapered patterns, serni- naturalistic semi-Persian ornament, Italian arabesques, and French ornament of the character with which we are familiar in Rouen ware. . . On the handle end of the fan, 2.e. the smaller semicircle, are either one, three, or more miniatures, often imi- tation Chinese subjects. . . (Rhead, 157-158).
Obviously, the eighteenth-century collector was not overly con- cerned with the ethnological accuracy of decoration. Even Rhead hedges. What is non-European is only somewhat so-"semi-per- sian," "imitation Chinese subjects." What is European is already linked to the Orient, whether "arabesques" or French ornamenta- tion similar to that found on china.
By MallarmC's day, fans were either imported from the Far East or manufactured in France itself. In 1867 in Paris alone there were over 4,000 workers involved in the manufacture of fans. We know that Mallarme's fourteenth iventail was originally inscribed on a Japanese fan "orni: d'un motif d'oiseaux" (OCP, 559) ("decorated with a bird motif"). The eighteenth e'ventail was written "2l'encre blanche sur un Cventail de tulle noir orne d'un dessin japonais" (OCP, 560) ("in white ink on a fan of black tulle decorated with a Japanese design"). And Mallarm6 addresses the thirteenth e'ventail, "0Japonaise narquoise" (OCP, 554). What were the characteristics of a "Japanese" fan? The fourteenth iventail is presumably of Japa- nese manufacture, but the motif of birds was not necessarily Japa- nese, since "the fans intended for the 'Europe Trade' are packed with figures, and meaningless ob-jects are introduced just for the sake of filling a certain space and giving an appearance of elabora- tion"-just as meaningless objects fill up space in salons (Percival 203). The eighteenth P'7)entail appears to be of French (or at least non-Japanese) manufacture, but with a Japanese design, which is not described in any way. Mallarme's address of the thirteenth iventail as Japanese does not resolve the difficulty of determining what was French and what was Japanese about it or any particular fan-it might be of Japanese manufacture with Occidental-inspired design or of French manufacture with Oriental-inspired design. To further confuse matters, the most famous shop selling Japanese goods, including imported fans, was La Porte Chinoise (Impey 189) ! As late as the 1870s, China andJapan were seldom distinguished. What chinoise?-ieand ja$onaise?-ie have in common is that neither has much connection to the aesthetic of the country from which it is ostensibly borrowed.
In contrast to the fans intended for export, a Chinese fan in- tended for domestic use "might bear simply a few strokes so placed as to indicate a thought of the artist's mind, or even a few characters of exquisite writing placed with consummate skill in absolutely the right spot" (Percival 203). This Chinese tradition dates back to at least the Han dynasty (Ecke xxi). Mallarme's inscriptions on fans, his P'ventails, thus to some extent efface centuries of chinoiserie and reestablish the link between poem and fan. Mallarmi of the 1880s and 1890s does not linger over physical descriptions of the fans upon which his eventails are written. Ife transforms the fan from the quintessentially French fashion accessory-de-bibelotizes it, so to speak-until it is only movement, gesture. This brings us back to the empty salon of "Ses purs ongles ..." We have already noted the wit of evoking an empty room in the bourgeois home of Second Republic France. As art historian Deborah Silverman points out, "The French Japonophile's chamber with its overwhelming amount of paraphernalia was as far removed from that of an Eastern aes- thetic as had been rococo chinoiserieinteriors" (149). Like the "salon vide" of "Ses purs ongles ...," the Japanese dwelling was better known for its "starkness" (Silverman 149). Mallarmk's stripping away of this paraphernalia forces the reader to reassess the relation- ships among objects and among words, just as his restoration of ges- ture and written word to the fan forces the reader to reexamine the relationship between an object and its evocation.
We should not, however, believe that Mallarm6 intends to evoke China or Japan or any other far-off land with his P'uentnzls. To evoke the fan as Chinese or Japanese object would undermine his at- tempts to force us to see the object, or better yet, its gesture, anew. A Chinesp or Ja$an~se fan was something his contemporaries could find all over Paris-in the same store where they purchased "um- brellas, change purses, hairbrushes and toothbrushes, fans, scissors, stationery goods, combs, bracelets, ribbons," etc.-but Xlallarm6's e'uentazls are nowhere to be seen (Miller 50). Mallarmi. does not want to entangle the P'uentnzls in the complexities of a specific geo- graphical location or historical moment. Not the fan itself, which by its design is inextricably linked to the Far East, even if purchased at the Bon Marchi., but the originating gesture is what the iuentazls force us to see-Mallarmi. says "i.ventai1," meaning the "id6e meme. . . l'absente de tous bouquets" ("idea itself. . . absent from all bouquets"), if we may borrow his famous evocation of the ideal flower (OC, 368). Beyond the "oubli" where his voice has relegated "aucun contour" is where the e'v~ntnzlness is extracted from what has become a rather common object.
Another poem, an untitled sonnet, which brings up similar problems of Oriental provenance begins
De l'orient passe des Temps
Nulie Ptoffe jadis venue
Ne vaut la chevelure IILI~
(~UCde loin des bijoux tu dktcilds. (OCP, 224)
That 1868 version was later reworked in the 1880s to become:
Quelle soie aux baurnes de temps 0ti la Chim2re s'extenue
Vdut la torse et native nue
Que, hors de ton miroir, tu tends! (OCP, 306)
The first version insists on an origin distant in both time and space for the cloth that cannot compare to the "chevelure." As Laurence Porter points out, the second version condenses the "Nulle 6toffe jadis venue" into "l'orient passe." Oddly enough, the image becomes more specific while the language "adds a nuance of modal uncertainty" (Porter 44). In other words, the shift from "itoffe" to "soie" narrows the many possible Orients to one, China, since it was the sole Oriental producer of silk. On the other hand, Porter hears "soie" as "soit," and "quelle soit" as "quelle que soit." Again, as with the other late Mallarme poems we have read here, these first apparitional readings disappear as we reread, but not without leaving traces, just as the Orient of the first version of this sonnet leaves traces of itself in the silk of the second version, as the Orient leaves traces of itself in the fans of Boucher and Xlme. Xlallarmi., which in turn leave traces of themselves in Mallarmi.'~ duentcczls.
So, returning to the ptjx,how do we account for an object that is not of Oriental provenance-in fact, not of any provenance what- soever? We allow Mallarmi. himself the first word on the controversy, taken from a letter of Xlay 3,1868 to Eugene Lefkbure:
Enfin, cornme il se pourrait toutefois que rythmk par le hamac, et inspire par le laurier, je fisse un sonnet, et que je n'ai que trois rimes en ix, concerte~-vous pour m'envoyer le sens reel du mot ptyx, on m'assure qu'il n'existe dans aucune langue, ce que je prefererais de beaucoup afin de me donner le charme de le crker par la rnagie de la rime. . . (Cl, 274)
(In the end, as it may still be that given rhythm by the hammock and inspiration by the laurel, I make a sonnet, and having only three tvords rhyming in it, could you arrange to send me the real meaning of the word ptyx; I've been assured that it exists in no language, tvhich I tvould much prefer in order to enjoy the charm of creating by the magic of the rhyme. . .)
Admittedly, the tone of the letter is somewhat whimsical, but we would not want to go as far as Noulet and declare this explanation for ptjx "pure coquetterie" (184). Another poet struggling with a rhyme might write to his friends in the hope that one of them would know a meaning for a word he fears does not exist, or that one of them could suggest a rhymeword that does exist. Mallarmi., on the contrary, wants to be reassured that this word does not exist, not only in French, but also in any other language. As we have already seen, until recent decades, most critics have taken Mallarmi.'~ letter as a challenge to find the real meaning of the word, as if he had asked them to reassure him not that it does not exist, but that, indeed, it does. Rather than attempting to define the word (or declaring it ~'ithout any meaning whatsoever), however, they should have
tried to look at its effects on the rest of the words in the poem. Let's look at the 1868version of the poem, since Mallarm6 refers to it in his letter.
La Nuit approbatrice allume les onyx
De ses ongles au pur Crime, lampadophore,
DLI Soir aboli par le vespkral Phoenix
De qui la cendre n'a de cinkraire amphore
Sur des consoles, en le noir Salon: nu1 ptyx,
Insolite vaisseau d'inanitk sonore,
Car le Maitre est all6 puiser de l'eau du Styx
Avec tous ses objets dont le R&e s'honore.
Et selon la croiske au Nord mcante, un or
Nkfaste incite pour son beau cadre une rixe
Faite d'un dieu que croit emporter une nixe
En l'obscurcissement de la glace, dkcor
De l'absence, sinon que sur la glace encor
De scintillations le septuor se fixe. (OCP, 221)
The situation of the pjx, both literally and figuratively, seems to be constant in the two versions of the poem. Literally, the word ptyx occupies the final slot at the end of the fifth line and is preceded by "nul" in each version. The object ptyx, if we can make a distinction between the words on the page and the fictional world of the poem, is not within a room, black in the first version, empty in the second. It does not sit on "consoles" in the first version, not on "cr6dences" in the second version. It is an "Insolite vaisseau" in the first version, an "Aboli bibelot" in the second version-but in both versions this vaisseau/bibelot is "d'inaniti. sonore." The final difference between the ptjx of the first and second versions concerns its place among other objects. In the first version it is among "tous les objets" with which the "Rsve" decorates itself; in the second version it is "ce seul objet" with which the "N6ant" decorates itself. Mallarm6 makes some changes in the world of the poem surrounding the ptyx, but the ptjx itself and its relation to this world is little changed.
Would we know that the ptyxwas not there if Mallarm6 did not tell us so? Would we notice the ptjx unless the Master had taken it away? The ptjx as object is the ideal bibelot; the ptyx as mere rhyme is the ideal bibelotized word. It has no function but to take up space, to elicit comment-it is a conversation piece. Like the P'uentnil,the ptyx is merely a gesture made word. It might be a "cin6raire amphore," a vase for holding the ashes of vesperal dreams. It might be a vessel for holding the waters of the Styx. But its only useful function is in its uselessness: neither the ashes of the Phoenix nor the waters of the Styx are real, but the ptyx would hold them if they existed. In- stead the ptjx holds "inaniti.," nothing. Michael Riffaterre reminds us that the
bibplot is a nonfunctional object . . . and yet the ultimate filler of emptiness during periods like Mallarme's, when household esthetics prescribe that every nook and cranny be stuffed with ornaments, that every bit of space be crammed with the shape of things. (Riffaterre 18)
In a sense, it is only the shape of the ptyx that continues to exist in the poem. We do not know what that shape might be exactly, but we have some idea of the shape of its relations to other words in the poem, some idea of the shape of its relations to the other objects of the world of the poem. To evoke emptiness, Mallarmi. could not simply leave a space blank at the beginning of line five. No one would understand the function of that blank space. Instead he fills that space with a nonexistent word, then declares that the object it purports to represent is not there.
The bourgeois French home of the 1880s is so crammed with the shape of things that the shapes of these things are no longer evi- dent. Mallarm6 makes us see them again, or at least makes us see the gesture accompanying them: the empty filling of space with the shape of things. Kiffaterre also reminds us that
Knickknacks, familiar little objects like packs of cards left lying on a table, are common motifs in scenes of intimacy-they presuppose members of the house- hold sometvhere about; when no one is there, these oejects symbolize the essential continuity of living, they are silentwitnesses, the rime de~chuse~.(Kiffaterre 69)
But what are we to make of unfamiliar little objects left lying on credenzas, let alone unfamiliar objects not left lying on credenzas? The absence of the ptyx presupposes that no one is somewhere about. It would seem that the ptjx strips living of its essential conti- nuity. It witnesses nothing, offers witness of nothing. The ptjx is a thing without a soul. As a conversation piece, it can elicit only a maddening discourse that does not even know what questions to ask. The ptyx is a bibelot without an origin, without a history, with- out attachment to a far-off land the representation of which it in- tends to avoid.
If bibelotization is "the bourgeois way of apprehending and un- derstanding art," then Mallarmi. provides the ultimate but ulti- mately confounding bibelot (Saisselin 73). Marx insisted that an object's value is not obvious, that instead it "is value that converts every product into a social hieroglyphic. Later on we try to decipher the hieroglyphic, to get behind the secret of our own social prod- ucts" (cited in Apter, 1).Bibelotization is a process by which the artifact distant in either time or space (that is, either antique or exotic) has "been uprooted from its ancient sites and functions [and] stripped of its religious and social and political associations" (Saisselin 131). The deciphering of the hieroglyphic of the bibelot, the reconversion of the object into value requires a reassertion (or, at least, recognition) of its rootedness in the past or the Orient, for example. As in the case of the duentails, Mallarmi replaces the thing with its originating gesture. The point of filling one's salon with bi- belots is to invite an assignment of value to them based on their age and provenance. The interminable attempts of critics to determine just what a ptyx is and, more importantly, where Mallarmi. found it, are precisely that gesture of determining and acknowledging a bibelot's value. In the poetry of Xlallarmi, however, as in classical Chinese poetry, a word is not an object with a fixed value that needs only to be determined. Instead, the ptyx. like the Chinese character, takes on meanings, even if fundamentally meaningless, through its multiple connections to the words of the poem surrounding it.'
Apter, Emily. Feininizing theFetish: P~pchoanalj~zs in 7'urn-uf-
and ~Varhrrrrtzue Ob~es~ion
Ithaca: Cornell Cniversity Press, 1991.
Austin, Lloyd. Poetic Principles rrnd Practice: Occasional Papers on Brrudelrrir~, ~VIallarmt and Valtrp. Cambridge: Cambridge University Press, 1987. Barbier, Carl P., ed. Colloqup 12ilnllnrmt. Paris: A. G.Ni~et,1975.
Bonnefoy, Yves. "The Poetics of Mallarme." I'alek'rench Studies 54 (1977) : 9-21.
Bowie, Malcolm. 12ilallrrrmt and the Art oj Being D~jicult. Cambridge: Cambridge University Press, 1978.
Burt, Ellen. "Mallarmt's 'sonnet en px': The i-\mbiguities of Speculation." I'ale French Studies 54 (1977): 55-82.
Cheng Qianfan, ed. Gushi,[inxuan. Shanghai: Shanghai Guji Chubanshe, 1983 Delfel, Guy. LEsthtfique d~ Sftphrrne ~Mallarmt. Paris: Flammarion, 1951 Ecke, Tseng Yu-ho. Poetry 011 the fi'ind: The Art uf Cl~ine~eFoldi~~gFa~~~
frum the "bling rrnd Ch'ingDpnrrslies. Honolulu: Honolulu Academy of Arts, 1981. Florence, Penny. ~\lrrllrrnni, ~bfrrnet rrnd K~rlon: Visiral and Aural Signs and th~
Gpnprrrtiun oflbfecr,iing. Cambridge: Cambridge University Press, 1986.
Fort\, Adrian Obj~cts ujDpsz?e New Yo1 k Pantheon Books, 1986 Houssaye, hrsene. lVIe?l rrnd M%ime71 oj the Eight~enth C~ntury, Trans. i-\nonymous. New York: Redfield, 1857. Impey, Oliver. Chinozseri~:Thp Imf~crct uf Orientrrl Styles un Tl'e~tern Art rrndDecoration. London: Oxford University Press, 1977. Kaufmann, \'incent. Le lii~re et ses rrdre~se~.
Paris: Meridens Klincksieck, 1986
Kromer, Gretchen "The Redoutable PTYX " ~VIode??lLanguage ~Vutp~
86 (1971) 563-72
'Many thanks to Vincent Kaufmann, Ann Smock, and Marshall Brown for their wise comments on early versions of this article.
La Charite, Virginia A. "Mallarme and the Plastic Circumstances of the Text." Pre- Te.vt/T~.vt/Cunte.vt .Ed. Robert L. Mitchell. Columbus: Ohio State University Press, 1980.
Mallarmk, Stephane. Oeuvre5 romnpldt~s. Ed. Henri Mondor and G. Jean-Aubry. Paris: Gallirnard, 1945.
--A , Oeu7~res coinpl2t~s: poisies. Ed. Carl Paul Barbier and Charles Gordon Millail. Paris: Flammarion, 1983.
---. Corr~sf~undance[Vol.I). Ed. Henri Mondor. Paris: Gallirnard, 1959.
Michon, Jacques. ~\lallrrrini et Les Mots anglais. Montreal: Les Presses de l'universite de Montreal, 1978. Millan, Gordon. A Throru of theDice: The Life ojstiphrrn~ 12ilrrllnrmLNew York: Farrar, Strauss, Giroux, 1994. Miller, Michael B. The Bon Mrrrchb: Bourgeuis Cultur~ and the Dppcrrtment Store, 186% 1920. Princeton: Princeton Cniversity Press, 1981. Minahen, Charles D. "From 'vers de circonstance' to 'circonstance de vers': Inti- mations of a Mallarmk Fan." Ualhou~i~French
Studie~25 (1993): 27-33. Myrick, i-\rthur B. "A Note on a Sonnet of Stkphane Mallarmk." ~bfodern Language .Votes 22.4 (1907): 127. Nelson, Robert J. "Mallarme's Mirror of Art: An Explication of Ses purs ungle~ ..."
12iloder71 Language Quarterl) 20 (1959): 49-56. Noulet, Emile. I'ingtpoPm~s de Stiphcrne ,VInZlarmi. Paris: Minard, 1967. Owen, Stephen. Tm-nclitiunal Chineje Poetry and Poetics. Madison: University of Wis-
consin Press, 1985. Percival, MacIver. 7'h~Fa?1Rook. London: T. Fisher Unwin, Ltd., 1920. Porter, Laurence M. Thp Crijis ufFm-enc11 Sjmboli~m. Ithaca: Cornell Cniversity Press,
1990. Rhead, G. Woolliscroft. Hi~to? uf theFnn. Philadelphia: J.B. Lippincott, 1910. Richard, Jean-Pierre. L'L'niu~rs iinaginaire de ,VIallannb Paris: Seuil, 1961. Kiffaterre, Michael. Semiotir~ ujPuetry. Bloornington: Indiana University Press, 1978. St. Aubyn, F.C. Stiphrr?1eAbfrrllarini.Boston: Twayne Publishers, 1989. Saisselin, Kkmy G. The Buurgeuz~ rrnd the Bibplot. New Brunswick: Rutgers University
Press, 1984. Silverman, Deborah L. Arl Suui~~au
i71 Fin-de-Siicle France: Politics, P~)cholugy, and Stjl~.Berkeley: University of California Press, 1989. Souffrin-Le Breton, Ellen "The Young Mallarme and the Boucher Revival."
Bnudelrrire, 1Mallrrrini, VrrlPry: Neru EJJ~~~J
in Honour of Lloyd Austzn. Ed. Malcolm Bolrie, Alison Fairlie and Alison Finch. Cambridge: Cambridge University Press, 1982.
Sugano, Marian Zwerling. The Poptics uf th~ Occation: !Mrrllrrrmi rrnrl the Poet? of Civ cumstance. Stanford: Stanford University Press, 1992. Tang Gaocai, ed. Trr71g~hiJirrnshrrng Czdian. Shanghai: Shanghai Cishu Chubanshe, 1989. 1354. Winspur, Steven. "Mallarmk of Our Times." DalhousieF~ench Studies 25 (1993): iii-vii.