Judith Gautier, Victor Segalen, translation and printing technology

While looking for something else (as it usually happens), I ran across these notes for a talk I gave a few years ago about Judith Gautier’s pioneering translations from the Chinese. I never had the time to write them out in full, but somebody may be interested to see the train of thought. A shout-out to Steve Yao for the invitation, and to C. Bush and T. Billings, who made the day memorable as fellow panelists.

“Not Illustration: The Book as Object from Baudelaire to Pound, Via the East”

Hamilton College, April 12, 2012


[everybody is paying attention to circulation of literature, as a form of “world literature”; history of reproduction techniques is also part of this; but gets less attention. It would be reductivist to overrate, but naïve to underrate, the importance of printing in making possible communication of literary ideas between China and Europe. Here I focus on a few examples of an early stage of that communication in which the reproductive technologies are (with or without the authors’ consent) thematized.]


Victor Segalen’s Stèles is a puritanical book in the best modernist way. I mean by that that it is troubled by its having to involve subject matter. The poems often express a flight from their ostensible subjects, either into allegory, materiality, or intentional unreadability. I’ll give some examples later on. And yet they do have subjects. They have subjects in a double way. There is a theme, topic or point of departure, and there is usually an original somewhere to be found in Chinese literature and history—an original that waited a hundred years for Timothy and Chris to track it down and show how its relation to the finished poem was almost never simple, but usually ironic and twisted; something I can’t believe readers of Segalen’s poetry were contented to go without noticing for so long. So the subject matter, and the Chinese pre-text, are both needed to activate the poem, but the poem is often constituted as an act of refusal to reduce down to those elements. What is left of the poem if we follow it in its refusal, and put between parentheses the activating seed? I see two things: irony and materiality.

There’s a strand of nihilism in every good aesthete. By aesthete I mean somebody who takes seriously the making of art objects, not as a means to some other end, but as an activity in itself. Consider a stone lying out in a field, with no one paying attention to it. Now consider the stone hauled into a sculptor’s studio and made into the image of something, say Venus or Apollo. There’s been a transformation of the stone and a downgrading of its stoniness; to be a proper observer of the sculpture you have to be willing to suspend your awareness of the stone as stone, you have to see it as a medium for rendering flesh; as both stone and flesh. You don’t completely forget that the stone is stone; you want to be able to see the making of stone into flesh, the sculptor’s technique, the wobbling or pivoting that goes on in your mind as you approach the object from both sides, the material of representation and the thing it is made to represent. Then and only then you can say, “Wow, this Bernini guy can really do flesh, he can make the marble seem warm and tender like nobody else.” But this example is too traditional. It assumes that the transformation that makes material into art objects is always of a mimetic, representational kind; and the twentieth century rejects that assumption. Let’s go back and re-envision our stone and its career from quarry to museum under the aegis of non-representational art. Let’s imagine that it undergoes a minimal transformation, such as being relocated in a new place where people pay it a new kind of attention, not the attention that the farmer pays it (complaining about it, plowing around it) but, say, the attention that people have paid for hundreds of years to the stones in the Ryoan-ji Zen temple in Kyoto [SLIDE]. You are free to see in these stones a miniature version of the Isles of the Blessed, or of the cosmological Mount Meru, or a set of human personalities, but I think none of these would amount to a very Zen-like contemplation. All would have a little too much human-interest narrative wrapped around them. Both Zen and modernism are impatient with that obligatory turn to narrative, anthropomorphism, mimesis. They are focusing on something else, trying to get us to ignore the temptations of mimicry. As Roman Jakobson put it all the way back in 1919,


The emancipation of painting from elementary illusionism entails an intensive elaboration of various areas of pictorial expression. The correlations of volumes, constructive asymmetry, chromatic contrast, and texture enter the foreground of the artist’s consciousness…. Thus the realized texture no longer seeks any sort of justification for itself; it becomes autonomous, demands for itself new methods of formulation, new material. Pieces of paper begin to be pasted on the picture, sand is thrown on it. Finally, cardboard, wood, tin, and so on are used.[1]


The point is to make it harder to think of a painting as a window onto a view, or a sculpture as a stone or wooden facsimile of something it’s not. Duchamp with his readymades achieved the minimal gesture of modernism, namely putting something into the context that assigns it the role of being an art object. Gustave Flaubert, the forerunner and hero of this particular modernism, said that his ambition was to write “a book about nothing,” in order to show that “style is an absolute way of seeing.”[2] His books about nothing, in his case, turned out to be full of things, but maybe it is only for the discerning reader that the subject matter of Madame Bovary and Sentimental Education disappears into the irony of Flaubert’s style, is completely negated as content. What’s left once artists are committed to the program of not making imitations of pre-existing realities? A minimal definition of the art object would retain the quality of being an intentional object, the consequence of some will or desire to call attention to something; putting an object in a space, like the Zen temple or the gallery or even the Kirkland College student center, manifests the intention, as do other features of framing and paratext like titles, signatures, and velvet ropes.

Segalen’s poems involve their subject matter, their paper matter, the proportions of the page and the inky illustrations that we must take on faith, if we don’t read Chinese, as Chinese titles or epigraphs that somehow relate to the French text. This happens in many ways—I could be here all day giving examples, but I want you to explore and enjoy this wonderful trilingual book and find them for yourselves. One case: [SLIDE] In “Stèle de l’âme,” where Segalen reproduces an ancient stele that gives a reversed image of a regular inscription, and comments on it, along with suggesting something about the realm of spirits and the language of the dead, he is commenting on the risk of illegibility run by a bilingual text that is being presented to a mainly French-speaking, non-Chinese-speaking, audience. For when a bilingual text like this one is sent to the printers, it may happen that “passers-by exclaim, ‘Ignorant engraver, or impious oddity!’, & not seeing, they do not linger.” Of course we’re supposed to linger. Just as Segalen is exhibiting a characteristic distance from the “exoticism” that he nonetheless needs to use to hook us in, as potential readers, he is also expressing, I think, distance and similarity, a kind of suspension of metaphor and antiphrasis, in regards to the earlier collections of Chinese poetry by Judith Gautier. Let’s put Segalen out of mind for a moment and ask ourselves: what is a book of Chinese poetry translated in French supposed to look like? One subdivision of that question is: how is such a book to be produced?

Judith Gautier was the daughter of Théophile Gautier, novelist, poet, critic and literary man-about-town. He pronounced the creed of art for art’s sake in the preface to Mademoiselle de Maupin, a novel that fell afoul of the censorship a few years before Les Fleurs du Mal, which Baudelaire dedicated to him. His daughter Judith, according to a story, was growing up in utter freedom from any organized education until one day her father met a beggar on the Champs-Elysées who was a Chinese mandarin fallen on hard times. He invited the mandarin back to the house, took him on as tutor for his daughter Judith, and so she became familiar with things Chinese. The first edition of her book of Chinese poetic translations came out in 1867. Its page format has a striking resemblance to that of Segalen’s 1912 Stèles. The poetry is rendered in prose, but a prose that has its particular slowed-down, isolating rhythm. [SLIDE. This is a poem perhaps better known to you from Pound’s version. Show the differences.] Each sentence or even sentence-fragment, each relatively complete clause is broken off as a separate paragraph, and a string of these short paragraphs, usually three to eight, make up a complete utterance. What sort of utterance this is, and what kind of act the utterance of a poem is in this special bi-language, we soon discover. It is a language in which poems first name things as external objects, then discover that those external objects mirror the speaker’s state of mind; a complete circuit of metaphor is thus established, passing through the stages of perception, recognition, and confirmation. For an example, see “La Feuille sur l’eau,” on page 23 of the 1867 edition. [SLIDE] The poems don’t rhyme, therefore, in the obvious sense. But they do rhyme, semantically speaking, by the way their images work, proffering an object, turning that into a resemblance and then into an identity. They are allegories of the work that poems are supposed to do. “Oh, quel accord serein résulte de l’union des choses qui sont faites pour s’unir! Mais les choses qui sont faites pour s’unir s’unissent rarement.” (29-30) The non-reader of Chinese, reading these, can quickly come to the conclusion that the Chinese do have poets, extremely poetic poets who put together the pieces of a dispersed world unified in metaphor, and with this semantic play of difference overcome by similarity the translation overcomes the limitations of prose to make recognizable poems even in French. I imagine that Judith Gautier has in mind too the recent invention of prose poetry by Baudelaire, whose Spleen de Paris had come out just a few years earlier and had redefined what it meant for a piece of writing to be poetic. “Quel est celui de nous qui n’a pas,” asked Baudelaire, “dans ses jours d’ambition, rêvé le miracle d’une prose poétique, musicale sans rythme et sans rime, assez souple et assez heurtée pour s’adapter aux mouvements lyriques de l’âme, aux ondulations de la rêverie, aux soubresauts de la conscience?” I think this program is somewhere in the background of Gautier’s Chinese translations.

Chinese writing appears in the 1867 book only as a foreign-language supertitle. It adorns the title page and the book’s separate thematic sections. The Chinese is sometimes stylistically awkward [SLIDE], sometimes apt [SLIDE]. The translations, we’d say today, are free—they add and subtract, they rearrange, they are more adaptations than translations, if we want to be strict about it. In fact, Gautier seems to acknowledge this, as she heads her translations with the word “selon” (according to, after). She is making texts that French readers will see as poetic, where a strict, literal or scholarly translation might have persuaded the philologists but done nothing to please the literary public. The décor of these poems is full of Chineseness: plenty of emperors, empresses, dragons, and rice wine, along with the more generally available river, moon, mountain and falling leaves. The 1902 version of the book differs from the 1867 version by emphasizing more strongly the exotic trappings. Perhaps this was possible because French culture had become less narrow-minded in the past 45 years, with Indochinese colonization, international expeditions, and writers like Pierre Loti in whom Segalen was to bewail a cheap and external “exoticism”; the difference between the two editions was also a matter of printing technology, which is a kind of invisible axis connecting the literatures of East and West in this work specifically. Gautier 1902 mentions, in her preface, the special charm of ideograms, that seem to beckon to us with their meanings before we have even looked them up in the dictionary. She includes a great many more poems by a great many more poets, including the most famous of women poets, Li Qingzhao. [SLIDE] The book even includes a poem by Confucius (pp. 205-207), which I am sorry to say is not authentic. But as an object, if the 1867 book gave a chaste and spare impression of prose poetry, working its effects on the reader by the evocation of images and the cadence of brief symmetrical paragraphs, the 1902 book is lush. The pages were printed with a kind of watermark or background image that comes out only poorly on this scan; it was an imitation of Chinese and Japanese figured papers that carried latent imagery and sang backup to the poems or letters they were to bear. The poems carry the names of their authors now in Chinese as well as in transcription, and their conclusions are marked by little colophon or cul-de-lampe figures taken from Chinese books or periodicals. [SLIDE] Where the 1867 Gautier had to evoke China through a minimal set of mainly verbal techniques, the 1902 Gautier can cut-and-paste and include Chinese printed artifacts in her own artifact. One of the Li Qingzhao poems, which ends with the speaker wiping away tears as she looks across a desolate landscape, is followed by a full-page Chinese illustration [SLIDE], which, as the caption on the left margin tells us, is taken from a series of One Hundred Beauties, published by the Shanghai Tongwen newspaper and not to be sold separately. The inclusion of realia here makes it seem as if the book is intimately, directly, undeniably Chinese, although of course the filtering effect of translation and misunderstanding is also everywhere to be seen. The usual mishaps occur: Chinese characters are printed upside down [SLIDE], transliteration mistakes are made, and of course translations are always subject to debate. Some of the mistakes, however, are new and could not have been made at an earlier stage of printing technology. Recall that the few lines of Chinese included in the 1867 version were typeset from one of the few character sets available in the West, presumably that of the Imprimerie nationale. In 1902, the availability of photolithographic technology meant that a series of Chinese characters could be reproduced the way we might reproduce an image file, as a little bloc, without any analytical intelligence needing to be used to “spell” out its different elements. And because it was a block, it could get reversed. Because photography was at the base of the reproduction, other mistakes could happen that will have a familiar look to us users of xerox machines and scanners. [SLIDE]

This matters for the idea of translation. If the words of the Chinese poets had to go through the mind of a French-speaking person and be transformed into French words (and let’s remember that Gautier was scrupulous about the amount of transformation that went on: her translations are “selon”), the images of Chinese writing and drawing could be reproduced directly, mimetically, photographically, without the touch of human hands, apparently, except where a mistake reminds us that somebody was involved in a human way. The translation is accompanied by something that we could call transfer. The overlay, on multiple planes, of transfers onto translations makes the Livre de jade of 1902 a specific and peculiar multimedia object, intertwined with the histories of Chinese and Western printing.

Photo-lithography is a very Chinese technology. (Brief history of printing in China. Metal type was not ideal for the language; lithography allowed for much better and more varied materials to be reproduced, like a return to the woodblock method.)

How do we understand the enterprise of Judith Gautier as translator? You could say, there’s a divide here, and Gautier is on the premodern side of it. Illustration, exoticism. But in her work there’s also something that anticipates a different cast of mind. The typography gone wild, the collage—maybe not in her intention, but in our retrospective sense of the exploitation of the possibilities of the medium, she’s doing something that exceeds illustration. Even when she tries to do translation as mimesis and imagery as illustration.

And I think we can see Segalen’s book as a refusal to play both of the games that Gautier is playing: his Stèles will be neither a translation nor a transfer, but something else—something defined by its desire to be something else, an ironizing translation with Flaubert’s “book about nothing” close to its heart.

Pound: two tracks, words again. Canto 53. [SLIDE]

I see Segalen as a pivotal figure. He’s pivoting from the Gautier model of translation to the Pound model, a change of which the use of illustrations and printing techniques is one aspect. He’s negating the use of images as illustrations, just as he would negate the idea that Stèles is a book about China, and proposing to us something like autonomous word art.


[Just as avant-garde writing turned its back on storytelling and personal expression, somewhere between Baudelaire and Mallarmé, so too the nineteenth-century aesthete’s book, when it included visuals, did so in a way that could not be called illustration. The pioneering translations of Chinese poetry by Judith Gautier include Chinese text as part of the page. But since this text would have been unreadable for most of her readers, it becomes a second stream or track, with an uncertain relation to the French-language text it allegedly produces. Victor Segalen’s two-track page layout borrows from Gautier’s and heightens its ambiguities. The practice of translation from Chinese and Japanese, based on a constant proximity (it would be overstating it to call it a “relation”) between poetry and visual art, was of paramount importance in the transition from aestheticism to modernism, or, as we might put it, from Gautier to Segalen to Pound.]




Greenberg, Clement. 1986. The Collected Essays and Criticism, vol. 1. Ed. John O’Brian. Chicago: University of Chicago Press.


Jakobson, Roman. 1998. My Futurist Years. Ed. Bengt Jangfelt, tr. Stephan Rudy. New York: Marsilio.



[1] “Futurism” (1919), in Jakobson 1998: 147. The autonomy thesis appears in many canonical vindications of modernism, such as Clement Greenberg’s, repeated many times over a long career: “The history of avant-garde painting is that of a progressive surrender to the resistance of its medium” (“Towards a Newer Laocoon” [1940], in Greenberg 1986. 1: 34).