MICHAEL EMMERICH
What is translation? Translation is an English word. Translation is, moreover, a somewhat peculiar English word. Peculiar, I would suggest, because it is incapable of definition. It is impossible to define because it is a sort of node—a point of intersection. Translation, used in the most ordinary of its many senses, refers to something that takes place, or at least seems to take place, between two languages. The English word “translation” has meaning only because we know there are other languages besides English from which one might translate into English, or into which one might translate from English. The word “translation” implies, that is to say, the existence of other languages. But it also indicates that other languages can be connected to English: it points to itself as the bridge, the carrying across that occurs between languages. We might say, then, that the word only means anything at all because it can itself cross its own bridge by being translated into other languages. “Translation” is defined, first and foremost, by its own translatability. Saussure would say that “translation” is defined by its difference from, for instance, “interpretation,” “adaptation,” “transnation,” et cetera. But that is only part of the picture. Translation might also be defined by its difference from the French word traduction. But that is still only part of the picture. More important is the fact that in order to mean what it does, “translation” must also be a translation of traduction, just as traduction is a translation of “translation.” It is this convergence that defines “translation.”
This is fairly easy to grasp when we are talking about French and English, in which the two words are used, as far as I know, in relatively similar ways—though the range of the French word is considerably narrower than the English one. French and English are, after all, similar in many ways: both languages use the Latin alphabet, for instance. But what if we bring in Japanese, about which I know somewhat more? How would we translate “translation” into Japanese? That would depend on the sort of translation we were talking about—even if we limit ourselves to the ordinary sense of the word with which we are presently concerned. And of course “translation” is and has been used in many ways that have very little to do with the ordinary sense of the word. It is also used to describe the movement of living bishops and the relics of dead saints, for instance, and people can be “translated” to heaven, as we see in this entry from Ambrose Bierce’s Devil’s Dictionary: “Gallows, n. A stage for the performance of miracle plays, in which the leading actor is translated to heaven.”
Setting these unusual usages aside, the most obvious Japanese translation of the English word “translation” would be 
hon’yaku. But the obviousness of this translation is misleading: it comes to mind first, I would suggest, not because it is a general category like “translation” within which other types of translation are included, but because it is the most nondescript, or the least specific in a series of terms denoting various sorts of translation. “Translation” in English is an overarching category that includes all sorts of translations, the act as well as the product of the act; hon’yaku can be used in a way that makes it seem like an overarching term—it can refer both to translation as an act and to a translation of a book, and is used to translate the “translation” in “translation studies”—but it isn’t exactly, at least not in the way that “translation” is. This is evident, for instance, in the fact that 
gendaigoyaku (the rendering of a work in a premodern form of Japanese into a modern form of Japanese, which is unquestionably a form of “translation”) is not generally considered a subset of hon’yaku. Hon’yaku also has considerably less of the ambiguously theoretical or metaphorical flexibility of the English term: one might classify transliteration as a subset of translation (indeed, Jerome J. McGann uses the term “type-translation” to refer to transliteration), but in Japanese one would simply be using the wrong word for the activity variously known as 
honkoku, 
honji, or 
hon’in. Hon’yaku refers specifically to translation from foreign (non-Japanese) languages into Japanese (or vice versa), sometimes more specifically still to translations from Europe or the United States, and its usefulness as a general term is thus limited. Those like myself who attempt to translate “translation” with the word hon’yaku are, in other words, subtly carrying out the type of translation (if it is a type of translation) known in Japanese as 
goyaku, or “mistranslation.”
Japanese has another word, 
yaku, that might seem at first to serve as a general term with a theoretical/metaphorical inclusiveness similar to that of the English word “translation.” Yaku appears in numerous multicharacter compounds that correspond to different types of “translation,” including hon’yaku and gendaigoyaku, and can be used to form neologisms: I have seen a recent manga translation of Genji monogatari (The Tale of Genji) by the artist Egawa Tatsuya described more than once as an 
eyaku (“pictorial translation”). But in fact it is only the Sino-Japanese character that has this general meaning; the word yaku itself refers to specific translations, as in “the Egawa translation of Genji,” 
Genji no Egawa-yaku. The general meaning of the character yaku, moreover—its theoretical/metaphorical inclusiveness—only arises through its use in compounds, and is thus limited by its uses. And so once again we find ourselves having to ask, in order to translate “translation” into Japanese, what particular variety of “translation” we are talking about. It will be useful, I think, to pause and consider a sampling of the answers we might give, if only as an exercise.
What, then, is “translation” in Japanese? If the translation we are discussing is complete, we might call it a 
zen’yaku or a 
kan’yaku. If a translator completes a translation, we might describe that instance of the act of translation with the verb 
yakuryō suru. If her completed translation is an excerpt, it is a 
shōyaku. A first translation is a 
shoyaku. A retranslation is a 
kaiyaku, and the new translation is a 
shin’yaku that replaces the old translation, or 
kyūyaku. A translation of a translation is a 
jūyaku. A standard translation that seems unlikely to be replaced is a 
teiyaku; equally unlikely to be replaced is a 
meiyaku, or “celebrated translation.” When a celebrated translator speaks of her own work, she may disparage it as 
setsuyaku, “clumsy translation,” i.e. “my own translation,” which is not to be confused with a genuinely bad translation, disparaged as a 
dayaku or an 
akuyaku. A cotranslation is a 
kyōyaku or 
gōyaku; a draft translation, or 
shitayaku, may be polished through a process of “supervising translation,” or 
kan’yaku, without it becoming a kyōyaku or gōyaku. Translations are given different names depending on the approach they take to the original: they can be 
chokuyaku (literally “direct translation”), 
chikugoyaku (“word for word translation”), 
iyaku (“sense translation”), 
taiyaku (“translation presented with the original text on facing pages”), or in the case of translations of works by Sidney Sheldon, Danielle Steel, John Grisham, and other popular American writers, 
chōyaku (“translations that are even better than the originals,” an invention and registered trademark of the Academy Press). When what has been translated is a word, the translation is a 
yakugo; a translation of a poem is a 
yakushi; if it is a lyric it is a 
yakushi; if you are discussing a translation as prose you say 
yakubun; if the translation is a book it is a 
yakusho or a 
yakuhon, and the translated title is its 
yakumei. When you translate as a mode of reading, you 
yakudoku suru; when you translate in order to clarify the meaning of a text, you 
yakkai suru; when you translate aloud, you 
yakujutsu suru. A Braille translation is a 
ten’yaku.
These examples should suffice to make my point. In order for “translation” to have any meaning at all, it must be translatable into other languages, but the moment it is translated, it is swept up in a system of differentiations different from the one in which it is enmeshed in English—indeed, it doesn’t even have to be translated, because the word itself implies its own connectedness to these other systems of differentiation. Translation must be viewed as a node within which all the ideas of translation in all the languages there ever have been or could ever be might potentially congregate, intersect, mingle. Or we could say that the word “translation” is haunted by all the concepts it might translate, the words with which it may be translated. A word like “dog” can be understood, if only provisionally, in terms of its difference from an (indefinite) string of other terms in English; “translation” is made doubly provisional by its inevitable connection to other, non-English ideas of translation that could, at any moment, be brought to bear on the English word, just as the English word can be brought to bear, through a subtle process of productive mistranslation, on the Japanese word hon’yaku. “Translation” is, that is to say, always waiting to be redefined, not through its difference but through its similarity to other terms in other languages.
If this is not as obvious as it probably should be, it is because too frequently we consider translation from a perspective that has nothing to do with translation. We focus almost exclusively on translations. There are originals and translations, source texts and languages and target texts and languages, domestic and foreign, those who commission translations and those who consume them—everything but translators engaged in the act of translation. There are several reasons for this. The most important, perhaps, is that it is difficult to get a handle on what exactly a translator is doing when she translates. Consider this description of the process by Donald Philippi:
Whatever happens after a translator sits down at the computer, it isn’t anything material. What realm do we enter when we boot up our computer, attune our mental faculties to that odd wavelength of ours, and ascend into the ethereal realm of the translator’s daily praxis? The translator’s consciousness is not focused on any object, but is rather liberated from the world of material objects. The translator’s realm is on a highly abstract plane, rather like that of a mathematician, grammarian or logician. The material objects are distanced. The domain of consciousness in which the translator operates is detached from the whole natural world. Abstracted from reality, the translator operates outside the spatio-temporal system in the world of pure consciousness. As Edmund Husserl would say: “Between the meanings of consciousness and reality yawns a veritable abyss. (Husserl, Ideas, p. 138.) [ … ] The translator’s world is a world of incorporeal experiences based on contact with non-material relationships and concepts. The habit of dealing with these incorporeal substances gives translators a good ability to attain high degrees of abstraction and to intuitively perceive relationships which are not obvious on the surface. Ghostly relationships are moving around almost imperceptibly in the ether; it is our task to identify and catch them, pin them down, then radically demolish them and reassemble them into an equivalent in the target language.1
This is a brilliant description of the experience of translating between “typologically diverse languages,” but it is also alien to the everyday life of anyone but an experienced translator, or perhaps even to the everyday life of the translator, as the opening words of the passage suggest. Translation as an act is itself so foreign, you might say, that we feel compelled to domesticate it. We accomplish this through metaphors, by anchoring translation firmly in the “world of material objects,” “the spatio-temporal system” outside of which the translator operates at the moment she is translating, speaking about translation as something that takes place between two languages, two cultures, two nations.
Translation comes from the Latin word translatus, the past participle of transferre, which might be translated as “carried across.” We speak of translating from one language into another language, and translation is often described as a “bridge” between languages, cultures, nations. Both the notion of translation as something that takes place in an “in-between” place and the particular metaphor of the bridge are so common, and cleave so well both to the etymology of the word “translation” itself and to the spatial metaphors implicit in the language we use when we speak of translation (again: we translate from one language into another), that at times it seems almost impossible to think of translation in any other way. And this mode of thinking about translation is, indeed, ubiquitous: it figures in translators’ discussions of translation, in the pleasantly optimistic advertisements of translation agencies, and in theoretically sophisticated treatments of translation of the sort one might read for a course on translation studies. I’ll give you a few examples of what I mean, from writers who are using the metaphor to very good effect in very different ways.
Translation is not the transfer of a detachable “meaning” from one language to another. It is a dialogue between two languages. It takes place in a space between two languages. And most often also between two historical moments. Much of the real value of translation as an art comes from that unique situation. It is not exclusively the language of arrival or the time of the translator and reader that should be privileged. We all know, in the case of War and Peace, that we are reading a nineteenth-century Russian novel: it should not read as if it was written yesterday in English.2
By the very nature of things translation is a bridge between two languages, and if we speak of the problem of translation with regard to the literature of one particular language we appear to be dealing solely with either a beginning or an end, rather than with an entire process.3
Translation is a bridge between cultures.4
All these quotations, and their invocation of spatial metaphors—the idea of the translator’s or the translation’s in-betweenness, the suggestion that translators and translations serve as bridges—dovetail neatly with the metaphors we use when we talk about language and “communication” (which itself contains a spatial metaphor): “Did you get my meaning?” “Did you catch what she said?” “Am I getting through to you?” “I hope I’m conveying myself.” Translation is represented, then, in the quotations I just listed, as a version of communication that takes place between two languages, cultures, and nations, rather than between two people. But there is no analogy whatsoever. When a translator sits down at her computer to translate, she is alone. There is no communication happening. Indeed, there is no transfer of a message from one language into another, because from the perspective of the translator at the precise moment she is translating, she is not between languages, and her languages are not separate. We might say, rather, that she is saturated with two languages—that she is a node for two languages. Both languages are living inside her, in the same place, at the same time, in constantly shifting concentrations and configurations. She is not a bridge; she is something like a ghost.
And this is the antidote I would like to suggest to our habitual, perhaps unintended but nonetheless inappropriate metaphorical representation of translation as another version of communication: not a move away from metaphor, an attempt to clarify what happens when a translator translates, but a shift from the metaphor of the bridge to that of the ghost. Rather than imagine the translator as someone who stands between languages, cultures, and nations, we would do better to cultivate an image of him as a ghost who haunts languages, cultures, and nations, existing in two worlds at once but belonging fully to neither. The translator, as a ghost, is neither wholly domestic nor wholly foreign, because he is simultaneously both foreign and domestic; she is neither entirely visible nor entirely invisible to those who stand in one world or the other, even in the finished form of her product, because she is in their world but not of it. The translator, as a ghost, sees languages not as discrete, autonomous, unproblematically present unities but as—what else?—ghostly signs or echoes of each other. I began by noting that the word “translation” is always haunted by other words in other languages that could be used to translate it, or for which it might itself be used as a translation, but of course those “other words” are most clearly visible, clustering around, merging and separating, to the translator, for whom the “other languages” are never really “other,” or fully “mine,” just as English is never really “mine” or fully “other,” and just as English and those “other” languages are never fully “other” to each other.
Conceiving of translation as a ghostly activity, rather than as a bridge—taking as a fresh point of departure the haunted, haunting experience of being simultaneously within two languages, cultures, and nations but belonging fully to neither—does not mean cutting translation off from the world, from economics, from global or domestic politics. It means reconsidering the relationship of translation to the world, economics, global and domestic politics, and so on from the perspective of a practice that cannot readily be assimilated to dreary nationalist narratives. From the point of view of the translator, translation should never have been conceptualized as something that takes place between two languages, cultures, and nations, because that is just the opposite of what it is: translation doesn’t “take place,” it is something translators do, and it isn’t done between languages and cultures, it is done in languages, by people in whom languages and cultures merge. When translators translate, moreover, we work with and within two or more particular languages. If translation can be conceived of as a node, a ghostly act performed at a point of linguistic and cultural intersection, then the nature of the node is inevitably defined by the merging of particular languages and cultures.
Considering “translation” from the vantage of an individual translator, seeing it caught up in a flurry of different versions of itself, each one trailing its own history, allows one to settle on one’s own working definition of the word, recognizing that no definition will ever be more than provisional. One might formulate a definition specific to the particular languages a translator engages with, emphasizing problems and issues that another translator who deals with different languages might never confront, or even imagine. One might define the word in a manner relevant to a given genre—translating a novel from one language to another is not the same, after all, as translating a cookbook or subtitling a movie. Or one might acknowledge that the notion of translation as an act performed only at the confluence of different languages is itself somewhat limited, and focus instead, for instance, on forms. Translating a script for performance may well require a different approach from translating a script for publication in a literary magazine. Translating a printed book into Braille is not the same as translating a poem from Punjabi into Swedish. Translating Dante Gabriel Rossetti’s complete writings and pictures into the form of a hypermedia archive is not the same as translating 138 penciled note cards into a facsimile codex edition of Vladimir Nabokov’s The Original of Laura. And yet each of these different acts is, or could be considered, an instance of translation.
I myself define translation, very simply, as any change wrought upon a piece of writing intended to make it accessible to a new audience with particular needs or preferences. This is an extremely broad definition. But from my perspective as a scholar-translator who works with Japanese books, it seems an appropriate one. Or rather, I suspect that for all its malleability, this definition feels right to me precisely because I am a scholar-translator who works with Japanese books. The particular nature of its breadth, the direction its openness takes, betray the influence of my experiences with the languages I know, above all the two languages I translate from: Japanese and classical Japanese. Not everyone will agree that The Original of Laura or the Rossetti Archive can or should be considered translations; my engagement with Japanese literature, especially literature in classical Japanese, is a large part of the reason I do.
The Japanese language has a long history. For most of that history, writing of the sort we would now describe as “literature” circulated exclusively in handwritten copies. Woodblock printing found its way into Japan as early as the eighth century, but until the seventeenth century its use was reserved for texts deemed more valuable than mere fiction. The Tale of Genji, for instance, which was completed in the early eleventh century, survived for its first six hundred years only because members of the elite and their scribes kept transcribing it, each in their own calligraphic style. Then, at the start of the seventeenth century, fictional works, including The Tale of Genji, began to be printed and sold in woodblock editions. These new printed books could be mass produced, unlike the old labor-intensive transcriptions, but because the blocks were handcarved from manuscripts, essentially they were facsimiles of handwritten, calligraphic copies. If a person wanted to read a printed edition of The Tale of Genji or any other work, she still had to be able to read calligraphy.
This situation changed dramatically in the final decades of the nineteenth century, when moveable type rapidly supplanted woodblock printing. People soon grew accustomed to reading typeset text, and schoolchildren no longer learned to read the calligraphic forms of earlier ages, whether written by hand with a brush or printed from woodblocks. And so it became necessary to reprint early works of literature in the new form of the typeset book. In 1890, the first typeset editions of The Tale of Genji were published in Japan, and new ones have been appearing ever since. These days, apart from a tiny group of specialists, hardly anyone would ever try to read The Tale of Genji in a calligraphic form, whether handwritten or printed, for the simple reason that they couldn’t. That old, calligraphic Japanese is utterly illegible to the vast majority of modern readers, even those rare souls who have a good knowledge of classical Japanese grammar and vocabulary. In order for modern readers to read The Tale of Genji, they need to have it transcribed into a recognizable form: the familiar, typeset Japanese of novels, signs, and menus. They need, in short, to have it translated—even when they can understand the language of the original.
But just what does that mean? What are we referring to when we speak of “the language of the original”? If someone were to read a sentence from a calligraphic copy of The Tale of Genji aloud, then read the same sentence from a typeset transcription, the two readings would sound the same. And yet most readers of classical Japanese would find the calligraphic copy illegible, and have no problem whatsoever reading the transcription. The two texts are the same, then—they are written in the same language—precisely to the extent that we ignore the visual form of the writing. They can be said to be the same, and to be written in the same language, only if we agree to ignore a difference so significant as to make one text legible and the other illegible.
The tendency to conceive of language phonocentrically and in terms of grammar and syntax, independent from its material forms, has deep roots and a long history, and it is hard to shake. My experiences with modern and classical Japanese, however, have impressed upon me just how much writing matters. I find it hard to ignore not only the visual element of writing, the marks on the page, but also its broader spatial dimensions: the physicality of paper itself, and of the book. Translation, as I understand it—as I define it for myself—is not simply an act of engagement with language as grammar and syntax, as mere recorded speech; when I translate, the original to which I address myself is not a specimen of Japanese language heard in my inner ear, but a piece of Japanese writing, a collection of pages, a Japanese publication.
In 2005, I published a translation of a novel by Akasaka Mari called Vibrator. About halfway through the book, the narrator, a woman with various psychological troubles who is riding from Tokyo up to northern Japan in the cab of a truck driver she barely knows, finds herself gazing out the window into the night:
Once more I looked at the map. Route 353 went around the southern side of the mountain. The shrine must be up at the peak, and that gateway we went under earlier must have been the outermost, the one that marks the entrance. The air down at sea level had been dry, but now that we were at a higher altitude I noticed tiny crystals drifting through the air, here and there, glittering. I kept gazing out at these crystals, and then after a while little white things started appearing, mixed in among the crystals. The white things soared through the air, weaving between the crystals as they dropped. And as I moved through these white flakes, it came to seem as if I were in a tube or a tunnel or something, cut off from the me I had been previously—here I was, moving within this space. Something before me leapt into action. The wipers had come on. It felt as if the meaning of the movement had seeped out from somewhere, that was how I understood it, and then gradually, ever so gradually, I began to remember my body.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . it’s snow.
The words appeared inside my mouth as if they had come bobbing up from an ancient layer of memory. I said them again: spoke the words clearly now, aloud. “It’s snow.” But then, as I continued staring out into the snow, the word “snow” started breaking apart—it disintegrated into s n o and w and the force that had hung there between the four letters, the force that had held them all together, was gone now, and there was no way to retrieve it. I no longer understood why s n o and w had been linked in that way, or why this was the name for that white stuff falling in front of me. Oh, OK, I get it. It’s because the things linking them have disintegrated, that’s why they’re so dry and swishy and flaky, that’s why they tumble down over everything—maybe.5
This passage is fascinating, in part, precisely because it deals with the materiality of language. The word “snow” has a physical dimension, and it can be broken down into its constituent parts. The disintegration of the word “snow” into the letters “s n o and w” is bound up, moreover, with the narrator’s mental state—she is falling apart like the word, together with the word.
Japanese is not written with the Roman alphabet. When the English word “snow” falls apart, disintegrating “into s n o and w,” it falls apart differently from the Japanese word for snow. Japanese makes use of two syllabaries called hiragana and katakana and a large number of ideographs known as kanji. In Japanese, words can generally be written in any of these three forms. The word “snow,” for instance, which is pronounced “yuki,” is most often written with the kanji 
, but it can also be written in hiragana as 
or in katakana as 
. The kanji 
, being an ideograph, inevitably calls up the particular meaning “snow.” Since the hiragana and katakana forms represent only the sound “yuki,” they are not tied to a fixed meaning. In the Japanese original of Vibrator, when the narrator first notices that “it’s snow,” her thought is expressed, not in kanji, but in hiragana: 
. In English, she goes on to explain that “the word ‘snow’ started breaking apart—it disintegrated into s n o and w.” A bilingual translation of this same phrase in the original might look like this: “the word ‘
’ started breaking apart—it disintegrated into 
and 
.” Here the kanji, expressing both meaning and sound, is not itself broken down; it is first transformed into units of sound that do not in themselves have any meaning, and the narrator then loses her sense of “force that had hung there between the two syllables.” In English, a word that has both sound and meaning is itself broken down into letters that have neither sound nor meaning. The relationship between sound, writing, and meaning is different in the Japanese and the English. Language hangs together, and falls apart, differently in the two languages. And the narrator, who falls apart with her language, falls apart differently in the original and the translation. In re-creating this passage in English, then, I had to translate one system of relationships between sound, writing, and meaning into another. In order for this passage to make sense, I had to deal, not just with the words, but with the words on the page—the very thingness of language itself.
As it happens, this same passage plays directly on the material, visual form of the words on the page. Novels in Japanese are usually read from right to left, and the text is printed vertically. English, needless to say, is printed horizontally and is read from left to right. Translation from English and various Western European languages has been so important in Japan that the act of translation is often described synecdochically with the phrase “from horizontal to vertical” or, in the case of Japanese–English translation, “from vertical to horizontal.” My translation of Vibrator is an instance of the spatial, linguistic shift “from vertical to horizontal.” This is especially apparent in the passage I have been discussing. The sentence that I translated as:
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . it’s snow.
looks like this in Japanese:
In both Japanese and English, the printed sentence itself suggests not only the time that passes as the narrator, gradually beginning “to remember my body,” remembers too what the white flakes drifting through the air around the truck are—but also, visually, materially, the snowflakes themselves. In Japanese, the snow is falling. In English, the snowflakes are whipping past the window as the truck tunnels through the darkness. It occurs to me that here, in the overlapping of these two sentences, translation and original, we might discover a visual representation of the act of translation. A depiction of translation not as a crossing over—not as something that takes place in an in-between state—but as an activity that a ghostly, disembodied translator does in the unstable, shifting confluence of the languages she lives within. In a place with no dimensions, a point of intersection, fleeting as a snowflake, falling through time, through history, mysterious and unique.
1. Donald Philippi, “Translating Between Typologically Diverse Languages,” Meta 34, no. 4 (1989): 680–681.
2. Richard Pevear, quoted in “Trials of the translator,” a letter to the editor in the Times Literary Supplement, February 8, 2008.
3. Paul F. Guenther, “Faithful Ugliness or Faithless Beauty: The Translator’s Problem,” The German Quarterly 35, no. 4 (1962): 504.
5. Mari Akasaka, Vibrator, trans. Michael Emmerich (London: Faber & Faber, 2005), 72–73.
