E Bruce Brooks
University of Massachusetts, 5 Dec 2002
Nine Maxims On Translation
No lecture was presented during a visit to the Translation course regularly taught by Lucien Miller (Chinese 660 = Comp Lit 695) on Thursday 5 Dec 2002, at the University of Massachusetts at Amherst. The time available went instead to questions by students about our 1998 translation of the Analects. But a lecture might have occurred, and if it had, it would probably have included some points made in a paper given more than a decade earlier for an American Literary Translators Association (ALTA) conference, held that year at the University of Massachusetts, along with several other points encountered in practical translation experience. That lost lecture, plus some illustrations made possible by the Internet medium, might have looked something like what follows. Clicking on a bar will take you to an explanation or an example.
Not much that is useful, in a theoretical way, can be said about translation. The basic concept is simple, even though implementing that concept in a specific case can be very difficult. The best discourse on translation is not a lecture; it is a successful translation. But the following nine maxims (some of which contain several subordinate suggestions) may give useful warning of some standard pitfalls, and perhaps suggest some practicable solutions.
1. Definition. A translation is the same text in a different language. Every statement made about the work by a reader of the translation should also be true of the original text. This is not possible. Not with poetry, and not even with prose, where the difficulty is generally assumed to be less. But it is what we aim at. The real thing, with the curtain of language somehow made transparent.
2. Ethical Aspects. Don't lie. You have an obligation to the original, and to your audience. Each of them is relying on you for what it knows about the other. Disclaimers in prefaces don't count; nobody reads prefaces. The audience will invariably take the translation as the real thing. Since it can't be the real thing in all respects, the translator's first duty is to rank the importance of the original's many properties, and put major effort on the properties at the top of that list. One of the points at which this issue comes to a head, and where the question of essential versus dispensable features gets asked, is the question of rhyme in translations of poetry.
3. Your Equipment. There are preliminary steps, before you actually sit down to work on your text. Some of them can involve much time and thought. The perfect translator does not exist, but one can try to be a reputable translator. Here are some background suggestions:
- 3a. Consider Your Temperament. Translators are experts at language, but they don't turn that expertise to their own creative account. They deal with literary works of high originality, but they are not entitled to insert their own originality into those works. They have the skill and sensitivity to be writers, but they are not writers; they are witnesses to other people's writing. There is a personality to which these demands, plus and minus, are congenial, and that personality defines the natural translator. Compare the literarily skilled person whose talents come out best in analyzing or anthologizing other people's work, or the musically trained person who is not a composer, but brings great insight to the task of realizing other people's compositions by performing them before an audience.
- 3b. Learn Your Language. If you haven't read widely in your own literature, you don't really know what your language is capable of. If there are pages of your own dictionary that you haven't visited, you don't really know your own language. If you work in rhyme, you will additionally have to reorder the wordstock in your head by rhyme groups, like a Tang examination candidate. You can afford to labor over a special passage, now and then, but in general, you have to work fast, and speed requires skill. If you have to thumb the thesaurus too often, you don't have skill. If you can't deliver the final result when you said you would, you aren't professional.
- 3c. Prepare Your Audience. What a translation has to do is a function of what its intended audience knows. A culturally prepared audience knows the background of the work, and the things the work alludes to, and is merely waiting (as was the original audience) for acquaintance with the work itself. Such an audience does not require the translation to carry information that the original work was not responsible for. To achieve that situation in the second language may need social background work, before the literary work can begin. That social work may take generations; it may require institutional effort and foundational funding. Meanwhile, do your best with the audience you have. Given that audience, ask what burden of self-commentary the translation will have to carry, and think how best to distribute that burden.
4. Your Text. Much of the literature of translation is not about errors in translation; it is about errors in understanding the original. The task of equipping yourself with the source language, and with the source culture in which that language operates, is here taken for granted. But even with that equipment, it is easy to miss the salient features of a particular text. Here are some precautions:
- 4a. Consider the End. Many poems only disclose what they are about in their last lines (Horace, Carmina 1:34). Read to the end before starting work. Some prose traditions like to cadence abruptly, with little formal preparation. Be ready to help your readers over these sudden culminations, which may not work well in their new context.
- 4b. Locate the Peaks. As Waley observes, there are some passages that just have to be gotten right, in order to carry the climactic weight that the work puts on them. Start at such a climax and work outward from it. Sometimes the high spots, in a poem or in an author's whole work, are not reached by formal analysis, but are chosen by a whole tradition. Either way, here are some essential parts of the task.
- 4c. Calibrate the Beginning. The end is crucial, but the beginning is also worthy of notice. It is often in the first few lines that a work signals its intentions (or artfully conceals them, while still inviting the reader's further attention). The first few words of the Mencius text establish the moral earnestness in Mencius's argument for a humane and yet successful government policy. The first few bars of the Heike Monogatari evoke the mood of impermanence against which the violent action to come is projected. The first few lines of Sterne's Sentimental Journey define an air of nonchalance which will persist thematically to the end. Translations too need to get off on the right foot. A defective beginning cannot be repaired later on. The reader's framework of understanding has already been set.
5. Getting Nuances. Beyond the words and linguistic problems of the text ("as for" is not a durably viable way of rendering topic phrases) are elements of more general atmosphere and contrast. Don't focus on the words to the detriment of these elements. For instance:
- 5a. Tone. Literary masterpieces are more ambiguous in tone than the general reader may expect. Features such as sarcasm are notoriously easy to overlook, even within the source culture. Some Soviet writers who did not simply go underground survived aboveground by comprehensive indirectness and obliquity, with one level for the censor and another for the reader. The right tone, or the right counterpointing of more than one tone, needs to be represented in the final product. Don't simplify it out. Don't play the censor. Nor it it your job to rewrite the original as though the censor of that time, or the controlling sensibility of that time, had never existed. You and the author are stuck in the same corner.
- 5b. Level. Don't render everything into a uniform "voice." Maugham noted that Henry James "got the sound of his own voice into every line he wrote." It was a criticism. Don't always colloquialize; don't invariably exalt. A citizen of the source culture, listening to your version, should recognize the social separations and linguistic distinctions that are part of the original. Don't transpose. If you do transpose (in recasting "Carmen" as "West Side Story"), label it as a transposition.
- 5c. Relationships. A poem which alludes to another poem should be similarly linked in your translation. Several poems close to each other in the original literature should be analogously close in translation. And prose can be as elaborately allusive, as tightly interconnected, as poetry (the machine translation people were disappointed to find that Russian physicists liked to allude to Pushkin in the technical articles; it messed up their hope of a simple semantic matching process). Getting these things right verges on the virtuoso level. But it turns out that the virtuoso level constantly obtrudes, in the translation business. Since you can't avoid it, try to get good at it.
6. Avoiding Fallacies. Translation is one of those topics that attract theories. Most theories, whatever their value as ideas, don't help the practical translator very much. Some of them are downright harmful. Here are three to watch out for:
- 6a. Literal Translation. This concept was refuted in 1975. What is called literal translation is usually the raw material for a first reading, as assembled by someone who doesn't know the language of the text very well. The process is inevitable and thus pardonable in a beginner. It is not to be enshrined as a model of the finished product which the professional should aim at.
- 6b. Neutral Style. Machine translation theory notwithstanding, there is no such thing as a neutral style. Everything has a style. The telephone book has a style, and a most prosodically demanding one it is. Departures from it, by forgers interpolating names into the telephone book, will be instantly detected. Find the style. When the style is found, write in the style.
- 6c. Charity Issues. Be an ambassador, but not an advocate. Don't speak for the work more warmly than the work merits. The "principle of charity" asks us to decide all questions in the work's favor. No; don't. If the work has gaps or lapses, leave them intact (Analects 15:26). If it has internal inconsistencies, don't smooth them out. A criticism which can be fairly made of the original should also be true of the translation. Your readers should be able to see what the fuss was about. The fuss is part of the piece, as the piece continued to exist and to attract critical attention in its home language. Be sure that the issues are also visible to your readers. Deal them in.
7. Final Revision. For simultaneous translation, the most nerve-wracking of specialties, there is no tomorrow. But if you work in a more leisurely mode, put your work aside when you think it is finished, and go back to it when the passions of discovery and the enthusiasms of ingenuity have had time to cool. Coming to it from outside, a week later, see if it still looks good, reads good (aloud, of course), and intangibly satisfies your remembered sense of the original. This is your last chance to let the original break through your idea of it, and to educate you about itself. Revision brings to a head some of the paradoxes of translation. It makes us choose between virtues. Here are some of the virtues your final product should have at the end, whatever else it may lack:
- 7a. Convincement. The piece must convey to its readers a sense of authenticity, or no other virtues are relevant. It should not only be accurate, but feel accurate. Have you found the right counterpart style? Does the piece move well in the framework you have created for it? Is it comfortable in its century? Is your "War and Peace" 1810-ish? If you are reading proof, has the typesetter broken a stanza across a page, making the poem stumble? Does the eloquent oration persuade? Does it inflame? Or does it read like a footnote of itself? If the latter, you need to have another try.
- 7b. Ease. There are limits on the effort your audience is willing to make. As a practical matter, your work must lie within those limits. One hard word can be handled by an elaborate explanation, and this may be worth while, especially if it is a central term, and if getting it right is crucial to the success of the whole enterprise, and if you have a page of preface to devote to explaining it. But the piece as a whole, though it may be stiffer than other things you are translating for the same audience, should be unlaborious. Some effects (Japanese respect language, Chinese tonal scansion) may have to be sacrificed so that what can be brought over, consistently and not as a momentary tour de force, is not fatally hampered. Don't drown your virtues in your failures.
- 7c. Strangeness. As a corrective to the preceding: Audiences come to translations in search of something they don't already know. Don't betray them. If you have revised out of your translation the qualities which attracted you to the original in the first place, you have gone too far. Don't translate down. Translations should add to the experience of their readers, and enrich the horizons of the cultures into which they are introduced. Otherwise, why bother?
8. Long Thoughts. The encouraging fact is that we get better at this by doing it. We and our colleagues in translation improve, over the years, by doing it together and by learning from each other, including each other's mistakes. The same applies to cultures, as they get more acquainted with each other. Read somebody else's translation of the piece you are working on, but done a hundred years ago. Notice the conceptual and presentational points that were hard to convey to the audiences of that time, the points that were labored over or missed altogether by the translator, but may be routine and easy now. Here is where progress can be felt. This also works in the larger dimension. Translation needs a world of listeners to resonate against, and that world of listeners is learning more all the time. Take the long view. Even if, being mortal, you can participate in only one stage of the long process.
9. Nontranslation. There are some things, such as the interminable cotton candy of Japanese academic prose, that should not be translated at all. Instead, summarize them. Recast them. Cultural contact eventually creates an interculture, above the more local levels at which particular cultures govern and particular traditions dispose. Translation operates within those local levels. But it is perfectly proper to contribute to the emergence of the interculture, along with, or instead of, the work of translation. To add to the interculture's precedents for ways of saying things effectively and efficiently. There is more to the world than the fascination of the past, or the lure of the exotic, or the gestures of the local. There is more to the world than translation.
As was noted at the outset, the University class in 2002 was more interested in our experience translating the Analects than in hearing the above paragraphs of good counsel. Other correspondents have also inquired about our experience, and our presumptions, in approaching that translation. It seems a shame to leave the Analects out altogether. Here, then, are some recent questions on the Analects translation, along with our answers, by way of a personal postscript to the more general hints given above:
- Questions from Di Aiying.
The bibliography of translation is endless. Some of it is just literary theory, and literary theory is mostly worthless, though there are some better moments. Missionaries over the years, and around the world, have given much thought to translation, as a practical and not a theoretical subject, and have found practical solutions, or clarified difficult issues, for some of the recurring problems. One window on that world is available here. Any local library catalogue, under "Translation," will provide other possibilities. Use your own judgement on what you find there. And good luck to you, in your thinking and your doing.
All lectures and abstracts posted on this site are Copyright © by their authors.
5 Dec 2002 / Contact The Project / Exit to Lectures Page