back toJacket2

This piece is about 11 printed pages long. It is copyright © Kent Johnson and Jacket magazine 2008.See our [»»] Copyright notice. The Internet address of this page is

Kent Johnson

Notes on Notes on Translation


The following observations respond to Eliot Weinberger’s now quasi-classic “Notes on Translation,” a two-page sequence of twenty-five maxims, published in 1988, in Clayton Eshleman’s late, great Sulfur magazine. Weinberger’s remarks are reprinted here, numbered, in a sans serif typeface colored blue. I’ve had a fairly substantial correspondence with Weinberger that stretches back to 1995, or so, and like many, I’m a long-standing, keen admirer of his varied work. When I wrote to tell him I intended to offer reflections in direct response to his text, he replied by writing, “Whoa, can’t believe those notes on trans(lation) are 20 years old. Haven’t looked at [them] in years, but I’m sure I don’t agree with any of it.” Forgotten as they may be by their author, however, those pithy postulates have had a fair amount of impact on not a few translators over the years, and I suspect Weinberger won’t mind if I now reflect on them, if with friendly difference here and there. In any case, perhaps not too long from now, I won’t agree with much of what I’ve written here, either. Translation, with all its paradoxes, does that to one.
      A more recent statement by Weinberger on translation may be viewed at Fascicle magazine, part of a large gathering of translation and essays on the topic that I co-edited in 2005.
      A slightly different version of these remarks was originally presented at the University of Texas/ Dallas in March, 2008, at the invitation of the American Literary Translators Association and published shortly thereafter in the Translation Review. KJ

Paragraph 1

Eliot Weinberger: 1. I’ve said it before: Poetry is that which is worth translating. The poem dies when it has no place to go.


Kent Johnson: An echo of Benjamin here: that poems “call out” for translation, so they may enter their “afterlife.”


And it’s worth saying, too, that deeply worthy poems often die, untranslated, through a kind of pointed neglect… This can happen — and on significant scales — when once-oppositional literary movements come into legitimate status and assume (the agonistic certainties of their convictions now enabled by position) the pursuit of institutional authority and control. Thus, in the name of poetic progress, entire traditions are relegated to the dust heap of the passé or the verboten. As took place under the ideology of the New Criticism, for example. Or as is taking place presently, within a former avant-garde’s rapid academic refashioning… Poems can die this way, too, having no place to go, being insufficiently “advanced,” too tainted by “superseded” forms or conventions of address, to be deemed worthy of translation.


2. The object of a translation into English is not a poem in English.


The assertion is counterintuitive, to put it mildly. But it’s true, or should be. Just as the object of poetic composition is the object of a poem, the object of an act of translation should be an object of translation. Not, that is, something that submissively hides its nature, but something that gladly bears the absences, contradictions, and ambivalences of its being. The ample folds of its robes, to allude to Benjamin again, are borne without deference or diffidence.


3. A translation creates a specific kind of distance: the reader never forgets that what is being read is a translation.


Again, this will be because the intense translation of a poem — regardless of its “fidelity” to the original — results not in the copy of a poem, but in an autonomous object of recreated language. Something ultimately other, singular — an instance, even, as Ortega y Gasset once proposed, of a wholly different genre.


4. A translation that sounds like a poem in English is usually a bad translation.


Or as Schleiermacher famously advised, the poem should not be moved toward the reader, but the reader toward the otherness of the poem… But Weinberger’s “usually” is a key word here: The unusual and laudable thing is the translation of a poem into English that does not English that which challenges the expectations of our English.


5. A translation that strives for the accuracy of a bilingual dictionary is always a bad translation.


Poetry is profoundly unfaithful to the accuracy of the dictionary, so how could translation be bound to the accuracy of the dictionary? A tautological question, of sorts, but one that is worth asking more often, perhaps. Which is not to say that dictionaries aren’t essential, nor that word for word renderings can’t be heuristically valuable exercises…


6. A translation must sound like a translation written in living English — and more, an English that takes advantage of certain possibilities that are not normally available to poems written in English.


And in this way, translation should expand our apprehension of what living English is, a category always in flux, heterogeneous, and without horizon, if it is to mean anything. Furthermore, and as should be obvious, there can be no such thing as “living English” in poetry without living translation.


7. The success of a translation is nearly always dependent on the smallest words: prepositions, articles. Anyone can translate nouns.


Yes and no. The smallest words are indeed essential to the success of a translation. But so are nouns, which anyone can translate, yes, but not necessarily successfully. In the transmission of linguistic and cultural energies from language to language, nouns can become — often must become — something quite dissimilar. I know the example is overused, but it’s true: The meaning of “bread” in Ethiopia is not the meaning of the “bread” you get at Cub Foods. Nor does a “witch” in Bolivia fly on a broom — and you wouldn’t want to dress your kid up like one on an Aymara holiday, either.


8. A foreign word with multiple meanings can — though few do it — be translated into several English words. One word can lead to a few, just as a few words can lead to one.


Or a single word can lead to a single word that is a completely different word. If a word in the original produces a certain kind of effect which the corresponding word in the target language does not, the translator should opt for a completely different word or passage that does. Well, that is fairly obvious, I take it.


9. Effects that cannot be reproduced in the corresponding line can usually be picked up elsewhere, and should be. Which is why it is more difficult to translate a single poem than a book of poems by a single author. Which is why a translation shouldn’t be, though it almost always is, judged on a line-by-line basis.


And it’s here, really — when “formal” approaches are abandoned for “dynamic” ones — that comfortable understandings of translation’s nature and task begin to break down, get messy, and the act begins to interface with the energies of poetic composition proper. There is more to a poem than meaning retrieved in left to right reading, and often a faithful engagement will have little to do with lexical accuracy… The translation of prose, with its relatively predictable syntactical structures and lexical values is one thing; the translation of poetry, where these structures and values interact at multiple vectors of equivalence and are made indeterminate and strange, is another. To be as faithful to certain dimensions of a poetic text may well mean making the translation unfaithful at certain dimensions, too. It is most always a negotiation.


10. Few translators hear what they’ve written.


“Hear” on the inside, that is, as a poet hears from the inside of her or his poem. A translator senses the original from the outside, and can never inhabit it as its poet had. The translator stands, as Benjamin put it, outside the language forest of the source, calling in. But the translator can and must be on the inside of his or her translation, must inhabit the echo chamber of its formal variables and designs, its rhythms, pauses, the meanings of its silences. In such sounding, the translator may begin to hear certain things in the original not previously perceivable from the outside.


11. Pound intuitively corrected mistakes in the Fenollosa manuscript. For the rest of us, it is impossible to translate from a language one doesn’t know. To translate through an “informant” is to paint by numbers: it’s their design, you merely add some color.


And then Pound wrote in his own mistakes, of course, over those of his first “informant,” Fenollosa. And thank goodness for many of the mistakes. Strange how misunderstandings, errant guesses and inaccuracies can enable great translations — great “inaccurate” translations that in turn change everything about a nation’s poetry. And the poetry of other nations, too, in the case of Pound. In fact, as Weinberger himself demonstrates in the introduction to his New Directions Anthology of Classical Chinese Poetry, Pound’s Cathay had a marked impact on 20th century Chinese poetry. Go figure.


Now, for the rest of us, it may be impossible to make translations that change everything, but it is certainly possible to translate from a language one doesn’t know well, so long as one approaches the poem, humbly, as a poet, and has a good informant. Here, then, I would disagree with Weinberger: All translators should have informants, of one kind or another. Someone, yes, will do a first design, but designs can be, and should be, redesigned. Just make it new.


No, collaboration is a good thing, especially in translation, which is always a collaboration, and all the way down. Plus, collaboration doubles the chances of felicitous mistakes.


12. Everything can be translated. That which is “untranslatable” hasn’t yet found its translator.


On a certain level this is true. But certainly great poetry often points (can this still be said?) to the untranslatable conditions of being, those that shimmer and hint just beyond the reach of language. Sometimes, it’s those moments of “pointing” that most seem beyond satisfactory translation. And here, in thinking about the “untranslatable,” I’ve sometimes pictured that well-known optical paradox: the figure of translation and the figure of the poem it faces as efflorescing into two illusory silhouettes, fixed, in turn, by a central illusory vase. Only when the silhouettes vanish, does the vase appear; only when the vase vanishes, do the silhouettes appear. But it’s just a hunch. When it comes to the untranslatable, I have no problem saying that I really don’t know what I’m talking about.


13. The original is never better than the translation. The translation is worse than another translation, written or not yet written, of the same original.


This would be so because the translation, as previously proposed, is something wholly other from the original, in the end. There is the mother and there is the daughter. And we don’t usually say that the mother is “better,” or more “original” than the daughter. Yet even most translators feel that what they create is a lesser, secondary thing, a shadow-copy of a “true” original. It’s time to get over this. There is no correspondence without difference; there is no poetry without translation. And we still don’t know, in any case, what either one is, or where the activity of one stops and the other begins.


14. Translation is not duplication. Every reading is a new reading: why should we expect a translation to be identical?


It is because we can’t expect this that translation practice, in the end, lacks true epistemic boundary or location. It is a spectrum, and translations will mark their distances and velocities along different spectral points of its red-shift range. To be sure, “faithfulness” is the normative ideal of our practice. But I’d propose that freer, imitative gestures — those speeding away, as it were, from fixed points of observation — can sometimes reveal senses and measures in the original that are otherwise lost.


The analogy has its limitations, of course… But poet-translators once practiced free imitation without restraint, infused their imaginations and visions into the source, in spirit of tribute and reverence. Why have we mostly lost this? Why is such practice no longer considered part of translation’s legitimate reach? As I asked in another piece, written on the topic of imitation, “If we can have the works of, say, both Brahms and Cage understood as Music, the art of both Watteau and Duchamp understood as Painting, the writing of both Tennyson and Mac Low understood as Poetry, why can’t we imagine that the task of Translation might extend, for the sake of certain purposes, beyond the relatively delimited protocols and horizons that currently “define” the practice? One never knows what might happen: Once upon a time, for example, a very unfaithful translation by a Scot named James Macpherson was translated into German, and German Romanticism was born.


15. Metaphor: from the familiar to the strange. Translation: from the strange to the familiar. The failed metaphor is too strange; the failed translation too familiar.


But to be sure, and I’m confident Weinberger would agree: The poems that most call out for translation are those that offer some profit of strangeness to the target language. Translation must seek to bring over the strangeness, obvious or latent, of the original, and always guard against the temptation to familiarize it. It is this, after all, that is the gain of translation — the new surplus, semantic or grammatical, that languages can invest in the general economies of others. Surely, and to draw from Roman Jakobson, if we are translating poetry of the Northeast Siberian Chukchees, say, we would do well to consider not rendering “rotating nail” as “screw,” nor “writing soap” as “chalk,” nor “hammering heart” as “wristwatch.” Not for the mere sake of embracing the exotic, but because the grammatical particularities of a language are deeply wreathed in the tenor of the poem’s total poetic field — in, that is, the dynamic field of linguistic and formal equivalence that makes poetry something different from prose. Which is not to say that this kind of transfer is usually possible… In languages with gendered nouns, for instance, where articles shoot-out deep traces of semantic charge, we are, in English, pretty much at a loss.


16. Many of the best translators know the original language imperfectly. All of the worst translators are native speakers of it.


Inevitably so, as native speakers of the source language will not — except in rare instances of perfect bilingualism — be native speakers of the target language. In turn, the best translators will be native speakers of the language they are translating into — and thus, ipso facto, imperfect speakers, to varying degrees, of the source language. But this does not mean that translations by native speakers of any language into another are of no value. In fact, in lesser known languages, they may be indispensible. This is especially the case today, where hundreds of little-known languages are on the verge of extinction. Given this unprecedented linguistic emergency, English-speaking translators of poetry should be seeking out those who can translate dying languages and offer their services as collaborators.


17. Translations are normally reviewed by members of the Department of the original’s Language. They are proprietary, and cannot help but find all translations from their language — -except those done by certain colleagues — to be travesties.


This is not a bad thing, since translation work gets little enough attention as it is. And some of the Foreign Language experts sometimes have important things to say. Today, twenty years after Weinberger made this complaint, the situation applies, when it comes to poetry, just as much to Departments of English and Creative Writing: from “mainstream” and “innovative” academics alike. And rarely, very rarely, believing, as they seem to believe, in the superiority of their direct lineages and traditions, do they comment on works of translation. I say let’s keep hearing it from members of the Departments of the Original’s Language.


18. Translation theory, however beautiful, is useless for translating. There are the laws of thermodynamics, and there is cooking.


Theory is much closer to translation practice than thermodynamics is to cooking. You can’t cook thermodynamics, but you can translate theory. And translation theory can be beautiful, even mystical, too, nourishing and sensual, like fine cuisine. It is true that no single theory can guide the translator’s practice, since different lines, even different phrases or words often require different enactments, most always unconscious, of so-called theory. In this way, a good translation will carry, like a train in a station, after long journey, diverse traces of the diverse “theoretical” regions it has traveled through (to paraphrase an industrious translator). And sometimes theories, even those founded on misprision, can produce epoch-making translations, to come back to Pound. Or they can inspire translators and poets in unexpected ways, even when the theory has no apparent practical use, to come back to Benjamin.


19. Most translators are capable of translating only a few writers in their lifetimes. The rest is rote.


Every poet who has published a book should translate or imitate at least one collection of poems by another in his or her lifetime. Alone or with an informant.


20. A translation is based on the dissolution of the self. A bad translation is the insistent voice of the translator.


But no, the translator’s self is never fully dissolved in the translation, nor should it necessarily be. Think of the Renaissance poets: Personality, voice, hubris, permeates their great translations. Our selves are always in our translations, and no less, I’d say, when we pretend they aren’t… But perhaps there are some little-tried vehicles of distancing we might employ. Perhaps, that is, it would be possible, here and there, for translators to dissolve their public, legal selves by assuming the anonymous cover of personae. Such a choice may have a liberating impact on the translator, making him or her less self-conscious, more daring, and the spirit of the act more intriguing to the reader. Why should translators of poetry not do what so many writers have done for millennia, and quite commonly, up until not so long ago? Over 70% of novels in England during the last three decades of the 18th century were anonymous or pseudonymous. Over 50% in the first three decades of the 19th were, too. Readers accepted this as normal. Translators of poetry might adopt the practice from time to time.


21. To translate is to learn how poetry is written. Nothing else is so successful a teacher, for it carries no baggage of self-expression.


To translate is also to more deeply learn, and marvel at, how grammar works — in one’s own language and in another. And it is to begin, however tenuously, to learn and marvel (this via J.H. Prynne) at the mysterious space or interval between languages — that shimmering area between repelling poles of grammar, which traces or residues of meaning cannot traverse. An area which is very much at the heart of poetry’s substance, perhaps…


22. Any poem should be translated as many times as possible, even by the same translator over the years. Only fundamentalists believe in a “definitive” translation.


And any poem should be imitated, translucinated, or traduced as many times as possible, even by the same translator over the years. That is to say, any poem is the raw material for forging experiments in translation. In fact, and to restate an earlier suggestion, a solid, so-called accurate translation may be the first step towards even better traductions, i.e., imaginative transfigurations that may well extend energies of the original, otherwise lost in more fundamentalist attempts to carry across, as Benjamin would have it, an “inessential meaning.”


23. A translation of classical Greek from 1900, say, is “dated” in a way that an English poem from 1900 is not, for we expect a translation to be written in a version of current speech, and refuse to make the mental readjustments as we would for a contemporaneous poem. With greater time, however — say, an Elizabethan translation from the classical Greek — -such recalibration becomes inescapable: the translation, in our reading, becomes part of the age in which it was written.


This is true, though the fact is not necessarily as depressing for the translator as it may, at first blush, seem… For translations, as I just said, can be translated too, as they often and variously are. A translation, too, if it is strong, will have its afterlife.


24. Nearly everywhere, the great ages of poetry have been, not coincidentally, periods of intense translation. With no news from abroad, a culture ends up repeating the same things to itself. It needs the foreign not to imitate, but to transform.


Which speaks, as my comment at the beginning, to a serious lack in our nation’s current poetry — where things have improved of late, as far as translation goes, but where what is translated tends to be that which reinforces our own national fashions of the moment.


25. An anonymous occupation, yet people have died for it.


Not just translators, but countless innocent, anonymous bystanders as well. Think of The Bible, for instance. History shows that the stakes in translation extend far beyond the page. In fact, and though the question may sound grandiose, what is literature, or history, without translation?

Copyright Notice: Please respect the fact that all material in Jacket magazine is copyright © Jacket magazine and the individual authors and copyright owners 1997–2010; it is made available here without charge for personal use only, and it may not be stored, displayed, published, reproduced, or used for any other purpose.