in-praise-of-mistranslationsIN PRAISE OF MISTRANSLATIONS
On Conversational Translation
by J.G. McClure

We all know Freud talked about the ego and the id. Except he didn’t. What he actually talked about was das Ich und das Er, which is to say, “The I and the It.” The words “mean” the same thing, except they don’t. When we translate Freud, we use the Latin pronouns for “I” and “It,” whereas Freud used the regular, everyday pronouns of his German.

It’s the same meaning, sort of, but the Latin “id” is outside our ordinary speech, and so it lacks the disturbingly uncanny mix of familiarity and otherness that “the It” conveys. “The id is made up of our primal desires—inaccessible and constantly influencing our actions, while the ego struggles to keep up.” “The It is made up of our primal desires—inaccessible and constantly influencing our actions, while the I struggles to keep up.” Hear the difference?

I love translating poetry. I’ve done many translations. But it’s my suspicion that translation is fundamentally impossible. As Cervantes said: reading even the best translation is like looking at a Persian rug from behind.

“Okay,” you might say, “but Freud is just an example of a bad translation. If we had used I and It, the problem would’ve been avoided.” But the problem is still there. Even when words signify the same thing, the sounds and usage histories of the signifiers (and by extension, the feelings they evoke) are very different. As poets, we learn early on that there’s a world of difference between a rock and a stone.

And that’s just within English. The problems are only compounded when we try to move across linguistic and cultural boundaries. In “The Task of the Translator,” Walter Benjamin puts it like this:

The words Brot and pain ‘intend’ the same object, but the modes of this intention are not the same. It is owing to these modes that the word Brot means something different to a German than the word pain means to a Frenchman, that these words are not interchangeable for them, that in fact they seek to exclude each other.

Even if the translator is able to find the supposedly perfect translation for a word, the “modes of intention” between the two words remain utterly different. Brot is not bread is not pain.

If I’m just asking you to go to the store and buy some bread, these distinctions aren’t so important. But when we’re dealing in poetry—where every word, every sound is crucial to the experience of the poem—then the idea of a “literal” translation seems like wishful thinking. How could a translation possibly replicate the experience of reading the original?

In his “On Linguistic Aspects of Translation,” Roman Jakobson argues that “poetry by definition is untranslatable.” The meaning of the poem cannot be separated from the words themselves; rather, the meaning of what is said is formed entirely by the way in which it is said, and untranslatable devices such as the pun “reign over poetic art.” Since, by definition, two languages cannot say the same thing in the same way, it becomes impossible to translate the sense of a poem from one language to another. Instead, “only creative transposition is possible”: poems are so untranslatable that, for Jakobson, taking a poem in Spanish and moving it into English is equivalent to taking a poem in Spanish and moving it into the form of an interpretive dance.

If that’s the case, one possible solution would be to perform such exhaustive etymological and historical research that the translation can include footnoted explanations of everything that is lost. Nabokov, in translating Pushkin, takes this approach. Arguing that “the clumsiest literal translation is a thousand times more useful than the prettiest paraphrase,” he claims that the translator “has only one task to perform, and that is to reproduce with absolute exactitude the whole text, and nothing but the text.” In order to do so, the translator must have an encyclopedic knowledge of the history, culture, and literary traditions of both the original and the target language.

That sounds good, but there’s an obvious problem. When Nabokov calls for “translations with copious footnotes, footnotes reaching up like skyscrapers to the top of this or that page so as to leave nothing but a gleam of one textual line between commentary and eternity”—he seems to ignore that such an approach utterly changes the experience of reading the poem. Even if it were possible to give footnotes that fully explain every possible etymological, historical, and cultural nuance of every word in the original text, thereby correcting for the problem of untranslatables, what we would be left reading is nothing like the original poem. The experience of reading this new thing—less poem than exegesis—would be wholly different than the experience of reading the original, so the translation has not been (and cannot be) literal.

Nabokov’s vision of translation fails to address an even more insurmountable problem: any translation is inevitably going to be influenced by the translator’s historical position. In “The Pedagogy of Literature,” Lawrence Venuti presents three supposedly-literal English translations of The Iliad, each from a different time period. All of them are totally “accurate,” yet they give three completely different versions of Achilles. He’s a different character in each version—three different Achilleses made to align with the values of three different societies. Although none of them are mistranslations, they all say more about the translator’s history and culture than about Homer’s. Objectivity is impossible.

So where does that leave us? I see two options. First, we could wring our hands and pull our hair and weep. Or, we could choose to celebrate the translation as a new piece of art. Rather than striving toward the impossibly asymptotic goal of “literal” translation, we can revel in the creative possibilities afforded by the meeting of two minds, author and translator.

In this formulation, “translation” doesn’t seem like the right word. “Adaptation” maybe. Or my preference, “conversational translation.” (Converslation?) In other words, translation is more like ekphrasis: when I write about Van Gogh I’m not making a Van Gogh painting, I’m interacting with his work and making something else.

Again, I find Venuti’s arguments useful here. He writes of the “remainder,” those literary effects that can only exist in translation and which therefore “bring the awareness that the translation is only a translation, imprinted with domestic intelligibilities and interests, and therefore not to be confused with the foreign text.”

Though the phrase “only a translation” wrongly suggests that the translated piece is somehow inferior to the original, Venuti’s point about new effects introduced by translation is a good one: it figures the translated piece as something new, something distinct from and not to be confused with the original. Rather than trying to exactly reproduce in English the effects of a Spanish or Greek or German poem—a task impossible from the start—we can choose to celebrate the new artistic possibilities of the conversation between two writers.

In “Taking Gendered Positions in Translation Theory,” Sherry Simon presents a theory in which “writing and translation practice come together in framing all writing as re-writing.” The translation is a new poem, one that comes after and is inspired by the original, yet makes no claim to literally represent it. Quoting Antoine Berman, she writes: “The translator has all the rights as long as he plays his game up front.” Though this isn’t conventional faithfulness, it is respect. Respect for the author of the original poem by not claiming to speak for her, and respect for the reader by avoiding any such deception.

The result of the practice of translation that I’m suggesting here is something like a poetic tightrope walk—freely and consciously re-writing while striving to avoid the potential ethical pitfalls that come with such a practice. And though truly literal translation is impossible, there is of course a spectrum of how radically the two texts might diverge. So I have to conclude that what is appropriate in translations must be considered on a case-by-case basis, and that some Grand Unified Theory of Translation is impossible. Instead of trying to give you one, I’ll close this essay with three “conversational translations” from various points on the faithfulness spectrum.

Here’s a version of Heinrich Heine’s untitled poem usually referred to as “Der Dopplegänger”. Although I’ve translated the sense of his poem pretty closely, I’ve added a new meta narrative. In Heine’s poem, the speaker comically rails against another man for aping his poetical feelings. I know well that blend of narcissism and empathy, so I found myself quietly resenting Heine for beating me to the punch by a couple hundred years. So in my version, the speaker rails against Heine while Heine rails against the man. Though my only major content changes are the title and the insertion of Heine’s name into the final stanza, these changes foreground the interaction between translator and author, and thereby allow for another layer only possible in translation.

Portrait of the Artist Translating Heine

The night is calm, the streets are quiet.
In that house once lived my beloved;
long ago she left this place, and still
the house stands where it stands.

A man is here, staring skyward,
wringing his hands in violent pain.
It makes me shudder to see his face—
the moon shows me my own form.

O Heinrich, mein doppelgänger, my pale companion—
what are you doing aping my lovepain?
It has tortured me here, on this spot,
night after night, since forever ago!

Heinrich Heine, “Der Doppelgänger”

Now for something looser. The poem refers to Catullus 8, and structures itself similarly to his poem. What I like from Catullus—his denial and prodding and masochism, his unrepentant vulgarity and some of his imagery—I took. What I don’t like—vague phrases, the degree of repetition, the gross misogyny—I’ve jettisoned. I’ve added a lot of new language and images, changed the structure, changed the focus, and have explicitly placed a version of myself in the poem so that it can’t possibly be misread as a faithful translation.

Little Anger Poem
Catullus 8

Poor fucked-up McClure, stop fucking up.
There was a time you’d go anyplace Ellie commanded—

Ellie, so loved you’ll never again love anyone like her.
There was a time when everything

you desired, you got. And? She wasn’t undesiring.
All suns shone brightly on you then

(sunstroke, saltwater desire)—

Now she says no. You, fuckup, say no too.
So be now hardened, sealed-off,

(pressurecooking desire)

toughen up your head. Goodbye Ellie.
I am hardened, toughened, sealed-off—

(I am I am I am I am)

What life waits for you, love? Who will want you?
Who will you want? Who will be yours

now? Whose lips will you bite?
Whose lips are you biting? Whose?

And finally, the loosest version of “translation.” I borrow and translate lines from Neruda’s famous “Poema XX [Puedo escribir los versos más tristes esta noche],” and incorporate them into the poem. They’re the engine that moves the poem forward, and the speaker is thinking of Neruda because he’s in the same situation as Neruda’s speaker: his lover is gone, and he’s tormenting himself with thoughts of her in order to get a poem out of it. The couplet format reflects Neruda’s poem, and the closing images of bread and stones borrow from some of Neruda’s favorite objects. But most of the poem isn’t from Neruda at all. Rather, it’s explicitly about the troubled conversation between the speaker and Neruda—whom he at once admires and criticizes. This poem is not at all Neruda’s poem, but it couldn’t exist without Neruda’s poem. In that way, it seems to me a species of translation.

Self Portrait with Pond and Lines from Neruda

Tonight, Pablo, we could write the saddest lines.
You and I on the ramshackle pier, under the infinite sky.

Beneath la noche estrellada: the starry night /
the shattered night, my soul

not content with having lost her
nor you content without suffering

rightly for the loss. I can’t quite hurt
the way I should. You understood.

So what if your couplets were bad?
All kisses and mourning and dew of the soul—

you had to twist the knife. I don’t love her,
that’s certain, but maybe I love her,

and how to be sure without hairshirt verse?
There was a time I could find passions

without digging—but we, the us from
that time, we’re not the same anymore.

Pablo, if you were here, we’d throw rocks
across the water. You’d tease my rusty Spanish

and we’d shake our heads with knowing sadness
at the little fish who rose to surface,
believing each stone a gift of bread.

Of course, there are many situations in which we’ll want a translation that is as close as possible to the sense of the original. Since, for instance, I can’t read Chinese, I’m wholly dependent on careful translators like Eliot Weinberger in order to access any Chinese poetry. It’s not the same but it’s the best I have. Likewise, in quoting Benjamin earlier I depended on a translation. That is, there are cases in which—although bound to fall short of true accuracy—we need the translator to try to get as close to that unreachable asymptote as possible. I’m not suggesting we abandon conventional translation altogether. Instead, let’s keep in mind that any translation will inevitably bear the mark of the translator—and that fact can be a good and generative one.

J.G. McClure

J.G. McClure is an MFA candidate at the University of California-Irvine. His work appears in Gettysburg Review, Green Mountains Review, Colorado Review, The Pinch, and The Southern Poetry Anthology, among others. He is at work on his first collection. See more:

Image credit: Unsplash




  1. Actually ‘er’ means ‘he’. The German word for ‘it’ is ‘es’, unless that was a typo? Or did you mean das Ich und der Er? (Using then the masculine article for the masculine pronoun.) I think it creates then a far different meaning then, that the id is actually then ourselves at a remove, yet not so impersonal as an ‘it’, with all the attendant ideas around third person narratives as applicable to our psyches. The way we tell stories about ourselves. Sorry to be picky but I found it difficult to read an article about translation when that mistake was at the very start.