Collaborative Translation

I’ve been recently very interested in translation from a language that you don’t know. Actually, this is far from a recent interest for me – its one of the ways I was initially introduced to literary translation as an art form. But only recently have I realized how controversial, how marginalized this practice can be in a practice that is already marginalized by creative writing and academia as institutions of cultural production.

Ezra Pound, W.S. Merwin, Robert Lowell, Martha Collins, and many many other brilliant writers and translators have chosen to work from languages that they don’t read. Their translations are among the most superb available of these poems. Case studies aside, or perhaps later, there are a couple of problems in opening any discussion on this topic. One is that it lacks a coherent term. I’m beginning with ‘collaborative translation’ though when using that as a research term you tend to get translation firms who work as a team to produce a technical translation. Another one that struck me as a possibility was ‘blind translation’ but that seems overly pejorative. ‘Team translation’ too sporty. ‘Co-translation’ seems like another contender. But for now, I’m going to stick with collaborative, and I’ll tell you why.

Collaborative translation involves at least two people, though they don’t have to be directly involved. One is a reader (preferably fluent or native) of the language of the original text. The other is a writer (fluent and native, and preferably gifted) of the target language. In Pound’s case, he didn’t have direct access to a native speaker for much of his work, and he relied on transliterations (trots or ponies) created by someone who was also not a proficient speaker of the original language. Still, his translations are arguably some of the most influential translations published in the 20th century. So in this case, the translator collaborates with pre-existing texts, and I would take this to include transliterations prepared specifically for the translator (as in Pevear and Volokovsky’s work) or transliterations prepared for scholarly or academic purposes, or even multiple literary translations of the same text.

The other kind of collaborative translation, one in which the translator and (to borrow and subvert a term from Spivak on translation) the ‘native informant’ work closely together in a literary exchange. This is often most fruitful if the translator is working directly with the author, but any native speaker with literary sensibilities and the desire would suffice. And this is something that I’ll want to come back to in another discussion.

The common factor that makes both of these processes collaborative translation is that which also requires any product of this process to be viewed as a work of creative originality as well as a work of translation. It is the requirement of this process that the translator not have linguistic authority in the source language, but artistic authority in his own. And in my mind, that is what essentially separates literary from technical or merely academic translation in any case.

So as I said, my first encounter with literary translation was a one-day workshop as part of the Joiner Center’s Writers’ Workshop (an excellent two-week workshop for writers interested in social engagement through their writing in Boston). Martha Collins, poet and translator from the Vietnamese, was teaching an afternoon master-class in translation, and my friends encouraged me to go. We began by looking at three different translations of a Neruda poem, and though I speak and read Spanish proficiently, most in the class did not. Then, with those three and the original Spanish before us, we set about making our own translations. I was confident that as one of the readers of Spanish in the room I was going to produce something that was ‘better’ – and my youthful hubris was immediately shattered when we shared our versions. We repeated the process with a poem from the Vietnamese, and I was stunned that my creative abilities were sharpened by my distance from the original.

This experience is not unique – the first several masters-level translation workshops I attended had us practice the same exercise. Using the excellent and sadly out-of-print book The Poem Itself edited by Stanley Burnshaw my first translation workshop had us render two different Chinese poems. And most recently, during the first meeting of this semester’s MFA Translation workshop, we began by rendering a short Chinese poem from a previous version and a trot. The merit of these exercises as exercises points to what I think is ultimately the merit of collaborative translation as a practice: that semantic fidelity is ultimately fallacious, that metaphor and image and dependent on too much to transfer between cultural contexts, and that rhyme and rhythm function differently between time periods within the same language, much less between different languages. And this means that poetry is eminently translatable, so long as the translator is freed from the burden of linguistic policing.

This is a massive subject, and I hope that somewhere someday I can listen to people far more brilliant than I am discuss it (I haven’t been able to find any scholarly or critical conversation on the subject, though there are a number of anecdotal pieces of work on the practice). For now, I’m leaving you with three versions of a poem by Ch’ang Chien (8th C A.D.), as rendered by three poets independent of one another with access only to the original Victorian translation (follows at end):

Way when June shown in

Translation by Frodebart Winslow

The dawn in the temple grounds
The dawn on many trees
A path to a covered place
A hut in dense trees
The dawn a happy mountain over birds
Images of a pool empty, a mind
of  10,000 noises, silence
A bell chimes.

Written September 11, 2001


Adapted by Blaz Lilienthall

All’s cold and still as dawn lips the convent
door and breathes a yellow breath on tree-topped
hills. Beneath the canopy, gold stops short
and limns a path of winding dark that’s sent
me to Dhyâna’s hall, where fir and birch
gate a palace for the birds. They cry my
heart from shadow I lose my shadow. Turmoil
roils away to nothing in this lake.
The altar bell tolls with each drowning wake.


Inscription at Broken Mountain
Translation by K. Ripley

Dawn dawns clearly     clear the sacred ground.
First sun shines through deep leaves path
slipping past the seen. Meditate here:
trees     flowers     dense     deep; broken mountain
light illuminates the birds     the earth
a reflection in the pool     empty images empty
heart and mind and silence ten thousand times
the all silent     stone bell chimes.


Translation by Herbert A. Giles

The clear dawn creeps into the convent old,
The rising sun tips its tall trees with cold,—
As, darkly, by a winding path I reach
Dhyana’s hall, hidden midst fir and beech.
Around these hills sweet birds their pleasure take,
Man’s heart as free from shadows as this lake;
Here worldly sounds are hushed, as by a spell,
Save for the booming of the altar bell.


  1. Kevin

    Great post! I, too, am very interested in this topic.

    Finding a term for the practice of translation from a language unknown by the translator (or, in some cases, by the translator-partner who is the principal artistic executive of the translation) is a difficult problem indeed. “Collaborative translation” is a functional placeholder, but I think it’s ultimately inadequate. My problems with it are:

    1. Its literal meaning encompasses, as well as the case with which we’re concerned, translation by partners who all know the original language. It’s inelegant and potentially confusing for jargon to signify only part of its literal meaning. Furthermore, to fix “collaborative translation” in this meaning would terminologically complicate discussion of translation by partners who all know the original language, or discussion of the practice of collaboration in translation in general, i.e. without reference to language skills.

    2. Its focus seems wrong to me. Semantically, there is nothing there to do with lacking knowledge of the original language. But lacking knowledge of the original language is the crucial distinguishing feature of the term’s referent.

    3. “Collaboration” with pre-existing texts is “collaboration” in a metaphorical sense — without the exchange that collaboration implies, it’s a completely different kind of thing than collaboration with a live partner. So “collaborative translation” seems biased toward the live partner case, similar to the way that “resource-based translation” would be biased toward the pre-existing texts case, though metaphorically applicable to the live-partner case.

    4. It pretty much rules out the possibility of literary translators translating from languages they don’t know without using any pre-existing text but a dictionary. I’d heard that this was how Georges Hugnet translated Gertrude Stein, though Googling around it seems like that might be unlikely. Anyway, even if very few excellent translations are produced in this way, I don’t think the term should exclude it.

    I think “blind translation” is moving in the right direction — it lacks the problems enumerated above — but I agree with you that it’s too pejorative. A less pejorative version might be something like “opaque-original translation,” although I don’t know that I’m satisfied with that formulation.

    • Kevin,

      Thanks so much for the thoughtful comment – I kind of can’t believe anyone actually read all that! The issues you raise with the term “collaborative” are exactly the ones that I’m dissatisfied with as well. “Opaque-original” is lovely, but somewhat cumbersome and still rings of a pejorative to me. The translator inherently has to understand the original (though I take that to mean not necessarily the language of the original, but the art of it) and therefore it can’t be wholly opaque. Or at least through the translation process it becomes less opaque. And I worry about the use of anything alluding to the ‘transparent’ metaphor, since the translator is so often expected to be transparent as a pane of glass through which the original can be viewed, and that is highly problematic for me.

      I think you’re right in identifying that it needs to specify the work of a translator who is translating from a language he doesn’t read without allowing for the potential of “team-translation” (which as I understand it is the specific term for a group of people who all speak the two languages in question working together). The focus on knowledge of language may in fact be the problem here, because to call it “monolingual” translation implies the translator doesn’t possess any other languages, which is certainly not the case in the majority of instances. And I think, and I’ll likely write more about this later, that one of the reason there is no discussion of this practice, to the point that we lack a meaningful term for it, is that literary translation has been for a long time under the police of language-studies and area studies who only accept translation as a legitimate discipline when the translator’s authority relies on his full command of the language. We see this happen in reviews of translations a lot, where the reviewer (presumably selected because they read both languages) picks apart the decisions, effectively policing the field of literary translation and preventing creative innovation, something that Venuti writes about quite a lot.

      There are a lot of issues at stake here, clearly… I’ll probably try to break them into parts and see if I can’t get at them bit by bit…and I’m glad to know that you’re thinking about this stuff too!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.