Martha Collins is the coeditor, with Kevin Prufer, of Into English: Poems, Translations, Commentaries (Graywolf Press, November 2017).

Some years ago, I was asked to teach a two-week summer workshop about translating poetry. Some of the participants were experienced translators, but most were not; some had never studied a foreign language. Along with other materials, I gave them Pablo Neruda’s “El Pescador,” in the original and two English translations. Chosen almost at random, the poem and translations proved to be the most valuable resource I offered that summer. Seeing what different translators have done with the same poem immediately eliminates easy assumptions that beginning translators (and non-translators) often make: that there is a single way, a most correct way, or even a best way to translate a poem. Although most of the students had a preference for one translation or the other (as did I, though I didn’t express it), they could see that the two translators of the Neruda poem had very different priorities.

As I continued to teach the summer workshop over the next few years, poems from other languages (French, German, Russian, Chinese, Latin, Greek) joined Neruda’s, and the number of multiple translations increased as well. At least three, I discovered, were far better than two; in the case of Catullus’s famous lxxxv (“Odio et amo”), I included a dozen or more.

Meanwhile, I began teaching semester-long translation workshops—first in a graduate English program, then in an undergraduate creative writing program. In both, I asked the students themselves to compile multiple translations of a single poem written in the language from which they were translating. Their compilations, added to mine, became our most essential “textbook,” and gave us an excellent basis for asking important questions about literary translation.

Useful as these makeshift textbooks were—and important as I continue to think it is for students to compile their own “multiples”—I began to look for other resources. There were, I discovered, a few related articles and other publications that focused on single poems, including Eliot Weinberger’s Nineteen Ways of Looking at Wang Wei and Hiroaki Sato’s 100 Frogs. But there was nothing in print that offered multiple multiple translations.

And so I began to think about a new anthology, which led to conversations with Kevin Prufer about how the translations might be compiled and presented, and what kind of commentaries might be added to them. The result, twenty-some years after that first summer workshop, is Into English: Poems, Translations, Commentaries, which Kevin and I coedited and which was published this month by Graywolf. Much of what I have written so far, as well as what follows, relies heavily on (indeed is often recycled from) the introduction to that volume.

Let me begin by returning to “El Pescador.” Here are Neruda’s first three lines:

Con larga lanza el pescador desnudo
ataca al pez pegado al roquerío
el mar el aire el hombre están inmóviles

And here are the first lines of the two translations:

With his long spear the naked fisherman
attacks the fish trapped in the rock pool
the sea the air the man are still
     (tr. Alastair Reid)

With the length of his lance the stripped fisherman
assaults the stricken fish in the rockery
man, ocean and air keep immobile
     (tr. Ben Belitt)

Even students who had no Spanish could see that Reid’s translation hewed more closely to the original in terms of word order, diction, and style, while Belitt’s seemed more interested in attempting to replicate the assonance and alliteration of the original.

The distinction implied in that recognition led to a couple of basic questions that we began to ask about other translations. First, on a continuum ranging from the most “literal” to the most “free,” where, more or less, did a particular translation lie? And did it lie consistently there? Which brought up a related question: Where, on another continuum between most loyal to form and most free of it, does a translation lie? What is gained by attempting to replicate meter and rhyme (or sound, in Belitt’s case), and what is lost?

And then we began to ask other continuum questions. What about levels of diction? More generally, what was the stylistic register of a translation, ranging from formal to colloquial, or was there a mixture of styles? If the latter, did this reflect the original poem, or was it an unfortunate (or deliberate) result of the translation? If the poem wasn’t contemporary, what was gained and what lost by moving the poem toward modern and even contemporary English? The latter question was particularly relevant to poems from ancient Greek, Latin, and Chinese, about which we could often note whether or not a translation substituted contemporary references for the original ones. At what point did a translation become what John Dryden in the seventeenth century and Robert Lowell in the twentieth called an “imitation”—or, beyond even that, a poem in its own right that might make reference to the original by inscribing “after Pablo Neruda” (or whomever) beneath the title?

In addition to these general continuum questions, students came to grapple with more specific ones. The most cursory example of multiple translations confirmed what any dual-language dictionary tells us: there are many ways to translate even single words, including simple ones like Neruda’s mar (“sea” or “ocean”?). This is of course especially true in English, which has, thanks to its multiple origins, an enormous vocabulary. At the same time, other languages have many words to express what we in English consider to be not only a single word but a single concept; that English, having dispensed with “thee” and “thou,” has only one word for “you” is an obvious example. But it was equally important to realize that translating word-for-word was not the best way to proceed.

For one thing, there is syntax, which differs considerably from language to language. A so-called literal translation of a French or Spanish poem would have us saying, in the Neruda poem, the “spear long” or the “fisherman naked”—a distinction that might be especially interesting if a line break occurred between the noun and adjective, since the Romance language would allow us to perceive the general term before its modification. Did a translation attempt in some way to replicate the order in which the original words appeared, or was it more interested in keeping the stylistic simplicity or complexity of the original syntax? German poems were particularly useful in considering this distinction.

All of this is not to mention the knottier problems that my students began to discover. A translation may go smoothly for a while, and then come upon a section or line that, for any number of reasons (semantic, syntactic, stylistic, cultural), runs into trouble. The trouble spots are the places where multiple translations are most apt to differ. Looking at them carefully took us more deeply into the nuances of both the original language and English—and, more generally, challenged assumptions about how language itself works.

Early on, I happened upon two published translations of a poem that exhibited more likeness than difference: they were so similar that it seemed obvious that one translator had used the other as a basis. Of course most poems contain lines that almost anyone who is not moving toward imitation will translate in the same way. But simply to improve or “correct” someone else’s translation became something my students saw should be avoided—not only because it leans toward plagiarism (my point), but also because the resulting translation is apt to lack consistency.

How would I use multiple translations in a translation workshop now that Into English will be available? As noted, I would continue to ask students in a semester-long workshop to make their own multiple translations. But reluctant as I would be to give up the constantly changing packet of my own multiple translations, I would welcome not only the compilations of multiple translations our twenty-five contributors made, but also their insightful commentary about those translations. I have learned as much from them as my students learned from my packets, and however many times I have read the text in the course of editing the book, I know I will continue to learn. Translations are endless, and so are their lessons.