The Anxiety of Translation: A Conversation between Ilan Stavans and Robert Croll

From a translator’s viewpoint (at least, from this translator), the best author is a dead author. That absence is a form of freedom.

Translation, by definition, is about dislocation. By traveling from one culture to another, our rootedness is turned on its head. In this dialogue on translation and anxiety, Ilan Stavans, Lewis-Sebring Professor of Humanities, Latin American and Latino Culture at Amherst College, the publisher of Restless Books, and the host of NPR’s podcast “In Contrast,” and Robert Croll, translator of Ricardo Piglia’s three-volume The Diaries of Emilio Renzi (Restless Books, 2017–20), ponder the responsibility the translator has toward the original text, the discoveries of how unstable the target language is, and the realization that translation is an essentially destabilizing experience.

Robert Croll: For me, the act of translation always involves an underlying anxiety: my feeling of responsibility toward the original text, which is bound to the knowledge that my words will be taken to represent the author’s intentions, leads to a constant fear of being discovered as an impostor. But can experience in translation destabilize the way we read texts in their original languages?

Ilan Stavans: It unquestionably does. Translators, out of experience, become extraordinary readers. The adjective “extraordinary,” in this case, doesn’t mean good. Translators are often terrible readers. That’s because they come to understand an author’s intention better than most people. They see the hiccups, the imperfections. I’m tempted to call that “the myopia approach.” As a translator, you hold the original as close as possible. Personally, while I am translating I frequently bring the source closer to my eye, almost touching my nose, to make sure I am getting the meaning accurately. I do the same with the screen: in dealing with a challenging passage, I enlarge the font. This, of course, is in part a result of my age: I am fifty-seven and therefore my eyesight requires a bit of help. But originals don’t want to be read that closely, e.g., they aren’t designed to be scrutinized with such care, to be tested as if under a microscope.

That act of destabilization, of course, extends itself to any other work coming before the translator’s eye, not only those about to be translated. The same happens to a tailor: she sees aspects in clothing the rest of us miss. Or to those dedicated to political polling. They develop an acute eye about human behavior enabling them to understand if an issue has traction.

RC: And, when we move from speaking a language that is not our own to translating from it, our native language is not left intact. When we attempt translation, we continue a critical process that many of us experience the first time we study a second language in an academic (rather than immersive) setting: we begin to apprehend elements of our first language that we could not have imagined before, simply because we had no vantage point. We must become aware of languages before recognizing that we inhabit a language—the same as belief.

IS: Learning a second language entails apprehending an entirely new epistemological way of approaching the world. But it also does something else: it pushes us to understand our origins—not only language but place—with a fresh perspective. That’s because all learning is comparative: we approach a phenomenon from certain assumptions, meaning with a set of preconceptions.

One of my discomforts with the way translation is often discussed is that it tends to look at languages statically: it is about making the content delivered in language A acquire an outfit in language B. But languages aren’t fixed. They are in constant change, not only as semantic codes but in the way individuals approach them. American English today is different from American English a decade ago, or in 1985, when I immigrated from Mexico to New York. And so is my own English: its texture, its granular taste, has evolved within me. I don’t speak it the same way as I did a decade ago.

Likewise, my Spanish. I really learned how it functions—syntactically—only when I embraced English as my daily language of communication. That is, I became aware of its metabolism, which I knew little about when, as a native user, I employed it day in and day out in my adolescence.

Of course, this transaction, e.g., the multilingual journey as well as the act of translation itself, makes you feel exposed, in a constant state of loss. In my first years in the United States, I used to wake up agitated in the middle of the night, with the sense that, linguistically, I was not yet a full-blown user of English while I was already rapidly losing my grasp of Spanish. At the time, I remember fixating on the expression “no man’s land” as a suitable description of this in-betweenness.

RC: I still recall one of the first times I spoke in Spanish after the untested environment of my (rather homogeneous) high school classrooms. I was speaking with a man in an airport; we could understand each other perfectly well, but he immediately started laughing at my speech, and I soon joined in when I realized that, from his perspective, it was like talking to a book (and, at that time, more of a textbook than a novel). To put it simply, my language then was only ever idiomatic in calculated ways. But we become flexible.

Depending on where we learn a second language, we may acquire regional characteristics, but something that fascinates me is how long we can remain ignorant of the associations that our words will evoke for native speakers. The first time I studied in Spain, teachers asked me if I’d learned Spanish in Mexico, saying it was because I habitually used the word ahorita, meaning “right now.” You’re right: we fixate on expressions. It reminds me of the way people latch onto terms in academia when trying to find their footing—the month or two when everyone uses a word like “liminal” or “multiplicity” until it begins to lose meaning.

But another effect of translation on our own language is that we can acquire various idiolects over the course of working with different authors’ voices. Translating a writer becomes easier—or seems to become easier—over the course of multiple texts, as we come to understand the kinks and contours of that writer’s language. But there is a danger here as well: the belief of mastery. If we come to think we have reached a point of certainty about a writer or a language, it becomes easy to read our expectations into the text rather than assess it on its own terms and remain open to surprise.

I often think about certain contradictions among the anxieties of translation. I am able to feel that what I’m doing is nothing original, almost to the point that it seems like a kind of plagiarism, even as I worry that I’m producing an inadequate representation. How can we combat these lurking uncertainties?

IS: It is more than possible. In fact, it is de rigueur. Translators, in my view, are usurpers. We are just like actors: falsifiers, impostors. We traffic on unoriginality, that is, on making copies. The copies aren’t exactly the same as the original but they are bound by it. We are shadows, ghosts.

RC: Does the anxiety of translation change depending on whether the author is alive?

IS: From a translator’s viewpoint (at least, from this translator), the best author is a dead author. That absence is a form of freedom. You don’t have anyone on your shoulder, overseeing what you do. The freedom I’m referring to, of course, is a mirage. The moment the author dies, your responsibility shifts. If the original is being translated, it’s because it has outlasted the author, meaning that it has a claim for posterity. The quality of the translation depends now on how posterity is being negotiated.

RC: And, when a text outlives its author, it is more likely to see future retranslations. In many cases, classics become Everests (or, in the case of more niche novels, at least regional peaks), drawing translators to their challenge. But I think the act of retranslation—though of course affected by such ambitions—is often a reflection of the unfixed quality of language that you mentioned earlier; novels and their translations do not age at the same rate.

One of your latest projects is the translation of languages you don’t know. What are the possibilities afforded by translating from a language you cannot speak, and what are the limits and ramifications of that act?

IS: I find that effort—to translate from unknown tongues—incredibly alluring. On the surface, it could be argued that such an enterprise is foolish. And on the surface, it surely is. But life itself is nothing but matching pairs of socks in the dark. Why then not take the bull by the horns, as we say in Spanish?

There is a long tradition of translators rendering work in languages they are not familiar with: Richard Wilbur translated from Russian, Samuel Beckett and Joseph Brodsky from Spanish, and so on. This isn’t the same as translating a text from a language in which it wasn’t written. Emma Lazarus translated medieval Hebrew poets from a German text by Heinrich Heine. Isaac Bashevis Singer translated from German novels originally written in French. What I’m talking about is doing translations from originals that you as the translator can’t decipher.

Singer, by the way, also translated his own work—alone and with the help of others, often women with whom he might have been involved sexually. He called the translations of his work from Yiddish into English “second originals,” often announcing to the world that they, and not the Yiddish versions on which they were based, should be approached as the source. And indeed, with the exception of the translations of Singer’s work done into Polish and Japanese, which are based on the Yiddish originals, everything else was based on the English versions he co-produced.

Academics these days talk about the impossibility of translation. Or else, about untranslatability, the recognition that, at its core, translation is a doomed intellectual endeavor. The focus actually reaches further, acknowledging that some texts are de facto untranslatable. This argument, of course, isn’t new. Religious debates (I’m thinking of basic Talmudic discussions, for instance) are based on this assumption.

Translating from languages one doesn’t know is about delving into the unknown head-on.

RC: You do it alone?

IS: With the support of an “informer,” so to speak. It is like spying: while an outsider, you need to get immersed in the local culture. You also need to pass. Since you don’t have the means, you depend on someone who, over a glass of whisky, will immerse you in the details of, and about the original. It is left to you, though, to recreate it.

Strictly speaking, this endeavor isn’t a translation per se but an interpretation, though it looks, feels, and acts like a translation.

The insider who isn’t a translator gives you the sense of how the original moves. You take that information and re-channel it in the translation. However, I don’t call these translations but adaptations.

RC: And that’s an important distinction to make. These translations seem like more of a creative endeavor, almost using existing texts as points of departure rather than attempting mimesis with your translations. By working laboriously through an initially incomprehensible text in this way, you may come to fully understand its concrete images and actions but not the intralingual resonances of the words.

But even when working with languages that we understand (or believe we understand), translators are in a unique position to appreciate untranslatability, particularly in the way language is constructed. This leads to a familiar question: What is to be gained in the attempt of translating (supposedly) untranslatable texts?

IS: Again, translation, in and of itself, is an impossible task. Something is invariably lost. Yet translation is a much-needed activity, otherwise each of us would exist in a solipsistic universe. To translate is to approximate, to render in another language a piece whose meaning will invariably be refurbished in another. In that sense, any text is untranslatable.

Obviously, we’re talking now about another type of untranslatability: rendering a text from a language the translator doesn’t know. The gains, in my opinion, are enormous. They aren’t about fidelity but about interpretation. My hope is that readers see them as approximations. And in the age of relativity, an approximation is as good as we might get in terms of appreciation.

RC: I have heard two equally confident declarations: 1) translators will shortly become obsolete figures, replaced by technology, and 2) literature can never be translated by machines; any claim otherwise demonstrates a fundamental misunderstanding of the nature of writing. But the wonderful property of such dichotomies is that, once inspected, they reveal little more than the speaker’s opinion. Yet the advent of machine translation—and, more broadly, the proliferation of digital texts—does change the conditions under which translators work.

IS: Machine translation is a joke . . .

RC: Because it cannot revisit a text. I’ve often thought that the concern about the effects of machine translation comes from a confusion of the roles of translator and interpreter. Interpreters perform a task with immediate political stakes. Often they do not operate near established borders, where mutual understandings already exist, but rather in sites of new cultural proximity: in cities invaded or among populations displaced (likely because of invasions). Their underlying goal, then, is not to repeat precisely what is said but rather to convey what is intended. But translators work with literary texts whose considerations are not exclusively semantic; they may even be intentionally ambivalent. The task, then, is to recreate the way meaning is expressed—or elided.

Of course, our new access to such a surplus of texts means that it is becoming increasingly difficult to encounter a word or phrase that can’t be traced elsewhere. The kind of mistranslation that results from isolation grows scarce. This is particularly true when working with widely spoken and widely digitized languages.

But a constant source of anxiety, for me at least, is the fear of reducing the original in some way. This ties back to the effects of translation on our reading. The way we read becomes prismatic as, over the repeated stages of translation and revision, the text refracts into a series of potential interpretations. We must combat the lazy, violent urge to impose a single, favored interpretation onto a text where the original is ambiguous. We must attempt, in the final translation, to retain the potentials. This is the stage to push against our own egos and try, if nothing else, to vicariously inhabit the ego of the author.

IS: I love how you put it: reading is prismatic. I also like your willingness to remain up to various interpretations while you’re translating the original. It is as if you were resisting the fact, unavoidable from my viewpoint, that the translator becomes the authority through which a piece of literature arrives to a targeted reader outside the original realm. Maybe it’s generational: millennials resist thinking of their presence as an imposition. There’s also the fact that, in the age of hypertexts, we want to give readers as many options as possible. It is an illusion, though. Of course, my own view is also generational. I believe, for better or worse, your translator becomes the entryway to the world it conveys. Since you are the conduit, you might as well assume full responsibility for your actions.

RC: For me, I think an awareness that my presence is an imposition on the text precedes much of my thinking on how to approach translation, and it actually produces and intensifies my sense of responsibility to the original; I want to speak for the text and not for the image I’ve constructed of its author. Of course, it’s impossible for the translation to recreate the same set of possible inferences as the original, but what I fear is representing something intended to be multiple as being monolithic.

But it’s also important to note that the role of translation has not been static. Historically, translation has often been associated with empire: translating from an imperial language in order to impose a specific political and religious ideology or, more broadly speaking, a cultural canon. On the other hand, there is another mode of linguistic domination that occurs through translating other cultures’ texts into an imperial language as if to claim them for the imperial canon: a manifest destiny of art and ideas. How can translation fight against or contribute to exoticization and propaganda today?

IS: It should fight, yes, but it’s a battle it cannot win. Intrinsic in the act of translation is the unevenness of the two languages, the source and the target. The reader accesses the translation because the original is either unavailable or undecipherable. That exclusion inevitably injects a dose of mystery in the original. And all mysteries, psychologically, are dressed in exoticization.

RC: And this perception of mystery doesn’t necessarily begin with the translation—just think of the layers of exoticization and reinterpretation around Carmen as the threads of her story evolve through the words and images of Alexander Pushkin, Prosper Mérimée, Georges Bizet, and Carlos Saura. Nor does it end there: readers, sometimes encouraged by publishers, will read assumptions into the text.

In many cases, at least historically, reviews of books in translation have treated them as fully transparent representations of the original works—this oversight is easy to criticize, yet it is also a deadening experience to constantly include disclaimers that rehash the same basic sense of unknowing that foregrounds any translated text.

IS: I believe this attitude is changing. I have noticed that reviewers of translated books are more sensitive than they used to be. Perhaps this is the result of how globalism permeates everything these days. Rather than simply acknowledging that a book has come from another language, as used to happen, reviewers nowadays champion that fact. But we might be pushing it in the opposite direction. I have noticed, since I launched Restless Books, how reverential reviewers are of translated books. Sometimes that reverence verges on naiveté: because a novel comes from Iran or Chile or Iceland, it must say something unique. That’s a trap! There’s a lot of bad international literature, just as there is a lot of native literature. Knowing how to discriminate the good from the bad takes more than education. It takes courage.

Robert Croll is a writer, translator, musician, and artist originally from Asheville, North Carolina. He first came to translation during his undergraduate studies at Amherst College, where he focused particularly on the short fiction of Julio Cortázar.

Ilan Stavans is the Publisher of Restless Books and the Lewis-Sebring Professor of Humanities, Latin American and Latino Culture at Amherst College. His books include On Borrowed WordsSpanglishDictionary DaysThe Disappearance, and A Critic’s Journey. He has edited The Norton Anthology of Latino Literature, the three-volume set Isaac Bashevis Singer: Collected StoriesThe Poetry of Pablo Neruda, among dozens of other volumes. He is the recipient of numerous awards and honors, including a Guggenheim Fellowship, Chile’s Presidential Medal, the International Latino Book Award, and the Jewish Book Award. Stavans’s work, translated into twenty languages, has been adapted to the stage and screen. A co-founder of the Great Books Summer Program at Amherst, Stanford, Chicago, Oxford, and Dublin, he is the host of the NPR podcast “In Contrast.”


Read more interviews on the Asymptote blog: