Article body

One Art

BY ELIZABETH BISHOP

(1911-1979)

The art of losing isn’t hard to master;

so many things seem filled with the intent

to be lost that their loss is no disaster.

Lose something every day. Accept the fluster

of lost door keys, the hour badly spent.

The art of losing isn’t hard to master.

Then practice losing farther, losing faster:

places, and names, and where it was you meant

to travel. None of these will bring disaster.

I lost my mother’s watch. And look! my last, or

next-to-last, of three loved houses went.

The art of losing isn’t hard to master.

I lost two cities, lovely ones. And, vaster,

some realms I owned, two rivers, a continent.

I miss them, but it wasn’t a disaster.

—Even losing you (the joking voice, a gesture

I love) I shan’t have lied. It’s evident

the art of losing’s not too hard to master

though it may look like (Write it!) like disaster.

1. What do we talk about when we talk about “loss” in translation?

1.1 A field guide to getting lost in translation

“The art of losing isn’t hard to master”—Elizabeth Bishop wrote in her famous poem One Art—“so many things seem filled with the intent/to be lost that their loss is no disaster.” Loss is the subject of this famous poem by Bishop where first the door keys are lost, then the hours are lost, then the places, and the names, and three houses, two cities, two rivers, one continent and one lover. “It’s evident,” Bishop writes at the end of all her losing in the poem, that “the art of losing is not too hard to master though it may look like (Write it!) like disaster” (2008, p. 166).

This paper is about the inflation of loss talk in a certain discourse on translation and how translation is very often presented as a disaster. My question is a simple one: why do so many discussions dealing with interlingual translation in book reviews, in lectures comparing source and target texts, in sites listing untranslatable words, in reflections made by the translators themselves, as well as in works by translation theorists—why do so many of these discussions often involve the notion of “loss” and embrace the assumption that perhaps there is no translation without loss? There are more than 37 million web pages referring to Frost’s alleged statement that poetry is what gets “lost in translation” (in Brooks and Warren, 1961, p. 7);[2] we seem to have an avalanche of “Lost in Translation” books, films, reviews of translations, as well as academic articles referring to some sort of loss in translation. The loss part might sometimes be paired with Lost and Found, or Lost and Gained, or Lost and Regained, or Lost and Relost. But they are all part of the same equation in which “loss” is a central concept. The assumption seems to be then that all translations involve loss simply because they are translations. So why has this notion of loss become such a dominant concept when talking about translation? Where does this rhetoric of loss come from? What are some of the assumptions that underlie this loss discourse on translation products? Can we go beyond a descriptive perspective of the phenomenon and try to come up with a few explanations for this loss talk on translation? Can we possibly account for the loss concept’s pseudo-naturalness and pervasiveness not only within the Western meta-discourse on translation but also within translation thinking itself?

This particular discourse on translation seems to be often saying “you translate therefore you lose.” So now that translators have gained at last some visibility it looks as if they are often seen as losers—visible losers.

1.2 Lose-lose situation

We can find references to the notion of “loss” in John Dryden’s writings on translation in the 17th century. Dryden was a poet, critic, dramatist and translator of, among others, the works of Juvenal, Horace, Ovid, and Virgil, and he is considered by many to be the first English translation theorist. In his Preface to Ovid’s Epistles (1680), Dryden lays out his famous threefold approach to translation and defends his preference for the “paraphrase” method thereby the “spirit” and “sense” of the original are conveyed and consequently “loss” is avoided: “By this means [paraphrase] the Spirit of an Authour may be transfus’d, and yet not lost” (Weissbort and Eysteinsson, 2006, p. 146; my italics). José Ortega y Gasset, the 20th-century Spanish philosopher, declared the following in his famous “The Misery and Splendor of Translation” regarding the translations of Plato:

Whenever a translation of Plato, even the most recent translation, is compared with the text, it will be surprising and irritating, not because the voluptuousness of the Platonic style has vanished on being translated but because of the loss of three-fourths of those very things in the philosopher’s phrases that are compelling […].

2000 [1937], p. 62; my italics

The word “lost” apparently comes from Old English los meaning “destruction,” of Germanic origin, and is related to Old Norse los meaning “breaking up of the ranks of an army, disbanding of an army” with the implication of soldiers falling out of formation to go home. There is a certain practice in lectures and discussions on translations that we may want to reconsider when teaching translation, namely the practice of comparing a source text with one or two target texts in order to expose—directly or indirectly and with all the best intentions—the multiple ways in which the translators are seen to be “disbanding” and “destroying” the source text. The problem is not necessarily in the contrastive method itself. In fact, comparison between source and target texts has been traditionally the main research method in linguistically oriented translation analyses.[3] This method has been applied, for instance, in various Descriptive Translation Studies (DTS) oriented analyses (Toury, 2012) whereby the so-called translation shifts in the target texts are explored in order to identify norms in translation. This approach has made an enormous contribution to translation studies.[4]

The problem arises, however, when the emphasis of the comparison is on showing how the translator in general, and the literary translator in particular, has inexcusably omitted or added or made explicit, or how she, for instance, misused dynamic equivalence, or misunderstood the skopos, ignored the norms, missed the context or the register or sociolect or tone, or perhaps was too subservient or too manipulative or too invisible, or too fluent or too stuttering, or was unaware of her ethical responsibility or was too domesticating or too foreignizing or perhaps anthropophagic or not sufficiently resistant or was simply wrong. One of the reasons for this translator’s lose-lose situation is precisely the centrality that the concept of “loss” has acquired in translation talk in the West. There is a sense in which losses in translation are emphasized and foregrounded more than gains. So how did the rhetoric of “loss” in translation become so powerful in the first place?

1.3 Lost-in-translation talk

Let me offer a few examples in order to illustrate the pervasiveness of “loss talk” in translators’ reflections on their work, in reviews of translations, as well as in scholarly material of theoreticians when referring to translation in general or when evaluating translation products in specific.

The author, translator and theorist Umberto Eco states in Experiences of Translation that “the translator must not waste too much time trying to avoid gaining something, because when translating, one is not so much likely to gain as to lose something” (2001, p. 47). “It is not enough to find words that match,” says the novelist and translator Haruki Murakami, “if images in the translated text are unclear, then the thoughts and feelings of the author are lost” (2013, p. 171). Jeffrey Green devotes an entire chapter to “Lost and Found in Translation” in Thinking Through Translation, where he remarks that “something essential is always lost in a translation simply because it is written in a different language” (2001, p. 149). Mark Polizzotti, in his translation manifesto Sympathy for the Traitor (2018), attempts to answer the question: “is something inevitably lost in translation, and can something also be gained?” (2018, p. xii). In her beautiful “The Art of Losing: Polish Poetry and Translation,” Clare Cavanagh examines Stanislaw Baranczak’s translation of Elizabeth Bishop’s One Art in order to “take a look at what’s been lost and found in translation” (2013, p. 236). “But then why not think of translation as the specific art of loss, and begin from there?” asks John Felstiner in the preface to his translation of Paul Celan (2001, p. xxxiii). Robert Alter likewise discusses his translation of the Hebrew Bible in terms of “Lost in Translation: The Challenge of Translating the Bible” (2008). Within the memoir genre and using a broader definition of translation as cultural movement we have the cult book by Eva Hoffman Lost in Translation (1989). And there is also the (2013) film Lost in Translation by Sofia Coppola with Bill Murray and Scarlett Johansson, as well as the illustrated Lost in Translation compendium book of untranslatable words by Ella Frances Sanders (2016). And if you followed the 2017 translation scandal in the press, the row over the English translation of the Korean novel The Vegetarian by Han Kang (2015), winner of the Man Booker International Prize 2016 for both the author and the translator, you could find titles such as “Lost in (Mis)translation” (in The Guardian where the argument might be summed up as “we didn’t like the book, the translation must have lost something”; Armitstead, 2018). There are more than 400,000,000 results on the web for “how not to get lost in translation,” the assumption being of course that one gets lost in translation.

Not only translators’ reflections and memoirs but also theoreticians working on translation often employ the loss-gain equation and reinforce the loss rhetoric. The renown historian Peter Burke compiled a survey on printed translations made in Europe between 1500 and 1700 which he entitled “Lost (and Found) in Translation: A Cultural History of Translators and Translating in Early Modern Europe” (2007). In this important article where he makes a strong case for finding a large place in history for studies of translation, he recurs at several points to the loss-gain equation. George Steiner in After Babel remarks that “unquestionably there is a dimension of loss, of breakage—hence… the fear of translation” (2000 [1975], p. 189). Also Antoine Berman’s “Translation and the Trials of the Foreign” (2000 [1985]) outlines the author’s major analytic of translation which is based on twelve deforming tendencies. These tendencies are all grounded in the comparative method, they constitute “the universals of deformation inherent in translating” (Berman, 2000 [1985], p. 296) and are strongly imbued with the idea of destruction and loss.[5]

Moreover scholarly materials dealing with the status of translated literature as well as specific resources produced for teaching literary texts translated into English contain discussions where the asymmetry between source and target languages and cultures is foregrounded and provide a long list of what is lost in translation. In “Lost and Regained in Translation,” Beverly Sherry compares several translations of Paradise Lost in order to show “the great difficulty of translating Paradise Lost and what is lost in translation” (2017, pp.32-50). “Of course something is lost in translation,” writes Martha J. Cutter in her notable book called Lost and Found in Translation where she analyzes twenty works written by contemporary ethnic American writers and examines “the simultaneous loss and gain of translation” (2005, p. 1).

“I get tired of hearing this kind of thing, frankly,” writes critically the cognitive scientist Douglas Hofstader on the loss rhetoric in translation: “is this false modesty, or is it some kind of misplaced reverence for the original text?” he asks impatiently (1997, p. 395). “Death,” says Alexis Nouss, “is an invitation to translate” (2011, p. 2). Hofstader wrote Le Ton beau de Marot: In Praise of the Music of Language (1997) after the loss of his wife—a real tragic loss—where he took up her liking for a 16th-century French poem called “Ma Mignonne” (by Clément Marot), sent a copy of the little poem to a great number of friends and asked them to translate it into English while respecting a certain number of constraints. The outcome was a book with more than 600 pages and dozens of different translations of the little poem which ended up acting as a beautiful eulogy for Hofstader’s beloved wife as well as a robust alternative response to Frost’s definition of poetry as that which gets lost in translation.

To come back to Hofstader’s questions above, I do not think that the “loss” discourse on translation of some practitioners and theorists is merely self-modesty or self-deprecation. I do think, on the other hand, that the lost-in-translation discourse might be a symptom of a deeply rooted and internalized equivalence paradigm. The loss rhetoric reveals in fact the extent to which equivalence—as a central defining concept in translation—is far from dead and actually intensely present. If Anthony Pym (2000, p. 165) is right and equivalence is an operative illusion necessary for the definition and social function of any translation, then one possible consequence of this illusion is the actual evaluation of translation products in terms of gains and losses. In other words, one first answer to our question would be that the equivalence paradigm, illusion or not, is very much alive and kicking through the gains and losses equation.

Yet we might still want to ask why it is that losses in translation are actually emphasized and dramatized more than gains. I suggest at this point that we play with a further explanation within the realm of psychology and behavioral economics.

2. Translation and loss-aversion

2.1 What is loss-aversion?

In the late 1970s Daniel Kahneman and Amos Tversky came up with a new theory of people’s decision under risk and uncertainty—the so-called prospect theory (they received the Nobel Prize in Economic Sciences in 2002). Instead of the standard rational choice theory (the predominant theory in economics and very influential in other fields such as legal theory) which says that people make choices that enhance their own well-being, Kahneman and Tversky argued that people do not necessarily choose the option that would maximize their expected utility (Kahneman, 2011, pp. 278-288). In fact, they found that people ordinarily perceive outcomes as gains and losses, rather than as final states of wealth or wealth fare. That was a major breakthrough in terms of understanding how people’s choices and perceptions crucially depend on the way they frame any choice. According to this theory, gains and losses are thus defined relative to some reference point as opposed to being determined in absolute terms. Moreover people take the status quo as the reference point, and view changes from this point as either gains or losses. The second important step in the theory was to actually realize that the weight or value of gains was far from being symmetrical with that of losses. Instead, the value is normally greater for losses than gains. That is, having a dollar taken away hurts more than not receiving a dollar expected. In other words, the effect generated by a loss is greater than the effect produced by a similar gain.

The important conclusion was therefore that losses loom larger than gains in people’s perception. Moreover the consequence of this perception is that people are generally loss-averse. That is, individuals will be willing to pay less in order to gain an entitlement than they will demand in order to lose the same entitlement. Loss-aversion has been found in many real-world contexts and in field experiments to be a robust behavioral phenomenon.

Although the groundbreaking studies of loss-aversion are more than thirty years old, the interest in this phenomenon is far from declining. The concept of “loss-aversion” has been important for behavior economics and legal scholars, among others, because it explains how loss-aversion affects the preferences and behaviors of individuals subject to the law and to individual economic decisions.

I would like to argue that these ideas on loss-aversion may help us to better understand how the notion of loss, and more generally the gain-loss equation, operate in the field of translation. The next section examines the main points in this theory and shows how they might be relevant to translation.

2.2 Three observations on translation and loss-aversion

If we apply these ideas on loss-aversion to the field of translation, we arrive at the following three main observations:

1. On gains and losses

The idea that people evaluate many outcomes as gains and losses, rather than as final states of wealth, seems to match the observation that in translation we are very often inclined to evaluate and frame the translation product as a whole in terms of gains and losses.

2. On the (moving) reference point

Gains and losses are, according to Kahneman and Tversky’s theory, defined relative to some reference point, normally the status quo, and not in absolute terms. Applied to translation, we could equate the reference point to the source text and the changes from this point, namely the source text, will be viewed as either gains or losses of the target text. There aren’t then absolute gains or absolute losses. They are all relative to the source text. This means moreover that we might eventually want to pursue the history of this gain-and-loss equation. We may further want to ask, for instance, what were the norms that governed the gain-and loss equation in different periods and societies? I believe we may find that just as we have different norms at different times for the what-why-how to translate so we might also have norms concerning that which should not be lost in the target text. For what we consider a “loss” today doesn’t necessarily correspond to what was considered a “loss” in the past. We could possibly want to write a kind of à la recherche du perdu en traduction. In the next section, I attempt to illustrate this point by looking at some biblical translations. I will examine and compare different translations of the Hebrew Bible from the gain-loss perspective and ask which specific elements of the biblical Hebrew source the various biblical translators apparently could not afford to lose? It is crucial to add here, however, that what translators identify as “unlosable” in a specific source text reveals rather more about their ideas of the source text and what they—from the target language and culture point of view—consider “unlosable” than properly about any “pure” absolute unlosable, specific to the original source text, which brings us directly to the following point.

An additional important consequence for translation theory has to do with the fact that the reference point itself is a moving one—not absolute and fixed—but rather relative and changing on the timeline. In other words, not only are the translations with their gains and losses moving forward in time but also the source texts constitute different reference points on the timeline. There is a sense therefore in which Cervantes’s Quixote, to use Borges’s famous Menard example,[6] does not constitute the same reference point for translators in different periods and societies. Each period and society not only has its own translation(s) of Quixote but also has its own source texts of Don Quixote, simply because the reference point has moved in time and space.

3. On “losses loom larger than gains” (“loss-aversion”)

The crucial observation that losses are experienced as more significant than equivalent gains—the “loss-aversion” phenomenon—helps us to better understand the pervasive presence of the loss discourse in translation. Loss-aversion provides a context within which to account for the fact that losses in translation are accorded more weight than gains.

2.3 The balance between gain and loss: a few implications for translation

What are some of the implications of the analysis above? In his book Descriptive Translation Studies and Beyond (2012), the eminent translation scholar Gideon Toury discusses, in a short excursus at the end of the book, the process of becoming an accomplished translator. He insists that “the We Know Better” stance underlying the implementation of normative conditioning in translation teaching should be replaced by an alternative “Everything Has Its Price.” His radical suggestion is that we train translators so as to violate norms (Toury, 2012, p. 293). Not as an end in itself however—and that is a crucial point he makes—but rather as a means of opening the students’ eyes and minds to the multiplicity of modes of translation, all of which may be legitimate (always according to a certain set of norms) in one context or another, and thus helping them to pursue their own way as translators. Now Toury, far from rejecting the loss discourse, apparently incorporates the gain-loss equation in translation when he declares that:

Students would thus be trained to consider for themselves what stands to be gained by a certain decision made, what would be sacrificed, whether the gain is worth the loss—and whether there are any more appropriate modes of behaviour in terms of the balance between gain and loss within the recipient culture.

2012, p. 292

One possible implication for translation teaching would be, in my opinion, not only to explore but also to problematize this gain-loss equation in the classroom—both in the process of translation as well as in the evaluation of the product—in such a way that might neutralize the bias against “loss” and allow for different attitudes and negotiations to emerge.[7]

Hence by emphasizing the relativity of both the reference point—namely the specific source text in the context of the target culture—as well as the relativity of the gains and losses of the target text, we are in a good position to discuss (both in the classroom but also outside it) the multiplicity of modes of translation, their culture-specificity and changeability, and their dense fabric of norms and constraints. We can thus use the contrastive method and also actively produce translations in class, not in order to mourn the losses, but rather to problematize given ideas on translation loss and discuss the relevance to translation of, among other things, these three (Buddha-like) basic principles: (i) Everything is constantly changing; (ii) Nothing has an enduring essence; and (iii) Nothing is completely satisfying.

3. Through the Losing-Glass

3.1 “Unlosables”

Let us now examine and compare a few translations of the Hebrew Bible from the gain-loss perspective and ask accordingly which specific elements of the Hebrew source text the different biblical translators could not afford to “lose”. Put otherwise, what did the translators of the Hebrew Bible at different periods considered “unlosable”—in the sense of that which should not be “lost”—in their translations?

What we find is that The King James Bible’s “unlosable” (1611), for example, is not the same “unlosable” as The Message’s translation by Rev. Eugene H. Peterson (1993) or Henri Meschonnic’s French translation (2002) or the biblical Hebrew scholar and translator Robert Alter’s “unlosables” in his recent translation of the Bible into English (2018).

The King James Bible translation could not afford to lose the specific cadence of the biblical Hebrew paratactic syntax—the ordering of words in parallel clauses linked by the coordinating conjunctions “and” (prefix we in biblical Hebrew)—which is why we get all the repetitive and and and in the English text and very little syntactic subordination.[8] Interestingly it is just this specific trace of the KJB translation that would, as we know, eventually provide the 17th-century English translation with its unique syntactic branding:

1. In the beginning God created the heaven and the earth.

2. And the earth was without form, and void; and darkness was upon the face of the deep. And the Spirit of God moved upon the face of the waters.

3. And God said…

When we compare the above with Rev. Peterson’s The Message translation of these same opening verses of the Book of Genesis we actually do not find even one coordinating “and.” On the other hand, none of the drama and performativity Rev. Peterson sees (“all you see, all you don’t see”) in the source text and that, most importantly, he needs in his target text, are allowed to get “lost” (“this is not a study Bible,” states The Message site, “but rather ‘a reading Bible’”; quoted in Long, 2007, p. 48):

1-2 First this: God created the Heavens and Earth—all you see, all you don’t see.

Earth was a soup of nothingness, a bottomless emptiness, an inky blackness.

God’s Spirit brooded like a bird above the watery abyss.

3 God spoke:

Robert Alter’s rendering of these two first verses of Genesis replaces the coordinating conjunctions “and” by the subordinating “when.” The resulting sequence—“When God began to create heaven and earth…God said”—transforms the paratactic syntax of biblical Hebrew (which is otherwise a very strong and central feature in Alter’s English translation)[9] into a hypotactic one (subordination instead of coordination). The latter, moreover, allows Alter not to lose an important traditional Hebrew interpretation of Gn 1:1[10] according to which there was no absolute beginning (creatio ex nihilo), but rather a transformation of preexistent matter into a becoming human life scene:[11]

When God began to create heaven and earth,

1

and the earth then was welter and waste and 1-2 darkness over the deep and

2

God’s breath hovering over the waters, God said,

3

Henri Meschonnic’s French translation of biblical texts, in its turn, could not afford to lose something else, which was the oral quality of the Hebrew Bible (conveyed through specific cantillation marks in the Hebrew Masoretic text, i.e., specific diacritic marks above and below the Hebrew letters which are never rendered in translations).[12] Meschonnic thus introduced in his translation typographical marks on the page (mainly blank spaces of different lengths) to convey this oral aspect of the Hebrew Bible:[13]

1
Au commencement que Dieu a créé
 Le ciel et la terre

2
Et la terre était vaine et vide et l’ombre sur la face du
gouffre
 Et le souffle de Dieu couve sur la face de l’eau

3
Et Dieu a dit

There is consequently a sense in which “unlosables” are facts of the target culture.

Like in the Japanese garden, where every object is carefully selected and set down so as to achieve the optimal effect, in these translations every sentence, every word, every space or punctuation mark is carefully considered and positioned in order to frame “that which cannot be lost” in the target text and thus achieve the optimal target-oriented effect. And if we stretch the Japanese gardening metaphor further we could say that there are similarities between Shakkei—the art of borrowed scenery, one of the central pillars of Japanese gardening—and certain strategies of translation. When designing a Japanese garden, the master gardener will make use of the neighbor’s trees, or the distant mountains and waterfalls; the gardener will even borrow from the clouds, the winds and the dew and make them part of the new garden design. Not unlike the Japanese gardener, the translator may carefully choose what will be seen and what will not be seen in her translation, thus transplanting her chosen elements into the translated text.

3.2 Translation and l’esprit de l’escalier

Looking at translation discourse through the losing-glass, we have found a few interesting things. “Some things,” writes Rebecca Solnit in her book A Field Guide to Getting Lost, “we have only as long as they remain lost, some things are not lost only so long as they are distant” (2005, p. 41). Together with such notions as “original” and “equivalence,” “loss” is often likewise taken for granted when thinking translation. It is as if translation not just “crosses borders” (another cliché often invoked when referring to translation) but in fact crosses borders by leaving always something behind. Translation, within this specific (melancholic) perception, is seen as an emblem of what is always missing.

The “lost-in-translation” cliché involves then an ideologically loaded value which, obliquely, tells us more about the target culture mourning its translation loss than about the source text. The history of the “unlosables” may lead us to better understand what is considered a translation loss in different cultures as well as what kind of norms exist at different times regarding that which should not be lost. The “unlosables” therefore reveal something about the source text only inasmuch as it is perceived and interiorized by the target culture. Moreover, at the explanatory level, our loss-aversion combined with the illusion of equivalence possibly provide the necessary fertile ground for the flourishing of the loss rhetoric in Western translation discourse.

And perhaps the loss rhetoric has something of l’esprit de l’escalier or staircase wit, a French expression credited to the French author and encyclopaedist Denis Diderot,[14] and refers to that situation in which you leave a drawing room and are halfway down the stairs—unsatisfied, melancholy, frustrated—before you suddenly think of that crucial comment you just lost the opportunity to make.[15] The loss rhetoric seems, at times, to embrace this illusion that if we could only go back to the drawing room and translate what we should have translated, nothing (“Write it!”) would have been lost.