Side 1

Susan Sontag concluded her essay “Against Interpretation” (1964) with the following dictum: “In place of a hermeneutics we need an erotics of art.” Her thesis was that too much concern with content, with latent meanings and with sociological and psychological subtexts had strangled the pleasure of art as a sensuous experience. Interpretation, she wrote, was a “revenge of the intellect upon art,” and while the visual arts had managed to frustrate the urge to dig under and behind the work through movements such as Abstract Expressionism and Pop Art, literature was still a vulnerable target – in fact the main victim – of the compulsion to interpret. Marxist and Freudian criticism were singled out as the most visible culprits of that urge, of that compulsion. Perhaps, if she were writing today, Sontag would have inveighed against Postcolonialism, Gender Studies, and Ecocriticism.

The sheer variety of critical responses generated by literature are a magnificent tribute to its inexhaustible vitality, and those responses can surely increase our appreciation of specific works, but the fact that critical fads come and go – picking up and dropping “useful” objects of attention in the process – suggests that literature, though not thrashed, has often been thrust into the role of a servant. The literary text, in other words, is sometimes just a pretext. Half a century later, Sontag’s warning is as timely as ever: have we not gotten sidetracked, to the point of missing out on the sensual pleasure of the text?

To get us back on track, she advocated criticism that focused on form and promoted transparency, or “experiencing the luminousness of the thing in itself, of things being what they are.” This loaded phraseology echoes Matthew Arnold’s contention that the purpose of criticism is “to see the object as in itself it really is”; it also evokes Kant, for whom the dich an sich could not be known. Sontag’s main concern, however, was not with philosophy or even with arriving at a refined critical understanding of the artwork “in itself” but with how we experience art. “What is important now is to recover our senses. We must learn to see more, to hear more, to feel more.”

Now let’s suppose that a translator, such as me, sets out to render a literary work, such as a poem, into English. Let’s call that original poem, written in Portuguese, the poem-in-itself. It consists in a certain number of words endowed with particular sounds and meanings and placed in a particular order. Clearly no English translation can ever come close to giving us the poem-in-itself. Fogo means fire, medo means fear, and piedade means pity, more or less, but the words sound different, they have a different number of syllables, and the history of their usage in literature and common speech varies, each word having its own peculiar retinue of submeanings, cultural memories, and perhaps religious or nationalistic associations. 

If we cannot get everything right, we can at least get something. A poem’s music, for instance. Homophonic translations let meaning shift for itself while going out of their way to reproduce sounds. Fogo becomes fog, and medo is a medal. Or, rather than trying to simulate the music, we can take our readers on a trip. Foreignizing translations convey not only what is idiosyncratic in an author’s lexicon and style but also what is distinctive about the language itself. I recently read a Russian novel in translation and was surprised when the main characters – who were gentry and intellectuals, not landless peasants indifferent to the fine points of etiquette – repeatedly threatened to spit on this or that object, person, or idea, right in the middle of their discussions. This is an idiom, I learned later, for accentuating one’s contempt. Although I found it jarring, I admit that it went a long way toward providing me with local linguistic color.

An obvious argument against homophonic, foreignizing and other theory-driven translational methods is that they privilege one aspect (such as a certain type of fidelity to the original) at the expense of other aspects (such as readability). But by applying a number of contrasting methods to the same work, we can arrive at a set of fascinatingly different translations that, taken together, will allow us to get closer to the original. Or they would allow this, if we were able to read them all simultaneously. Our reading experience, alas, would be fatally marred by so much extra noise and busyness. 

So let’s imagine a very special translation of a poem that, through great persistence and a good deal of luck, succeeds in creating near-perfect equivalents of sound, meaning, tone, and texture. Would we then have something close to the original, to the poem-in-itself? Perhaps, but such a translation could only exist in a Borgesian fiction even stranger than “Pierre Menard, Author of the Quixote,” in which the protagonist recreates Cervantes’ masterpiece in the very same language, using the identical Spanish words. To achieve even an approximate verbal equivalence between two different languages is an infinitely more absurd proposition, given the tremendous number of details and nuances that could never be adequately replicated.  Any such attempt would be additionally doomed, or foredoomed, for being founded on a false assumption, since the “poem-in-itself”, be it signed by Camões, by Pessoa or by Carlos Drummond de Andrade, is itself a fiction.

The original poem in Portuguese would not exist without readers – actual or potential – fluent in Portuguese. Independent of whether the poet intends to publish the poem, and even if no one ever reads it, it is radically (rootedly) implicated in a common syntax and a shared vocabulary. The poem was never a pristine object, existing in a hollow cave or on a solitary mountaintop, no matter that the poet happens to be a hermit. There is an intimate relationship between the original poem and (in my case) the Portuguese language, between poem and poet, between poem and reader (actual or eventual), between poet and reader. Meaning does not inhabit the written page; it is an event, or a series of events, which occur during the writing of that page and in every reading of it. By meaning I mean linguistic signification but also the pith, the music, the mystery, the most delicate reverberations, and almost everything else that matters.

To enter that intimacy and to translate that meaning I propose erotics – not as a theoretical program but as a risk-taking attitude crucial for getting close to the text, in the same way two people get close. Human closeness, as I know it (and only through personal experience am I able to explain myself on this point), requires affective, playful, quasi-sexual engagement. Sometimes there is a remote, or not so remote, idea that I could enjoy being physically entwined with the person I feel close to, or drawn to; often there is not. But that entwining, or the inclination for that entwining, needs to exist at a soulish level. You and I may have much in common, plenty to talk about and even sincere mutual esteem, but without an erotic element, a spark, an effervescence, there can be no real closeness.

An erotics of translation assumes there is an erotic relationship between the writer and her poem. I admit there may be worthwhile poems written without any creative eros, but I haven’t come across them. Even a poet who makes poems by applying aleatory schemata to the printed text of a phone directory is apt to feel passion, to be a little or a lot aroused, when setting up the rules of her language poetry game. 

An erotics of translation begins at the beginning, when you flirt with the poem to see if you like it, to see if it likes you. You fondle the poem, try to seduce it, win it over. But only because you have been seduced, only because it has conquered you. (I speak of translations undertaken out of love, not of work for hire, and yet work for hire – like an arranged marriage – can lead to love.) After these preliminaries, you begin to get to know each other in earnest, since the erotics of translation can never be an ejaculatory one-night stand, much less a masturbatory enterprise. It involves a relationship whose consummation is a complex transport, a word I choose because of its double meaning: conveyance (from one language to another) and ecstatic emotion. I mean these last two words literally. The erotic translator experiences ecstasy on arriving at a satisfying solution. This is perfectly appropriate, since the poet probably reached something like ecstasy in the act of writing the poem. But even if she did not, ecstasy is still a good sign for the translation, as it guarantees that the translator, provided he is “faithful,” has stepped out of himself to join with the other – with the poem and with the poet who wrote it. 

“Faithful” does not mean linguistically faithful – or it means this as well, but that is not the main thing. The erotic translator is a faithful, engaged lover of the poem. To be a good lover, he has to know the poem’s textual body inside and out. He will inspect – and fondle – every article, noun, object, comma, lack of comma; every verb tense, every tense verse, strange verb, word bursting with Camões, word pulsing with Latin, word reveling in the favela, phrase basking in an ad campaign from 1954, ellipsis dancing with the unknown... And there are curves to explore, cadences to ponder, breaths to count, heartbeats to feel, that funny tattoo from a voyage made to astral circle number 3, and all the sounds that the body makes.

Eros in translation entails imitation, but not according to the duplicating methods of a counterfeiter. It is a mimesis kindled by admiration, a desire to have more of that wondrous object made of words. It is the need to say it absolutely in my own words, yet without divesting it of its sublime otherness. The honest truth is that I don’t appreciate, don’t feel the poem in Portuguese as much as I do in my English translation – certainly not because my translation is a superior poem but because my English is from the womb, I feel it in my nerves, my blood, my skin, my sex, in a way I will never feel Portuguese, however well I speak and write it. That’s why I translate – to read better, to possess the poem, to be possessed – and that’s why translation is an erotic endeavor.

But to put it more rigorously, what the translator imitates is not the poem, which is inimitable, but the effect that the original Portuguese poem has on the Portuguese-language reader. The particular meanings, sounds and rhythms of words in their individuality and when combined into verses, and those verses into stanzas, will never make the crossing wholly intact. And yet translation happens. Happily enough, poetry is what is not lost in translation, as long it’s good poetry and a good translation. Some of the poetry, amazingly, will even survive a bad translation.

The erotic translator is a kind of acrobat, leaping between the place of the Portuguese-language reader, the place of the original poet writing in Portuguese, his own role as a poet writing in English (whether or not he thinks of himself as a poet), and the place of the English-language reader. He will do well to survey the effect his translation has on English-language readers other than himself, since promiscuously erotic translations – involving other readers and sometimes even other translators – have a better chance of success. The realm of eros is not a democracy, however. Ultimately it is just the translator and the text, battling it out all alone. (In the case of a joint translation, the battle is jointly undertaken, requiring an additional balancing act.)

The translator is like a singer who learns all the notes, the proper accentuation of the words, when to breathe and from what part of his body, every crescendo, every diminuendo and every rhythmic subtlety, and then forgets it all, singing straight from the heart, without which there would be no song worth listening to. 


Side 2


From the highest window of my house
I wave farewell with a white handkerchief
To my poems going out to humanity.

These are the opening lines, in my translation, from the penultimate poem of The Keeper of Sheep, a cycle attributed to Pessoa’s heteronym Alberto Caeiro, who reportedly lived in a white house in the country. Further on in the same poem he explicitly acknowledges a feeling of loss, as if the poems he’s releasing to the world were people he had dearly loved and will sorely miss.  

There they go, already far away, as if in the stagecoach,
And I can’t help but feel regret
Like a pain in my body.

Who knows who might read them?
Who knows into what hands they’ll fall?

And who knows into what languages they’ll be translated? Who knows with what precision or what emotion, or according to what theory or technique of translation? Poets are apt to have scant knowledge of how their work will be treated by others, and less and less as time goes by. Their poems no longer belong to them; they have become public property. But if Caeiro’s regret over their loss feels like a pain in his body, it is because a part of him was placed in those poems and will remain in them. 

In fact the translator cannot, with impunity, ignore Caeiro’s biographical “substance” as a shepherd whose only sheep are his thoughts – thoughts that are after all just sensations – and who insists on seeing things exactly as they are, without filters or philosophy. These particular details about Caeiro are mentioned directly in his poems, and I suppose no one will dispute the advantage, for a translator, of surveying a poet’s entire output, or at least that part of his output where the poem-to-be-translated happens to hang. (Poems grow on trees, since each poet is a kind of tree, by which I mean that all the poems of a given poet are organically connected, even when they don’t look alike.) I further suppose that no one will dispute the usefulness of reading what the other heteronyms and Fernando Pessoa had to say about Alberto Caeiro’s life and opinions. I suppose this, since Alberto Caeiro is a mere fiction, his “existence” is itself a poem, and it is part of each poem attributed to him, since his is the voice that speaks each poem. 

In the case of poems signed by Fernando Pessoa, the biographical details of the poetic speaker will ideally be found in the work itself. It is Pessoa of the poems that matters, not the Pessoa who wakes up at a certain time, eats what he eats for breakfast, reads one or more newspapers, and so forth. But the translator quickly realizes that the poetical Pessoa is inseparable from the biological, anthropological and psychological Pessoa. Ignoring the man and looking only at his poetry, he will discover a richly complex tapestry of political, religious and sexual narratives, but he will still miss out on – and might mistranslate – inconspicuous allusions that a study of the poet’s life and interests would have revealed. Sontag argued that readers and critics should resist the temptation to interpret, even when symbolic meanings have been deliberately embedded into a work by the author. But won’t the translator, without grasping those meanings, be liable to accidently distort or subvert certain symbols in his translation?

And what about directly autobiographical references? These are strewn throughout large swaths of Pessoa’s, and most poets’, poetry, and the translator will obviously want to track them down. Obviously. And yet for the majority of literary critics working in the second half of the 20th century and into the 21st, the idea of inquiring into a writer’s life to throw light on her text was anathema, and authorial intention was regarded as an irrelevancy, at best. To go unhesitantly down that path was to court banishment from the English department.

In recent years the taboo has eased up, as the worthwhileness of contextualizing a given literary work has been recognized, but: “Be careful! The author cannot always be trusted and may not be the best guide to her own work!” These caveats make sense to me, yet as a translator I have to ask at every word and every turn, “What does it mean? What did the author have in mind here?”

Or perhaps I don’t. Although I cannot avoid the incessant repetition of the meaning question, however “against interpretation” I say I am, might it be enough to consult the text of the poem, rather than its author?

It might be, but if the poet is alive, I should probably consult her, and most definitely if I have specific doubts. Why not avail myself of her privileged relationship and understanding? Inside information can be extremely helpful; it just needs to stay inside. Once I was translating a long poem about a series of encounters between two men who were affectively or sexually interested in each other, but no names were given, there was just the pronoun ele (he), and something felt odd to me, the various encounters were not adding up. I consulted the poet, who explained in detail the elaborate subtext, which involved not just two but three men – aha! I did not explain to my readers what was explained to me, and so the threefold he in my English rendition is as potentially confusing as its pronominal counterpart in Portuguese, but had I not discovered the back story, then the front story – by which I mean the actual poem – would probably have gotten muddled in my translation. 

If anything for me is sacrosanct, it is that “front story.” The published poem is a public poem, there for all to see, and may legitimately be interpreted in ways the poet never dreamed of. My job is not to provide one or more interpretations – authorized or unauthorized – but to produce, as far as possible, an equally complex poem in English. If the poet says, “I meant this,” her this does not entitle me to delete ambiguities from the poem that she may not have intended. The poem’s “intentions” count more than the poet’s.

The author-is-dead school of criticism, by which I refer to those critics who play loose with the text on the grounds that it has no meaning until they endow it with one, made me suspicious of Wimsatt and Beardsley’s “Intentional Fallacy” (1946; revised 1954), a precursor to Roland Barthes’ 1967 essay, “La Mort d’auteur” (whose translators, it is hoped, avoided the faux pas of submitting questions to its author). I finally read “Intentional Fallacy” and was surprised to find W. and B.’s critical approach to poetry not very different from how I approach it as a translator. They take the poem away from the author, but without handing it over to the critic. Like so: “The poem is not the critic’s own and not the author’s (it is detached from the author at birth and goes about the world beyond his power to intend about it or control it).”

That parenthetical comment is more or less a prose version of the above-quoted lines from Alberto Caeiro. Since W. and B. admit that the poem is engendered by an author – rather than transcribed by a scriptor, à la Barthes – I wonder if they would admit the possible benefit, for a translator wanting to repeat the same trick, of asking the author how his or her child came into being. 

Messrs. W. and B. continue: “The poem belongs to the public. It is embodied in language, the peculiar possession of the public, and it is about the human being, an object of public knowledge.”

I can only second these reminders that people – rather than dictionaries or academies – are the masters of language and that the poem, however cut off it may be from its author, “is about the human being.” It is the humanness of literature that pleads for an erotic approach – in the writer, the reader, and the translator.

Elsewhere the essayists allow that a poem may express a personality or “state of soul”; all they insist on is that this personality be found in the poem itself. Logic dictates this position, since: 1) if a personality or state of soul was successfully embodied in the poem, then we will find it in the poem, and 2) if it was not, then we have no business asking the author what she wanted to express but failed to express.

This is a clean argument that would convince me as a critic, but translation is a messy, material business. I said in the first half of this essay that each word has a retinue of submeanings and cultural references. Let me rephrase that: each word is dirty from the using, coated with the “meaningful” dust of all the roads it has traveled, the dross of all who have spoken or written it, adding to it their own shadings, twists, intimations, emotions – things extraneous that cease to be extraneous, they become encrusted, or even essential to the word. When words are strung together into verses and those verses are piled on to each other to make a poem, then we have a potential dust storm of meanings. I need to translate that storm – agreed. Only that storm, and nothing that is not in the poem – agreed. But I need to find my way, or the way of the poem, and to do that I’ll take whatever help I can get. I won’t let theories or principles stop me. 

I exaggerate about the dust storm. Fact is, concatenated words often attain incredible clarity, like crystal. It all depends on the kind of poem. But whether it is a surreal journey to obscure domains or a straightforward tale leading to a definite conclusion, it does not exist until it is read. On this point Barthes is absolutely right. The reader described by his essay, however, is an impersonal megareader, proficient in viewing a given text from every conceivable angle, in receiving it from every possible intellectual (or anti-intellectual) and emotional (or unemotional) vantage point. This megareader subsumes, let’s suppose, a million different readers, who would each read the text in a (however slightly) different way. Such an enormously diverse receptor decisively dwarfs the writer, or scriptor, in the paradigm Barthes has drawn with his usual brio. But this reader does not exist. What exists are local, individual readers with height and weight and sex and a name – one million such readers, according to my example, each with their own relationship to the text and the author who will forever haunt it.

I’m with Barthes for trashing the notion that a text has a “secret” (the author’s secret) waiting to be deciphered. The service he performed is comparable to Wimsatt and Beardsley’s salutary denunciation of the intentional fallacy and to Sontag’s impassioned crusade against interpretation. It seems to me, though, that he has reinvested this hypothetical secret in the reader; now the reader holds the key, or the right to determine the value, or whatness, of the text. But any final determinations – wherever they come from – are lethal to a living text, as are all attempts to own it, since the living part is not in the words, not in the author, and not in the reader, but in the relationships between them all. Stripping the author of her authority and the text of its meaning, or secret, was a useful corrective, but Barthes, by placing all authority in a nonexistent Everyreader, also stripped the text of its humanity. No wonder he proposed that the word literature be ditched in favor of the less sentimental and more rigorous, more neutral (neutered?) writing. (One has to credit Barthes for consistency; his essay defines literature as a composite “neuter,” the “trap where all identity is lost.” I cite Richard Howard’s translation.)

For some kinds of writing, including pre-fab literature manufactured according to molds used over and over, automatic translation tools will soon do an adequate job. And it is possible that those same tools, vastly perfected, will in some distant future be able to produce viable and even admirable translations of Proust, of Pessoa, of Akhmatova, and of all literature that provokes and touches us in ways not wholly describable or even perceptible. That future will be the same future in which virtual affection and virtual sex will have successfully replaced our need for human love.


Share this article:
Facebook, Twitter, Google+.
Read later: