omer, who composed in Greek—and who as far as we know never translated, or according to some, even wrote1—was, nevertheless, also the first great translator in the West. Pseudo-Longino bears witness to this fact. The author of On the sublime at one point tells us that in the Iliad the blind bard made men seem like Gods, and vice-versa. Or to put it another way, Longinus understands Homer's task as a sort of translation: translating divine behavior into human, and human behavior into divine: "he made the men who went to Troy gods, to the extent that he could, and the gods he made men. But for us, in our unhappiness, there is a refuge, which is death; while it was not much the gods' nature as their misery which Homer made eternal."2

This eternal misery is also that of the translator, and particularly of the translator of poetry, he who takes on a task which has already and repeatedly been cursed: to reproduce in another language the echoes of the unsayable.

In his hand-to-hand combat with the metaphysical tradition, Heidegger arrives at the extreme of viewing all of Western thought over the last two millennia as a bad translation of the concepts of pre-classical Greek into Latin, something which squeezed out the sap, and dried out its vigor. Something that was already foreshadowed in Nietzsche. And to continue in this vein, the German language could be the noble filter more adequate to the translation of archaic Greek, and thus is exclusively capable of giving it a voice in post-modernity. There is a strain of humor running through this claimed exclusivism.

In Wittgenstein, the question of translation is much more implicit, but not less decisive and anti-exclusivizing. It is authorized in the analysis of the use of certain everyday expressions and slang (which he calls "language games"), which have been removed from their respective axes and point to modalities of translation, aporia and exile within the same language. Besides that, Wittgenstein's own thought is shaped at the watershed between two languages: the German of his childhood, early youth and reading, and the English of his academic training and later exile. But also on the delicate and labile line where the traditions of Jewish and Christian thinking meet.

Benjamin was, without a shadow of a doubt, the most important theorist of translation of the last century. His essay "The Task of the Translator" (which perhaps more accurately ought to be translated—preserving the dense ambiguity of the German term Aufgabe (task, but at the same time vexation) the Feat of the Translator) has no peer in density and exegesis. In it Benjamin argues for translation as an independent literary genre or form. And the way in which he argues for its autonomy—in spite of points which are polemical and on occasion mis-understood (as, for example, his affirmation of the untranslatability of the translation)—is, as a whole, dense, convincing and plastic.

As far as Derrida is concerned, he adds his own spice to certain notions of Benjamin's. He argues—possibly mistakenly—for the translation of the translation (perhaps through not paying attention to how much of the metaphorical there is in this prohibition by Benjamin, which looks at translation always as a modal experience, as a relation, as a unique task which is exhausted in the collision of translating itself). And, especially, he brings together some important questions of detail: the question of the resistance of proper names, the reinscription of metaphysics, "anasemic translation," et cetera.

As we have seen, there are many routes by which we can derive and discontinue concepts. There are many keys in which we can recompose and wander. And yet, on neither side of the Atlantic has the European language which first ventured into globalization yet produced a single theorist of true renown in the field of translation.

In the case of Brazil, to be fair there is, as a mitigating factor, the emphasis placed on translation by the poets of the concretist group in São Paulo, epigones of Pound. The concretists produced an extensive list of good translations and introduced a range of authors to Portuguese who had been hitherto unpublished. A breath of fresh air. An opening to everything which did not come from the chastest French tradition. But essays, investments in the creation of the concept are—though important and revelatory—episodic. There is no truly innovative theory, one that had not been foretold by Pound on the one hand and Oswald de Andrade on the other. In addition to which, the anthropophagy of Andrade hardly includes the question of translation among its primary tasks (a grave omission, and one very little commented upon.)

In the case of Portugal, the most prominent is Pessoa—always Pessoa—a bilingual poet, putting forth meditations on translation which are as graceful as they are epigrammatic and contingent. But, from another point of view, his is the only voice which resonates outside Portugal. And there are many specific and relevant aspects which are only touched upon tangentially. In particular, the specters with which the translator of fiction, and especially of poetry, into Portuguese is confronted. Something like the distance which voice alone will not measure between spoken and written Portuguese—especially in Brazil and in Africa; but also the musicality and the vocal quality of the language, which threatens to entirely transform into music poetic projects as arid and alliterative as those of a Gerard Manley Hopkins or a Seamus Heaney.

And good translators are rare. Those of the most honest stock, who seek to draw their conclusions beginning from the procedures adopted in their combat with the task—or feat. We do not possess many. Those who do not devalue that which may pop up unexpectedly. Paulo Rónai was one of them. But, in general, the lineage of those who have sought to identify the recurring questions in translation from other languages into Portuguese in particular is rather sparse.

That is to say, before we try to work at a solution in terms of creating a more general and methodical systematization we must first identify the problem to be solved. And the problem here, would be precisely those little practical snares—superabounding in gains and losses—with which the translator is confronted before offering the eyes of the reader a commercially published translation.

Among these "snares," one which springs into view—and, more importantly, into hearing—has to do with tone.

And what might the tone of a poem be? It is more than its register, is a more abstract aspect, an certain approximation to the exact axis (or context) in which a poems clamors to be heard. Its voice. Its most intimate voice. But a voice conversing, in movement. Not the shade of the voice -that would be closer to register. Not a sample of it. But the voice in conversation. If the register varies between, let us say, a review, a soliloquy, onomatopoeia, a drunken ditty, the tag of a jingle, an academic thesis, a soccer fight song, lullabies—or even all this together, alternating, in the weave of the same poem—the tone would be that something indefinable in which alone the poem can be expressed in its exactness, in its most interior secret. The tone would be the strongest insinuation of how to read to the poem. Its best form of being spoken. Something that is heard even when it is not read aloud. And that is something much more abstract (and thus less teleological) than its register—which is yes, something more palpable and verging on stereotype.

But the tone is also expressly constructed by the ordering of the words. Perhaps this order is itself the strongest support for the tone, in terms of the written word—without the aid of the voice, of the gesture. Which leads to the thought that syntax—especially in poetry—is something less cerebral than common sense would imagine. And it is not by chance that Benjamin attaches so much importance to word order when he is considering translation. He himself assures us that the clarity of the true translation "does not cover the original, does not block its light, but allows the pure language, as though reinforced by its own medium, to shine upon the original all the more fully. This may be achieved, above all, by a literal rendering of the syntax which proves words rather than sentences to be the primary element of the translator. For if the sentence is the wall before the language of the original, literalness is the arcade." 3

"If someone wants to move to a new experience, syntax is needed, a new syntax. A new syntax is a new cadence of uncovering, a new cadence of logic, a new cadence of music, a new structure of space"4 the American poet George Oppen informs us, in a sentence in which cadences and tones—departing from the same musical radiation perhaps want to say exactly the same thing.

Our effort, then, after this raw definition of tone, is that of, by means of analysis of two poems translated into Portuguese, and published by important presses in São Paulo, detecting, in practice, some "snares" into which translation may step—particularly those that arise from the difficulty of hearing zones of conversation "in reading," that is to say, from the difficulty of apprehending tone. The authors chosen are two acclaimed postwar American poets: Elizabeth Bishop and Robert Creeley. The translators are two contemporary poets and translators who are well established in Brazil: Paulo Henriques Brito and Régis Bonvicino.

It is important to remember here that, in spite of the reservations we may express, translations constitute, in themselves, and a priori, an act of courage. And, in time, it is better that there be translations, even with the possibility of slips as far as finding the tone is concerned, than otherwise.

In the case before us, deliberately, Bishop and Creeley belong to somewhat different moments and trends within the rich panorama of American poetry in the twentieth century. They were contemporaries, but only seemingly so. Bishop, who preceded Creeley, was always at the center of the canon, and considered to be heir to a tradition that made its way through Eliot and Auden. Creeley, on the other hand, followed after those who were on the margins, and carved out their own spaces over decades. His predecessors are Pound, Williams, and Zukofsky, that is, poets, who in one way or another, suffered considerable rejection over the course of their careers. (Though of course this commentary is very schematic).

Let us look at how the following famous poem by Bishop is translated:

The Shampoo

The still explosions on the rocks,
the lichens, grow
by spreading, gray, concentric shocks.
They have arranged
to meet the rings around the moon, although
within our memories they have not changed.

And since the heavens will attend
as long on us,
you've been, dear friend,
precipitate and pragmatical;
and look what happens. For Time is
nothing if not amenable.

The shooting stars in your black hair
in bright formation
are flocking where,
so straight, so soon?
—Come, let me wash it in this big tin basin,
battered and shiny like the moon.

O Banho de Xampu

Os liquens—silenciosas explosões
nas pedras—crescem e engordam,
concêntricas, cinzentas concussões.
Têm um encontro marcado com
os halos ao redor da lua, embora
até o momento nada tenha mudado.

E como o céu há de nos dar guarida,
enquanto isso não se der,
você há de convir, amiga,
que se precipitou;
e eis no que dá. Porque o Tempo é,
mais que tudo, contemporizador.

No teu cabelo negro brilham estrelas
cadentes, arredias.
Para onde irão elas
tão cedo, resolutas?
—Vem, deixa eu lavá-lo, aqui nessa bacia
amassada e brilhante como a lua.5

The poem, seen as a whole, is an a wholly intimate and somewhat solemn register. An oblique declaration of love. Written, as we know, for her Brazilian lover, Lota de Macedo Soares, with whom she shared an edenic house in the mountains in Teresopolis.

Even if spoken by a poet in absolutely flat and colorless reading, this poem would still be iridescent, precariously sentimental, with its elegant diction which connotes a reserved and vaguely embarrassed confessionalism. Here is a love letter as intimate as it is impressively well-made. Its tone is rather straightforward. And yet never in a loud voice. It is as if. Its tone is something explicit, although never spoken out loud. It is as if one were reading a score, expanding on the figured harmony. It presents its own way of being spoken. It does not attack it, so common today. It is disinterested, it does not deconstruct it, parenthesize it, Ashberyize—this is not its game. It does not collapse in pirouettes before the reader, and at the end, beg for applause—as the epigones of John Ashbery do without half the grace of the master. It is more generous and genteel with itself. It is less interested in showing its intelligence because it is in itself as much a declaration of love as of intelligence and grace.

Thus, even when simply seen on paper, a precarious equilibrium between solemn and intimate resounds in the ear. As indeed is suggested by the alternation between abstract and concrete, immeasurable and measured of the entities glossed by the poem: rocks, lichens, halos of moonlight, memories, the skies, Time (with a capital T as the broadest abstraction, almost an allegory), falling stars, finally focusing on her friend's hair, which for their part, shines like these stars, until washed in the prosaic tin basin—which we know so well here in Brazil, and which until the seventies was an emblematic object, present in nine out of ten homes. Yes, the tin basin is a symbol of the home. A sort of hearth alla brasileira and which only someone with Bishop's sensibilitiy and her enormous zeal for becoming one with the "other" could draw upon for a poem about domestic love. I cannot recall a strong Brazilian poet who has made use of this prosaic image. In Drummond, there is, at most, a "tin mug"—something more Bohemian, more from the wineshop, the bar, than from the house, the home. Something more stereotypically masculine. But this tin basin is also the moon, that in some way blots out the stars—milky powder in her friend's hair—during the washing: ablution, baptism, engagement. But also the feminine auto-circumscribing itself in love.

It is clear that this initial image of lichens, of explosions on the rocks, of the concentric shocks converging toward the halos of the moon have to do with a shampoo. And especially with a shampoo which happens over a tin basin "battered and shiny like the moon."

But the losses are evident from the beginning of the translation, and already begin in the title itself, since in Portuguese the word xampu can be stretched to include a bath with shampoo. And even if it is not found with this meaning in the most popular dictionary, a poet might well surmise as much, and employ it thus. Hence the superfluity of being so explicit ("O Banho de Xampu"), since the axis of the poem itself invites one to read shampoo in this sense.6

The translation begins by linking the lichens unavoidably to the "silent explosions," where in the original this is subtly imprecise. Though the sentence could also be read in a more enumerative manner. And further the lichens, so masterfully delayed in the original, are anticipated in the translation. The verb "grow" is translated by "crescem e engordam," rather adiposely. The word "concussões" is entirely outside the epistolary and elegant colloquiality of the register of the poem. But the absence of "our memories" (6, I) is no less grave. "Momento" is a term which spoils the colloquial and quite stylized solution of the two last verses of this strophe in the original. Possibly, in this context, it would be better represented by "agora." Just as the tense "tenho mudado," translated literally to the Portuguese, sounds, at the very least, awkward. The meter is almost always not the same. Would it be possible to reproduce the same level of condensation in Portuguese? Not likely. And if, in English, there is a continuing alternation of longer verses and shorter verses from strophe to strophe, the same cannot be said of the Portuguese, particularly in the first two strophes, in which this alternation is not found, and the verses have almost the same length—between eight and ten syllables.

In the second strophe, the term "guarida"—though certainly not to the same extent as the "concussões" of the preceding strophe—is somewhat out of alignment, though it finds a nice solution to rhyme, meter and balance, taking in the first three lines. Perhaps the best part of the poem in Portuguese. The rhyme schemes chimes on its own, in contrast to the previous strophe, and the obvious problem with the closing rhymes in this one ("precipitou/contemporizador). And perhaps the "for" of the penultimate verse might have been more effectively translated by something like "Pois" or "Pois que," since it is moderately anti-colloquial, as in the English. At the end, there is a rare jewel of inverse alliteration: "Nothing if not," which the translator has not even attempted to approach—and how could he have? The end of the this strophe is simply somewhat troublesome, with its "contemporizador"—so long, rhythmically awkward, and so distant from the genteel tone that the word "amenable" helps to produce in English, and for which, with its gentle meeting of vowels and consonants, there is no lack of more laconic and phonetically more adequate substitutes in Portuguese: "ameno," "docil," "suave"?

At the beginning of the third strophe, the subject of the sentence ("estrelas," stars) is subordinated. The repetition of the same duplex and flabby stratagem of paraphrase from the first strophe ("crescem e engordam") is repeated ("cadentes, arredias"); this is not even to mention that they are far from creating the shining impact of the original (something like "em formaçao esplêndida" in Portuguese)

To sum up, the poem sounds rather awkward in Portuguese. Why is this? In particular, because it has a difficult time finding its register in the language. A manner of speaking which is fairly colloquial, filtered through a gentle solemnity, something intimate, and related to that which used in certain personal letters, notes, or messages. And this is why the reading of the poem in the original leaves us with a sense of unity, of flowing and transitive speech. There are no highs and lows, as there are in the translation.

There are good moments in the translation, nonetheless. The best of them is doubtless the three verses which begin the second strophe. Now, let us move on to Bonvicino's task in translating one of the most famous poems by Creeley:

If in the case of the Bishop poem, by reason of its size and scope, some missteps as far as tone are concerned can be forgiven, in Creeley's poem, whose condensation and metareferenciality are yet more pungent, they threaten to suffocate the extreme minimalism of the piece, the delicate balance between spoken and written.

And this threat is already present in the first verse. In place of "penso," a more effective solution would be "acho" (I, 1). But the suppression of the pronoun is praiseworthy, the right decision. "Grow," in English, does not have as much proximity to the written style as does "cultivar" in Portuguese. A more appropriate equivalent perhaps might be "planto." But this is still conjecture. "Bosque" is an inadequate term that might better be replaced by "mato." The translation of "blossom" by "pétala" is mistaken from the point of view of both meaning ("bud") and metrics, but not for its sonorous emphasis on the bilabial. "Esta"(that) would be better glossed by "essa" (this). And this apparently trivial detail speaks much in relation to the subtle equilibrium between spoken and written in the original. Its ably stylized colloquiality is what does not comfortably finds its own space in the translation.

On the positive side, there is the ingenious suppression of the auxiliary verb ("cada ferida—perfeita") to maintain the rhythm. And, nevertheless, "minuscúla" seems excessively long to stand in for the compressed "tiny."

Here, we insist, more attention would be necessary, precisely because of the minimal size and scope of the original, since every almost inexpressive error in the translation will clang much louder than in the case of Bishop. Although, the direct mode of expression of the poem makes it, at a quick glance at least, easier to bring into Portuguese than in the previous case. Which is, in part, owing to a certain dry and empirical generalism of the original. However, to compensate, there is Creeley's unequivocal and sophisticated sense of measure, that same sense which led Ezra Pound to opine that Creeley possessed "the keenest sense of measure of his generation." And, in fact, Creeley shows us the measure of the lyricism that is possible, in an epoch that has already sung its last.

A sense of measure literally referring to the size of the verse. It is sufficient to note how ably the thought stops short at the end of each line in the first strophe. To then begin again with surprises. An able sequence. Tensions are compared to flowers. What follows "where" is lacunary. If written in medical jargon, "The Flower" could be a treatise on cancer. But it is a poem. About the kind of cancer that breaks in between life and writing. And given tone by a serene and even polite voice. And there is nothing like this ending: everything ringing, reticently suggesting, so that it is even possible to hear it (as it fades out) even after one is no longer reading.

To understand what Pound sees in Creeley as "measure" can be, imperfectly, translated by Creeley's assertion by which "in a poem, I tend to hear that which can be called its melody well before arriving at an understanding of what it all might mean."8 Something which passes through the experience of the eye, the ear, the mind. In a certain sense, the melody of a poem is the length and sequence of its words. "The possible better than the perfect," according to Creeley's own conception.

On the other hand, it is at the least suspect that, in translating unrhymed poetry without a regular meter, the Brazilian translator, by and large, does not pay close attention to the length of the verses. Bonvicino, in contrast, took care here. Less in the case of the initial verse, since its whole length is already contained in "Penso que cultivo." Nevertheless, the length of the remaining verses, which in a more or less regular way, reproduce the metric fluidity of the original, is praiseworthy.

"The Flower" is an extremely efficacious metapoem. It is more or less obvious that Creeley is speaking of his own task. Tensions, cultivated like flowers, are also poems—in the most classic sense of the florilegium and the anthology. And, at the end the alternation of "this" and "that" suggests the well-known game of "he loves me, he loves me not." There is a breath of stoic favor. A sort of kindness. There is melancholy but not misery. There is an almost coherent resignation. And a broad and complex accepting of life as written. Of the written translating life. This possibility. There is a subtle equilibrium between philosophy spoken out loud—like a daydream or a solitary exclamation—and an expressly colloquial manner of speech. Although a colloquial manner that is hard to find in the everyday. And, thus, ably reprocessed. There is dignity, in short.

And yet, all these modes are hardly to be found in Bonvicino's translation, which, to the same extent as that by Henriques Britto, does not really stand on its own, unaccompanied by the original, and precisely because it pays so little attention to the tone of the piece.

The question of the tone of a poem is a mysterious one. It must have to do with a certain refined capacity to hear conversations "on the page." But also to "read" speech. To decipher. It is a matter of an equilibrium that few translators know how to apprehend in its minimal equivalence, in its complex subtlety—which demands that he be not only an intellectual but also—and above all—an artisan, a practical man, who knows how to listen to everything from the radio to conversations in the elevator, by way of political speeches, sports reporting and impassioned harangues. To listen while occupied in slow and repetitive work, which is also the classic time of one who writes books. Or, as Ecclesiastes says, "of making many books there is no end; and much study is a weariness of the flesh." To listen in wise entropy.

But all this, it is understood, would still slip into register. For tone, in this case, would be simply the affection with which this is said, is conversed. The degree of affectivity with which this was mysteriously abducted to a zone of truth. Something less apprehendible. Very difficult to capture the tone of a poem in its greatest breadth. Something that the two translations simply brush on, without leaving marks of teeth, or even of lips; they simply outline, in their courageous failure. Interrupted gesture.

Tone, by reason of its inapprehensibility, its ethereal and scarcely didactic character, through there is in it that cannot easily be learned in a more orthodox way, is exactly that something more that makes a poem from a jumble of lines. That makes him who arranges them in order, a poet. And if a poet is one who encodes against time, tone is his strongest code. It is a matter of the most stable translation affection which can be encoded. It is the dignity of the poem in its most truth-saturated glimmer. It is something capable of mocking stingy esotericism on the one hand, and marketing on the other. To sum up, something that in itself already embodies resistance not only to time, but also to the space furrowed by different languages and cultural shadings.

Thus one can say that tone is the part which is most easy to identify and least easy to talk about in a poem. It is what is least accessible to paraphrase. And what is most resistant to being dissected by a theory. In a poem the tone is the mysterious nucleus in which affection, intelligence and chance are mixed in a zone midway between ear and eye. A pulsar.

But on the labile boundary that exists between tone and register, it is possible to perceive that the tone -although to a much lesser extent than the register—is also an aspect which is conditioned by differing languages. And so drunkards' songs are more syncopated and alliterative in the Anglo-Saxon tongues. And likewise ballads of tender love are much more at home in the Romance languages—after all, the the dolce stil nuovo was originally something from Provence and Tuscany. There is a tone which extends to genres. But, it should be understood, it is only partly determined by genres or periods or styles, since it also transcends them.

In Brazil, to digress, the writers from the Northeast to Rio, by way of Minas Gerais, are the inheritors of a tonal sensibility which is much more Iberian, Mediterranean, Mozarabic, marrano, Moorish. Archaic, mixed. Mestizo in all senses. While those from the south are more turned toward northern and western Europe, or toward Italy, the Middle East and quite recent immigration. They are groping for their own voice. And in this groping one can sense novelty. What those in the south have not yet noticed is the hybridity that comes from the simple fact of expressing themselves in a language that is far from transmitting their most immediate atavistic longings as far as their own descent: Italian, German, Ashkenazi Jewish, Swiss, Slavic, and even Arab and Oriental. What that is new and fused can come from this southern antinomy? Much. And even because, by a historical irony, Brazil Portuguese still is much more focused on the archaic—in the sense of the Luso-Atlantic cosmopolitanism of 1500 -than that of Portugal Hence the preservation of the gerund, of the vowels, and of an organic, mellifluous and marinated lassitude in Brazilian speech—much less hurried than Lusitanian speech.

It is not by chance that a poet conscious of the limitations of his medium, such as the paraense Age de Carvalho would feel at home with Paul Celan. Celan was one who struggled against an entire accepted tradition of literature in German. Perhaps so "that savage not mate with tame," as, indeed, Horace tell us in Ad Pisones. Celan glimpsed the extent to which the best promises, the urbanity, gentility and deep thought of this tradition were only concealing crime. Thus, Carvalho seems to suggest to us that Brazilian Portuguese—and especially that from Rio northwards—is excessively macerated, frayed, weak. Excessively complicit, in other words. That nothing so impactful was produced in it after Machado, Rosa, Drummond, Graciliano and Lispector. There is inertia and exhaustion in the air. When he opts for this self-exile, for this Jewish urging by Celan to self-renew,9 Carvalho is pointing to the vital necessity for fresh air in the Brazilian language. A language that confirmed too many injustices of epic scale, to be able to, suddenly, confront them without first twisting completely around. And so one of the antidotes against this state of things might come from contamination and from exile. From the necessity of seeing anew with unshod eyes—a radically poetic challenge. Whether Carvalho manages to reach his goal, or is frustrated in the attempt, is a conversation for another day.

The parallel that can be drawn, specifically, is with the soccer (Eng. football) of the Brazilian team. Like Brazilian Portuguese as far as expression is concerned, our football team seems inert, through bureaucracy and accommodation. It has lost its tone, and lacks in boldness, improvisation. Players who will be more than mere aggressors. The touch of the ball. Dribbling. Triangulations. All the syntax of the rapid attack in individual and enveloping touches. Space for individual expression that transcends looking for penalties or kicking the opponent to get a miserable lateral or corner kick. We need a more lucid direction capable of saving the immense repertoire of plays promised us by the past. Our traditional love of the attack and the inventive, fortuitous, chance goal is lacking. And even the failure of 1982 has gained mythical proportions, with the finest flower of incantatory soccer was decimated on the fields of Seville. Incalcubable, perverse irony. And exactly there, across from Ceuta.

The media has not helped matters, since it projects the spectacle of a man gesticulating foolishly on the side of the field. And the game itself, within the lines, is ever more forgotten.

But let us return to Carvalho's uneasiness and to the question of tone.

It is important to remember here that supplementarity is what follows from the juxtaposition of the different languages. Benjamin's famous image of shards. In its exclusion they form the so-called absolute language—of which they are nothing more than a refraction. A thought which is rather Platonic in its origins.

One should bear in mind, as well, that in British or American English (like that of the poems discussed above), broadly speaking, the difference between spoken and written is considerably smaller than in Brazilian Portuguese. This almost always means that Portuguese is at an advantage, as far as breadth of registers and tones is concerned. But for this reason, it poses greater difficulty.

As far as the advantage is concerned, it is one which our translators rarely use in all its vigor and power (in the same way that the writers from the south of Brazil have still not noted the goldmine which they have in their hands). And one which, it must be emphasized has not yet been seriously explored. It is as if in English there were fewer chromatic possibilities. And that in Portuguese, by reason of their excess, it were more difficult to identify the correct register and tone for a translation: the space for accommodating that which is, strictly speaking, unsayable. That wandering of voices highly saturated with meaning. That indefinable space in which one hears the conversation of the poem even when it is no longer capable of conversing. That proto-utopia, that true Sebastianism implicit in the well-made poem taken as a whole. And exactly because it stems from an artisanry which is so well-finished that it is capable of creating an abstract hearing, somewhat removed from the eyes.

No one, in Brazilian, not even Manuel Bandeira, knows so well how to calibrate this hearing/seeing in the matter of tone as João Guimarães Rosa. And his research was so arduous, that is even possible to conceive of translating a good part of his books into the same language in which they were supposedly written. A paradox. In the same way as Homer—who he knew amply, and in the original Greek—Rosa made gods of the sertanejos of Minas. This is why one would not be surprised if one day Grande Sertão were to receive an exemplary translation into Portuguese.

In its Latin root, the word tone (tonus) refers originally to the strength of a muscle, but also to the sound of the thunderclap. In the end, everything is ripped into light and flash. Potency. And it is Camões, another translator of men to gods—and the poet par excellence of the Portuguese language—who, in invoking the Tagides, at the opening of his epic, calls upon "a high and sublime sound," a "great and sonorous fury." And he reminds us of the fear that the word arouse "When high Jupiter, speaking thus/ begins in a tone of voice which is grave and awful...." The sound of thunder. In contemporary poetry, the meaning of tone is not so different, since it also refers to an invocation with a strong voice—even when the matter is apparently far from the epic.

It must be made clear that tone is not only affection, a zone of feeling. It is much more than this. It also takes in the organization of the idea. This is why it suggests syntax. As Wittgenstein says, "only one who can converse, can converse in his imagination. Since to converse in the imagination implies that what one lets be said in silence can later be communicated"10 And, in reality, to feel or apprehend is only half the job, since "sometimes one wishes to speak of belief and certainty as tones of thought: and in fact, often, these are expressed by the tone of the voice. Nevertheless, do not think of these as "feelings" which accompany the words"11.

In this sense, tone is the bait. It is the tone which summons the abstract hearing or resonates in the memory, even if perceptible words are lacking. To tune his hearing to the tone is the chief task (or feat) of the translator of poetry. And it is really the first task to be addressed, since every attentive reader is capable of tuning it in. Of attuning himself to the tone. Even before thinking of the words that might embody it in another language.

In a 1914 letter to Ficker, commenting on the poems of Trakl, whom he was supporting at the time, Wittgenstein says: "in fact, I don't understand them, but their tone fascinates me. It is the tone of genius." 12