- AP Aleksey Porvin
- MLR Michael Lee Rattigan
- MDP Mario Domínguez Parra
AP I want to start somewhere unfair. Trilce was published in 1922, but much of it was written around, or shaped by, Vallejo’s imprisonment in Trujillo in 1920–21. The language of the book seems to resist its own stability — words deform, syntax fractures, meanings refuse to settle, and even the title itself appears to have been invented. How do you translate a language that seems to be breaking from within?
MLR Yes, a keen sense of disintegration and of “a language that seems to be breaking from within” in Vallejo’s collection is pivotal. This is inextricably bound up with loss in his personal life: loss of loved ones, his brother Miguel in 1915 and his mother in 1918, together with a loss of freedom during his incarceration — on trumped-up charges — in Trujillo, from November 1920 to February 1921.
Vallejo heightens the personal to the point of combustion in Trilce and any translator worth their salt needs to subject their language to a similar temperature, work into those fractures, be as restless and limber as possible. It is worth bearing in mind what Vallejo himself says about “new poetry,” insistent as he was on the personal (intransferable) element, a direction toward life rather than literature: “its creator enjoys or suffers a life in which these new relations and rhythms are made blood, are cell-based, something in the end that has become vitally bound up with sensibility.”
Relations and rhythms made blood, something vitally bound up with sensibility — this is what the quickened reader feels every time they expose themselves to the poems in this collection. This is further complicated in Trilce by the shadow-play of other languages: Quechua, Latin, French and English, whose structural phrases and rhythms permeate the work. So the question isn’t just how to carry meaning across, but how to carry the breaking itself.
MDP It seems that, starting from the title of the book (an invented word outside any language, except the moment Vallejo thought about it), we have a lot of freedom in order to translate/rewrite the poems in English, but in fact we do not. Language here is codified into a chasm, but chasms are part of the rocks themselves, which for us are Vallejo’s lines. We have studied their geological (with the “logos”, the word, inside the rock) composition and have decided how to reproduce each nuance into English, intended typos included. Typos are so personal here that we had to create our own ones in order to be paradoxically faithful to the original.
AP Both your answers converge on something I want to press further. Michael, you cite Vallejo’s own theoretical language — “made blood,” “cell-based” — from his writings on the “new poetry,” which insists that formal rupture must be lived rather than merely performed; and you locate that rupture’s origin in specific, dated grief: Miguel in 1915, the mother in 1918, the cell in Trujillo in 1920. Mario, you answer with a geological register — chasm, rock, strata — and the striking claim that your invented typos are “paradoxically faithful” precisely because they are personal rather than letter-for-letter.
Vallejo’s fractures are not uniform: some seem to come from biographical pressure (poems tied to his imprisonment, such as Trilce XVIII, where the four walls are addressed almost like a body), others from the language itself (Andean lexicon and, as some critics argue, syntax; coined words and broken orthography; Roman numerals standing in for titles, which the early magazine versions of III and XLVI still had). Benjamin, writing about translation in 1921–23, imagined original and translation as fragments of a vessel that must fit one another in the smallest details so that a larger, “pure” language becomes visible. Does a fracture in Vallejo point to a lost wholeness, or to one still to come? But your typos, Mario, are invented in English; they answer to the Spanish ones without copying them letter for letter.
So: when a fracture in the Spanish is grief pressing outward from inside a life, and when a fracture is instead the residue of Quechua or other tongues pressing sideways from inside the language itself, do these require two different orders of “typo” in your English — one that behaves like a wound (biographical, load-bearing, irreversible) and one that behaves like a graft (linguistic, foreign matter forced into the tissue of the sentence)? Or has your method deliberately refused that distinction — treating the mother’s death and the Quechua substrate as geologically the same kind of pressure, so that in your English, biography and etymology fracture the line in the same gesture?
MDP The decision to create typos implies a personal choice in its creation. I think (unless Michael thinks the opposite) we were too focused, each time, on the word itself to think about anything else. Another translator may have focused on different letters or syllables of the same word in order to create them (same thing happens with neologisms: I translated a novel by Nikos Kazantzakis, which was full of them, into Spanish and there was no question about ignoring Kazantzakis’s original neologisms, so I had to elaborate them myself into Spanish in order to be faithful to the linguistic intentions of the Greek writer). Another example: I translated a book of poems by Andreas Embirikos, the Greek Surrealist poet. One of the poems had a typo in its title, related to the word “crow”. Instead of feeling mortified and angry at the typographer, Embirikos felt elated, because he said it was part of the Surrealist method that things like that happened in a text. So, in order again to be faithful to the original, I had to create a typo in Spanish.
Typos in the translation of Trilce were to be present for sure if Michael and I wanted to be faithful to the original. While I was reading, Aleksey, your excellent remarks leading to your pertinent question, I remembered Marcelo Zabaloy’s translation into Spanish (or into Zabaloyan, because if the result was to be faithful to the Joycean original, it cannot just be Spanish) of Finnegans Wake and the criticisms he received for including too much “rioplatense” Spanish in it, which would be, according to the critics, incomprehensible for Spanish speakers from Spain, for example. But what was Zabaloy supposed to do? You have to bear on your linguistic identity or identities. Narrow-minded readers would then not enjoy such great works as José María Arguedas’s Yawar Fiesta. Are you supposed to have a Dámaso Alonso beside you all the time in order to read for you, rewritten in prose, Góngora’s Soledades? There is not just one way to translate a difficult part of a text and when you are obliged to take decisions in a jiffy, then something just clicks in your mind and, in a certain way, at least at that moment, you know you have taken the right path (of course, the day after you may think otherwise, as it happens with the use, or not, of a comma).
Michael mentions an example in which Vallejo uses a type of Spanish that could be considered archaic: could we ignore such an intervention in the language as that? No, we could not.
MLR Every instance in which a fracture occurs must be taken on its own terms and, in terms of translation, is very much bound up with the sensibility of the translator. Uppermost is the instinctual response, a hearing with the inward ear, and this is what gives any translation a certain uniqueness.
This breaking, or breaking open, allows for a meeting of sensibilities, no? I’m thinking of Ted Hughes’s translation of Pilinszky, a definite sense of meeting sensibility wise, while there’s a strong whiff of Hughes throughout.
To narrow in on one example in the second poem of Trilce (II), a prison poem, the phrase Vallejo uses in the line “Qué se llama cuanto heriza nos?” — there’s a freshness of wound to heriza — herida translates to wound, but eriza refers to bristle or a hedgehog (erizar), which immediately brings to mind a smarting spikiness in terms of feeling, of hackles being raised — you see how Vallejo loads and shadows each phrase, viscerally concentrates every word. That’s without mentioning the inversion of word order and the odd use of Qué rather than Como.
Now my rendering of this line “What’s that called which up-hackles ūs” falls into the molten core opened up by Vallejo’s deeply suffered original. The (e)strangeness is all.
AP What strikes me is that the “wound versus graft” distinction I proposed is itself a bourgeois formal category, one that assumes language has an interior (biography, feeling, the intransferable self) and an exterior (borrowed syntax, foreign substrate) that can be kept apart.
But Vallejo’s Spanish doesn’t have an inside. Indigenous Andean speech isn’t a graft on his language; it is sedimented in it, but never fused, the way Cornejo Polar would insist against what he saw as the synthesizing drift of Rama’s transculturation: not harmonious synthesis but heterogeneity, a literature written from within an unresolved colonial fracture, where Spanish is already the imperial graft, laid over Andean grounds, Quechua among them, itself once an imperial language. The voice that results belongs wholly to neither. How far that substrate reaches into Vallejo’s syntax is a matter of critical dispute; what no one disputes is the lexicon, and guano, in the very first poem, is one of the Quechua words the book hands us. It is also the seabird dung on whose export the Peruvian state of the mid-nineteenth century built its revenue and its foreign credit, with largely Chinese contract labourers digging it out by hand, and the book comes back to the “guano islands” in XXV, so even a word from the book’s opening poem is already a word of ledgers.
So when Michael reads grief in the prison poems and Mario reads geology in the typos, I don’t think you’re describing two orders of damage. You’re both describing what Aníbal Quijano called the coloniality of power: the state that puts Vallejo’s body in a cell in Trujillo on “trumped-up charges” is continuous with the state whose Castilian was the language of the writ, the file and the sentence, the language in which that imprisonment was written up before it could be written about. The cell and the substrate are the same apparatus seen from two distances.
Benjamin’s broken vessel, incidentally, is often depoliticized into an aesthetic figure — fragments fitting together toward a “pure language.” Benjamin’s essay is theological in its terms, but if we read it against the grain, with the later Benjamin, translation appears as labour on a historically produced language. Your typos, Mario, take Benjamin at his word: the original is itself a fragment, so there is no whole vessel to restore, and Trilce was never whole in any case — it was written by a man the Peruvian courts had locked up. So “fidelity” here can’t mean fidelity to an origin. It means fidelity to the fact that the fracture was put there by history, by incarceration, by empire — and your job isn’t to heal it in English, it’s to refuse to let English pretend it never happened.
MLR This fracture “put there by history” makes me think of Joyce’s words — “history is the nightmare from which I’m trying to awake”. Any flight through those nets has to enact the sedimentation you mention, Aleksey. No “harmonious synthesis”, only rupture and renewal. Octavio Paz says as much in considering Latin American poets writing in Spanish. Here we have a convergence between Joyce and Vallejo and there’s no coincidence that the most unsettling, obstinate and inventive novel and poetry collection emerged in the same year, 1922. Eliot in The Wasteland mythologizes the fracture, whereas Joyce and Vallejo live inside it.
I’m fascinated by what you say: “Trilce was never whole” as this is certainly what the experience of translating the collection underlines. Fidelity means carrying the wound from one border to another, as Vallejo did in his own life and just as so many refugees are doing now.
MDP There is also the fact that Michael & I had to deal with more archaic varieties of the Spanish language, which is the colonial, dominant language, maybe, since the sixteenth century, when Spain had already a big clutch on big chunks of territory (geology, again).
AP Yes — and that “big clutch” Mario names isn’t background, it’s the operative fact. The archaic and historically marked varieties of Spanish you and Michael were negotiating in Trilce are already a state technology, not a neutral literary register that happened to arrive early. Nebrija told Isabella in 1492, explicitly, that language is the companion of empire — siempre la lengua fue compañera del imperio. So when we call Vallejo’s Castilian “archaic” we should be precise about what kind of archaism it is: not organic sedimentation like coral, but a language whose standard had been explicitly imagined, by Nebrija, in relation to imperial power, and which reached the Andes already carrying that charge. The clutch Mario describes on territory is the same clutch on syntax. That’s the same colonial infrastructure: tribute rolls, encomienda, catechism, orthography. Much of it was built on Inca foundations (the mita was an Inca labour levy the Spaniards repurposed, and the Inca state kept its own ledger in knotted cords) and administered, when it suited, in Quechua itself; after the Third Council of Lima (1582–83) the Church printed its catechism in Spanish, Quechua and Aymara (Lima, 1584). But the instruments that decided a life, the writ, the file, the verdict, spoke Castilian.
Which is why I resist even the word “archaic” a little. Archaic suggests a language aging naturally, like rock strata laid down by time alone. But what you, Michael and Mario, were negotiating in Trilce is closer to what Rama called the ciudad letrada — the lettered city — Spanish as the medium of bureaucratic and ecclesiastical power, the language of the escribano, the notary, the bureaucratic and judicial institutions inherited by the republican state that prosecuted Vallejo. So when Mario says “colonial, dominant language,” I’d push further left than dominance: it’s the language of accounting. Empire’s early administrative acts in the Andes were acts of inventory — of bodies, of land, of tribute owed — and the notary came with the conquistadors. Castilian arrived with the escribano already at work. Vallejo is writing poetry — anti-poetry, really — inside the accounting language of his own dispossession. That’s the material condition Benjamin’s “historically produced language” names, and it’s why fidelity, for a translator, can never be fidelity to Castilian’s fluency. Fluency is the crime scene.
MDP I think I told Michael (correct me if I am wrong) some time ago that Vallejo already did what Joyce achieved 17 years after Trilce, that is, Finnegans Wake, in the sense that Trilce is written in Vallejoan, if you allow me, which is not exactly Spanish, as Joyce’s last book was written in a language of a sole speaker (made out of 60 to 70 languages), which I like to call Joycean. As far as I know, that Vallejoan language was neither spoken nor codified by anyone during that time of avantgarde explosion. Juan Larrea wrote his great poetry in French, Francisco Pino (1910–2002) had not started yet his outstanding career as an experimental poet and I cannot think of anyone else in the Spanish language, at that time, dealing with such an extreme approach to poetic language, having Spanish just as a base from which Vallejo eroded (again, geology) its foundations into a new poetic expression.
MLR And with Vallejo the fault-lines run through his family — both his grandmothers indigenous Chimú women and both his grandfathers Spanish Catholic priests. A reckoning has to be made along the fault-lines of language and colonial depredation as lived experience. In that sense “fluency as a crime scene” takes on a very real meaning and calls for a new syntax, a subversive charge. How many images appear to suffer themselves into being, with Vallejo glass-blowing the language into ever more personal shapes. Yet nothing feels decorative, everything feels essential, or essentially suffered. Is that what you mean, Mario, by Trilce being written in Vallejoan — an extreme approach that is an act of renewal from within?
MDP Yes, Michael, a language no one before, and after, intoned in verse. It is not just a personal use of the language already codified for centuries, a new approach to things poetic, it is the development of a language in poetry previously unborn and not given birth since then by any poet.
MLR That rings true, and such singularity is what drew us to a fresh translation — a felt sense that even more nuances were possible in the struggle to thread ourselves “through the needle’s eye”. In a certain sense a poet is someone who out of one language creates another. With Vallejo the stakes couldn’t be higher; he takes on the accretions of history alongside his own personal experience — he literally sweats the historical burden into a new, vitalized language.
AP Mario, I’d amend your Joyce comparison, and the amendment is where the politics is. Stephen Dedalus fretted in the shadow of the Englishman’s language; Joyce answered by making Babel out of the master’s tongue and sixty other people’s. Vallejo has no sixty languages to hand. He has one, and it is the one that was issued to him, with the sediment of others lodged in it. So “Vallejoan” is not a private language, and I’m suspicious of the adjective “private” anyway, since it is the one form of property this whole apparatus was built to protect.
Wittgenstein’s private-language argument concerns a language whose words name only what the speaker alone can know, his own sensations, and he argues that such a language makes no sense. Vallejoan is nothing of the sort: it can be learned, which is exactly why it is not private. A language that others can share is never really of one speaker: it is the idiom of everyone the ledger didn’t bother to enter. Chimú grandmothers, priest grandfathers, the whole unrecorded arithmetic of the household. Vallejoan is what Castilian sounds like when the debtors start dictating the grammar. Deleuze would say the language stutters. I’d say it is finally being cross-examined.
And Michael’s glass-blowing is more exact than he lets on. Mario gave us erosion, the standard’s foundations worn down grain by grain. What is sand, once eroded? The raw material of glass. Vallejo takes the pulverized residue of the imperial norm, fires it in his own lungs, and blows a lens. Through it you see the crime scene the fluent surface was built to hide. That makes “needle’s eye” a delicious choice for the two of you, since the phrase recalls the Gospel image, transmitted through the catechetical world in which Vallejo was raised, though there the camel stands for the rich man, never the poor. What Trilce threads through the eye is not the camel but the tribute, the whole caravan of it, thread by thread, in a syntax that refuses to be counted. That is what fidelity means for a translator here. You don’t restore the language to the accountants; you keep the books wrong on purpose.
Everything we have said about Vallejo so far makes him look like the solitary artisan of his own dispossession. And yet, in España, aparta de mí este cáliz, the solitary voice gives way to something very different. In “Masa,” one man approaches the dead combatant, who goes on dying, then two, then twenty, a hundred, a thousand, five hundred thousand, then millions, and the dead man rises only when all the men of the earth have gathered round him. Somewhere between the enclosed, intensely private world of Trilce and the trenches of the Spanish Civil War, the “I” that suffers its way into existence becomes a “we” that can act. Vallejo’s movement out of the lyric self is not necessarily a departure from the radical syntax of Trilce. It may be one of its consequences: once language ceases to belong securely to a single speaker, it can begin to make room for others.
So here is my question for both of you, since you are two people speaking through the same book. Is Trilce already a collective poem that happens to appear as confession? And if so, what does it mean that its English is the work of two translators, with two sets of hands working on the same poem?
Translation is often imagined as solitary work: one translator alone with the dictionary, trying to decide what can be carried across and what must be left behind. Yours is a different kind of arrangement, and I am curious about the labor involved. How do you divide the work? Who decides when a line has been pushed as far as it can go? When you disagree, does the disagreement leave some trace in the translation, or does it have to be resolved into a single voice?
MDP Yes, Aleksey, maybe my comparison with Joyce is far-fetched and you are right. But the mixture of different aspects of the Spanish language made me think about a new language.
MLR Yes — the pivot for both Joyce and Vallejo is that fretting in the shadow of another tongue, with all the historical freight and accounting that goes with it. Vallejo bridges the personal and a wider alienation. His friends caught it in that phrase he used, “voy a mi pueblo” — communing with an acute sense of loss that is palpable in the poetry. Is it any wonder the language stutters? It is within the stutter and the abruptest falling away that the needle is threaded.
To your question — is Trilce already a collective poem that appears as confession? Certainly, for the “I” that suffers itself into being is never only one.
Which is why the English had to be two. The collaboration came after many a trial. I tilted at every word and line until I scaled the peak of poem LXXVII with an acute sense of failure. Failure isn’t a bad thing, indeed it is necessary here, when you consider that this collection on its arrival effectively fell “into a void”. How could it do otherwise?
Mario traced the impression of my rendering with a poet-translator’s sensibility and a lifetime’s grounding in Vallejo. Over two hundred footnotes followed, in which every missed wobble, stutter and spurt in the original was set against my translation. I had missed much — double, even treble readings. Vallejo is a master of shadow-play, relentlessly fracturing the language from within.
So we considered each nuance on its own terms and stoked the language of our translation accordingly. Who decides when a line has been pushed as far as it can go? Failure does. Any disagreement is the translation — it stays in the English as a wobble that both is and wasn’t resolved into a single voice.
Which brings me to a running approach we had — Vallejo, or the vital importance of awkwardness. During one exchange in which I asked Mario, “does that sound too awkward”, he responded, on-point, “with Vallejo there is no too awkward”.