Rethinking the Art of Translation

Faithfulness is still considered the defining feature of translation, while creativity is reserved for original authors. That needs to change. By

One version of the original. Paula Modersohn-Becker: Portrait of the poet Rainer Maria Rilke, 1906 (detail). Source: WikiCommons

In 2006, the Scottish poet Don Paterson produced what he called a ‘version’ of Rainer Maria Rilke’s Sonnets to Orpheus (German: Sonette an Orpheus). He took great pains to frame his work as something separate from translation, focusing on his own creative process in crafting English poetry from Rilke’s German sonnets. Here is one of his poems, alongside a conventional translation from 1987:

A tree rose from the earth. O pure transcendence–
Orpheus sings: O tall oak in the ear!
All was still. And then within that silence
he made the sign, the change, and touched the lyre.

One by one they crept out from the wood,
emptying each set and form and lair;
and looking in their eyes, he understood
they’d fallen quiet in neither stealth nor fear,

but in their listening. Growl and bark and roar
died in their breast as each took to the clearing.
Before this day, there hadn’t been a shack

that might have held the song, a plain earthwork
hollowed by their most obscure desire:
today the temple rises in their hearing.

Don Paterson, Version, Sonnets To Orpheus, 2006

A tree stood up. Oh pure uprising!
Orpheus is singing! Oh tall tree in the ear!
And everything grew still. Yet in the silence there
changes took place, signals and fresh beginnings.

Creatures of stillness crowded from the clear
untangled woods, from nests and lairs;
and it turned out that their light stepping
came not from fear or from cunning

but so they could listen. Shriek, bellow and roar
had shrunk in their hearts. And while before
there was scarcely a hut where they might stay,

just a shelter made of the darkest cravings
with shaky posts for an entrance-way –
you made a temple for them in their hearing.

David Young, Translation, Sonnets To Orpheus, 1987

This is the original poem by Rilke:

Da stieg ein Baum. O reine Übersteigung!
O Orpheus singt! O hoher Baum im Ohr!
Und alles schwieg. Doch selbst in der Verschweigung
ging neuer Anfang, Wink und Wandlung vor.

Tiere aus Stille drangen aus dem klaren
gelösten Wald von Lager und Genist;
und da ergab sich, daß sie nicht aus List
und nicht aus Angst in sich so leise waren,

sondern aus Hören. Brüllen, Schrei, Geröhr
schien klein in ihren Herzen. Und wo eben
kaum eine Hütte war, dies zu empfangen,

ein Unterschlupf aus dunkelstem Verlangen
mit einem Zugang, dessen Pfosten beben, –
da schufst du ihnen Tempel im Gehör.

Rainer Maria Rilke, Die Sonette an Orpheus, Erster Teil (1922)

Young’s translation is slightly closer to the terms used in Rilke’s original. The final line of the first stanza, to take one example, mentions ‘neuer Anfang, Wink und Wandlung’, which is much closer to Young’s ‘changes […] signals and fresh beginnings’ than Paterson’s mention of a lyre. When looking at the two as a whole, however, it’s hard to see any major differences: they are both recognisably taken from the original.

So why is this worth discussing? Looking at Paterson’s attempts to define his version as a separate category from translation really emphasises some of the problems with how we define translation – particularly when it comes to poetic translation.

Translation is usually defined as a kind of equivalence: a ‘faithful’ translation reproduces the original accurately in another language. The practical reality of translation, particularly non-literary translation, does work like this – someone translating an instruction manual is simply aiming to convey the same instructions as the original.

When we look at literary translation, however, things get a little more complicated. The idea of accurate reproduction not only obscures the extent to which it’s impossible to truly create a ‘faithful’ translation, it also reduces the act of translation to little more than an empty vessel for conveying the original author’s ideas. I’d like to suggest that, even if nobody is going to be discarding the practical definition of translation any time soon, it is worth acknowledging that there is more going on in the act of translation than this common metaphor suggests. By confining ourselves to concepts of singular authorship and strict definitions of originality, we are missing out on productive conversations about the interrelatedness of all kinds of texts.

The first and most fundamental question is, of course, what exactly is being translated, a question which is very closely embedded in discussions of where exactly the meaning of a text lies. It seems obvious that the process is not simply word-for-word, but calling it something like sense-for-sense if very hard to pin down.1 A translator has to consider a bewildering range of factors: the original text; the cultural context in which it was written, and the context of the time period and place in which it is likely to be read; even the needs of the publishing industry (such as if it will be read in a popular or academic context). To take a very conventional example, translators of Harry Potter have to decide if they want to translate the many British concepts that litter the text. The Chinese translation by Su Nong explains the cereal Cornflakes in a footnote, while the Israeli translator Gili Bar-Hillel chose to convey a sherbet lemon as a chocolate sweet; some transplant the setting into a boarding school in their own country, while others choose to keep it all on British soil.

And these kinds of negotiations are only the start. When we look at poetry – such as Rilke’s Sonette an Orpheus – the tensions really come to the surface. There is a lot of debate about if poetry can ever be properly translated, because the meaning is conveyed by much more than simply what the words say. 2More so than prose, poetry is about specific word choices, rhythm and rhyme, and a whole host of linguistic and stylistic decisions. Broadly, translators usually have to choose between conveying the meaning of the words and losing something of the structure and style, or conveying the style with a looser translation of the meaning.

Poetry is often considered untranslatable precisely because people focus on the idea of accurately reproducing an original. To get around this limitation, poetic translators take different approaches. To come back to Don Paterson, he wrote an entire afterword to Orpheus explaining the difference between his ‘versions’ and conventional translations:

[A]rtists also put themselves in the way of a dangerous kind of sympathetic resonance. (Of which Rilke writes approvingly, and rather unwisely, in ‘be the glass that shatters with its own ringing.’ ‘Zerschlug’ I was tempted to render as ‘shiver’ or ‘tremble’, as a shattered glass is no use to anyone.)

He places the emphasis firmly on his own creative choices and judgements, his own responses to Rilke. It is hard, however, to see this as anything more than a more confident version of translations taking the translator’s own interpretation into account. Paterson may be moving beyond the usual meanings of zerschlug, but the result is still a combination of the original text and his close engagement with it – as is the case with translation.

Paterson only tries to separate his work from conventional translation because he takes the usual view that translation is derivative: if he wants to be regarded as an author, not a translator, he has to be seen to be doing something more. In his afterword, he says that ‘translations date. Free versions, on the other hand, can never be definitive; but nor do they date in quite the same way.’

This comes back to the question of what a translation is conveying. New translations are published because older ones appear out of date in a way which originals do not: originals reflect their own time period, whereas translations are a negotiation between the original time period and the one in which the translation is being written. (Wolfgang Krege’s translation of The Lord of the Rings in 2000, for example, sought to update the language by using German slang: Sam calls Frodo Chef or boss, and Pippin references the lost property office.) Paterson wants to place his Orpheus in a category of works that don’t become dated after a few decades; he wants the poems to be judged in their own right, rather than by how well they reflect people’s expectations of what Rilke should sound like in English.

People’s expectations of poetic translation change. Earlier translations of Rilke rhymed; by the early 1980s, however, the translator Albert Ernest Flemming was arguing that making Rilke rhyme in English was an ‘impossible task’ that only leads to choosing ‘wrong or unsuitable words’. Does this mean that earlier rhymed translations should be a separate, less faithful category? And how far can Paterson’s versions actually be separated from previous translations?

If we come back to the Rilke poem I cited earlier, it’s clear that Paterson isn’t doing anything categorically different to previous translations. The changes he makes – such as adding the lyre to the first stanza – still reference the original work and its context. His version is still a combination of Rilke’s original and his own interpretation, and still takes into account all the other factors – even previous translations. To him, the notion of translation can’t properly convey his level of engagement with the text, with existing English translations of it, and with his own authorial decisions, such as how he renders words like zerschlug.

Paterson’s attempt to create a ‘version’ rather than a translation emphasises how difficult it is to define translation in terms of faithfulness, to see it as a very different process to the originality of authorship. Any description of translation as ‘faithful’ fails to encompass everything that happens when someone tries to convey a poem in another language. We need a less restrictive concept that better conveys all the processes at work within translation. Rather than viewing translation as between two languages, I think we should try to see it in terms of the prefix ‘trans’: translation, transformation, even transgression. This not only covers the idea that translation moves ‘across, from one place, person, thing or state to another’, but also encompasses ‘beyond, outside of, transcending’.

This image of multi-direction connections is embedded in discussions of translations, but is often ignored in favour of thinking about translation as simply ‘between’ the original and the translation. The concept of ‘trans’, however, lets us see all the connections between translation and original, author and translator. It acknowledges Paterson’s creative input without having to separate what he does from translation. It refutes the idea of a single authorial influence over a text, seeing translations not in terms of faithfulness, but with a recognition of everything that influences them. It is not simply a one-way transmission from original to translation, nor even a dialogue between the two – it encompasses everything that influenced the original author and the translator, and even previous translations.

It is not too far distant from the way we consider original works in light of the other works and authors than influenced them; both are an exchange of voices. There are countless texts that retell Shakespeare, for example, who himself was retelling – transforming – earlier texts.

We should see translations as multiple rather than derivative, a license to reinterpret that is not so far removed from those retellings of Shakespeare into our current cultural context. Translators not only respond to the original, but also to other translations and texts, as texts respond to other works.

It is true that translation is a bit of a different process to creating a work without a direct source. However, much like Paterson reworking Rilke’s sonnets isn’t so far removed from Young translating them, so is Paterson not so much further removed from Margaret Atwood writing a female-centred version of the Odyssey in The Penelopiad, or Jean Rhys rethinking Jane Eyre in The Wide Sargasso Sea – both of which are judged as original works of authorship.

When we judge translations by their faithfulness, we lose the opportunity to more fruitfully examine the myriad transformations at work within the process of translation – meaning poets have to go to the lengths of Paterson creating an entirely separate category of ‘version’ to have his work judged as poems in their own right. When we allow works to transgress the boundaries between translation and original, when we see them as a continuum of related subjects rather than as superior and subordinate categories, we will open up more productive discussions of translation.

  1. See The Principal Works of St. Jerome: Letter LVII. To Pammachius on the Best Method of Translating, translated by  Philip Schaff and Henry Wace: „For I myself not only admit but freely proclaim that in translating from the Greek (except in the case of the holy scriptures where even the order of the words is a mystery) I render sense for sense and not word for word“.
  2. See Matthew Reynolds, The Poetry of Translation.

Leave a Reply

Your email address will not be published. Required fields are marked *