There’s a well-known saying that poetry is “what is lost in translation”. If I set aside any number of sophisticated caveats, I think there’s something essentially fair about that. In poetry above all, I don’t think you can really say that you’ve read a given work, or a particular poet, if you’ve only read a translation, though you might of course — if you’re lucky — have read a good poem in its own right. The translation of poetry is peculiarly tantalising, because it points always, even more than other kinds of translation, to its essential impossibility, the gap that can’t be bridged.
I think a lot about translation both for pragmatic reasons — because professionally I work almost exclusively on early modern Latin which hasn’t been translated before — and because I find verse translation particularly creatively satisfying. There’s an obvious justification for quite pedestrian types of translation in genuinely bilingual contexts: things like Loeb classical editions, the Murty library, older Penguin classical editions some of which were in parallel text, or indeed the Penguin edition of Herbert for which I translated all the Latin and Greek verse years ago. Any edition of a text which presents the original alongside the translation has to keep uppermost the needs of readers who have some knowledge of the language of the original and are using a translation to pick their way through it. You are doing readers a disservice in this context if you stray too far from the basic meaning of a text in your efforts to represent other elements of it (such as style or sound or form), though even here the very best translations — as I discussed briefly last week in relation to a fascinating French anthology — at least attempt to keep all aspects in view. This kind of translation ‘points towards’ that unbridgeable gap in the most straightforward kind of way: it constantly asks the reader to move back and forth across the gutter of the page to the other side.
What I find rather dispiriting about the majority of translated poetry in English at the moment is that most of it reads like those ‘parallel text’ translations, even though in almost every case the original text is not provided, and there is no real expectation that readers will have any knowledge of, or be referring to the original. I’m hardly the first to complain about this. Geoffrey Grigson, in the bracing introduction to his 1973 Faber Book of Love Poetry, explained why he had decided not to included anything in what he called ‘the unmeasured, thin-rolled short crust of translation – useful, but now so easily accepted in itself as verse’.1 Instead, he just went ahead and printed a good deal of medieval and early modern French verse without translations at all. I can’t (understandably) imagine any press agreeing to this now. Don Paterson says something related to this in his Aphorisms (2000):
Translated verse is usually given away by the strenuous informality of its delivery. As if you had presented your passport and visa before anyone had asked for them: such behaviour only arouses suspicion.
I’m not sure quite when the near-consensus emerged that the proper way to translate poetry is always into the kind of lineated prose that would be appropriate in a parallel text edition. Here’s a rather fascinating comment from an ancient British textbook, Essays in Translation from French, first published in 1941:
In translating verse we should have preferred to use prose. But the result was always a hybrid genre, to which a verse rendering, with all its drawbacks, seemed in the end preferable.2
Two things struck me here: firstly, because I can always be distracted by grammar, the now pretty much obsolete observance of the shall/will rule for the first person singular and plural.3 Would anyone still write ‘should’ here rather than ‘would’?
The second was this rather useful idea that a prose version of a poem creates a hybrid genre, neither poetry nor prose. (I think we can include here the sort of ‘lineated prose’ in which most modern English translations are presented.) Richie and Simons’ book is the very definition of a parallel text edition — the whole point is to compare the facing pages of French original and English translation in detail, with the help of the notes — but back in 1941 they weren’t comfortable about putting verse into prose, even for these didactic purposes, and despite all the obvious advantages (“we should have preferred to use prose”).4
These days nearly all published translations, regardless of the form of the original, and however purely ‘literary’ (rather than educational) the context, are now in the kind of ‘hybrid’ genre which these editors rejected. Sometimes, of course, this works perfectly well: if the original is a prose poem, it would obviously be silly to translate it into rhyming couplets. If the most distinctive features of the original are particular images, or juxtapositions, or metaphors, then these too may come across pretty well in a prose version. But a lot of poems don’t work in those ways — and, generally speaking, the proportion of poems that work primarily in those ways is a good deal smaller in languages other than English than it is for Anglophone poetry today. Something interesting is going on here, and it must be related, I think, to a contemporary Anglophone discomfort with verse, and a creeping sense that what distinguishes a ‘poem’ is often, or mainly, or usually its particular (prose) style — what in old-fashioned terms we’d call its genre of rhetoric, rather than any particular formal features.5
But enough complaint, because there’s another way in which poetry and translation are connected, and in this respect we might almost say that poetry is what is gained by translation. Poets are (almost always) translators. It is hard to think of a really important poet who has not also translated poetry, though what this means naturally varies widely. For some poets (like John Dryden, John Ashbery, Richard Wilbur, or A. E. Stallings) translation or paraphrase or ‘rewriting’ other poetry has been a large part of their work.
The collected works of a great many poets contain examples of translation or paraphrase if you look carefully: sometimes these are just a handful of experiments, often at the back of the volume or scattered around, not always indicated as such by notes or titles — so we find bits and pieces of translation in poets as various as Eavan Boland, Rudyard Kipling and Thomas Wyatt. The lack of indication might be because we are (or were) expected to recognize a canonical source, such as a psalm or a passage from Virgil. I mentioned when I wrote about Horace that until fairly recently, pretty much every major poet in English had had a go at at least one Horatian ode, for example. (This would be true of poets in many other European languages as well.) In early modernity, it’s hard to find a poet who didn’t produce at least one psalm paraphrase.
Verse translation is in that sense a traditional writing exercise, a form of discipline, and clearly one that most poets have found at least sometimes salutary. Don Paterson, again, in a recent conversation printed in Poetry Review, admitting that he had ‘traded on identity points’ when he was younger, adds, half-jokingly perhaps but not I think wholly, ‘I had to go and translate Machado as penance’. The recent publication of Seamus Heaney’s collected translations has brought their extent and, as it were, persistence in the course of his career into focus for many readers who perhaps hadn’t thought much about Heaney as a translator before.
I have written previously about C. H. Sisson, a rather unfashionable poet whose reputation has survived, insofar as it has, largely on the basis of translations, many of which were commissioned. I don’t think Sisson’s translations are his best work, but it’s obvious, if you read through his poetry as a whole, how influential they were upon the rest of his work. Something similar is true about Bunting: his ‘overdrafts’ of Latin and French verse (I’m not qualified to comment upon the versions from Persian) are mostly not quite successful as poems in their own right, but as commentaries upon and interrogations of the original texts they are often extraordinary. From that interrogation emerged, above all, Bunting’s two remarkable books of Odes, the most authentically ‘Horatian’ verse, for my money, in all of English literature, and as a result quite unlike anything else.
Poets just do translate in a way that simply isn’t true of novelists or playwrights. No doubt there are practical reasons for this, too (most poems are a lot shorter than a novel, and commissions are always welcome), but I think it’s more than that. All literature makes use of the language it is written in: exploits its distinctive strengths and peculiarities. But poetry (good poetry, at least: real poetry) is always in part about the language it is written in in a way that I think it is not required for good prose. You can learn all you need to know about how to write well in your own language from reading in it very widely (and practising, of course). But it is very hard to interrogate a language itself, to feel for its boundaries and perhaps push at them, to taste what is particular to it, if you have no sense of an alternative.
Gwyneth Lewis articulates a version of this idea in her comments on writing poetry, as reproduced in the — fascinating and highly recommended — Bloodaxe anthology Strong Words:
Here I take it that the whole point of translation is to introduce a new element - of rhythm or thought - into a literary tradition. The point isn’t to produce a version so culturally smooth that nobody would ever guess it was imported. There has to be something strange, novel and fascinating either about the style or cast of mind of the new piece.6
Sometimes the alienness of the original message, as well as its language, is part of the point. Just after 9/11 Seamus Heaney published a poem, ‘Anything can happen’ which is actually quite a close version of Horace Odes 1.34.7
Anything can happen. You know how Jupiter
Will mostly wait for clouds to gather head
Before he hurls the lightning? Well, just now
He galloped his thunder cart and his horses
Across a clear blue sky. It shook the earth
And the clogged underearth, the River Styx,
The winding streams, the Atlantic shore itself.
Anything can happen, the tallest towers
Be overturned, those in high places daunted,
Those overlooked regarded. Stropped-beak Fortune
Swoops, making the air gasp, tearing the crest off one,
Setting it down bleeding on the next.
Ground gives. The heaven’s weight
Lifts up off Atlas like a kettle-lid.
Capstones shift, nothing resettles right.
Telluric ash and fire-spores boil away.
The poem is about being shocked by events, by a change of direction or a twist of fate which in its suddenness and violence feels like the intervention of the gods. In the Latin original (which you can read here), the examples of unexpected events and changes of fortune takes up stanzas 2-4: this is the core of the poem and the part Heaney has adapted in his stanzas 1-3.
But Horace frames his poem with an opening verse describing the poet’s own historically half-hearted piety, and what seems like an experience of ‘conversion’, of returning to religious observance. This is Conington’s translation from the late nineteenth century:
My prayers were scant, my offerings few,
While witless wisdom fool'd my mind;
But now I trim my sails anew,
And trace the course I left behind.
The implication, though it is not spelt out, is that the speaker has been shocked by some reversal of fortune, whether of good or back luck, into greater religious observance. I have been led astray, he admits, but now I am turning back to righteousness. Heaney has left this opening stanza aside, which is interesting — it’s much more common to start a poem with a passage of translation or allusion, and then move away from it. Anyone who knows Horace well enough to recognise the borrowing is, however, likely to remember how it begins. (It’s also one of the relatively few biographical or quasi-biographical fragments in the Odes, so memorable too from that point of view.)
Heaney’s poem, it seems to me, is quite provocative either way. Without the Horatian opening, the poem risks making 9/11 sound more like an unexpected natural disaster than a wilful act, especially if we do that automatic thing of “reading out” the classical gods (“You know how Jupiter . . . "), viewing them as so much stage machinery, or grandiose metaphors for annoyingly extreme weather. The final stanza Heaney added at the end seems designed to recognise both that everything has changed (‘nothing resettles right’), and to make the transformation sound at once enormous and oddly impersonal. But if we hear the silenced opening, as I think we are meant to, we are at once challenged (how do we respond to what has happened? are our hearts changed?) and reminded what piety of certain kinds can lead to — it puts the people that planned and carried out those attacks as well as all those who responded to them, back into the poem.
Overall, I think the poem is a very uncomfortable one, and the discomfort arises in large part from being a translation of a poem about a type of religious experience and understanding that is alien to us. It’s as if Heaney’s choice of Horatian text — his implicit borrowing of an Horatian perspective — allows into the poem something he doesn’t quite want to acknowledge: the impressive otherness of it all.
A bit of housekeeping this week. As a few people have now “pledged” subscriptions, I have turned on paid subscriptions for Horace & friends. Everything is still free, and I have no plans to paywall any content for now. This just means that if you are burning to pay me, you may now do so . . .
Geoffrey Grigson (ed. and intro), The Faber Book of Love Poems (Faber and Faber: London, 1973), 15.
Essays in Translation from French, by R. L. Graeme Richie and Claudine I. Simons (CUP, 1957), but the introduction is dated July 1941, the date of the first edition.
I was taught this rule in the first year of secondary school in 1990, but I must surely have been one of the last. If any reader younger than me was also taught it, do add a comment! For anyone in the dark, the rule is (or rather, was) that shall/should is used ordinarily for the first person singular (I) and plural (we) vs will/would in the other persons — so “I shall, you will, he will, we shall, you will, they will” — but that this is reversed for emphasis. “I shall go the shops tomorrow”, but “I will go to party tomorrow, even though you don’t want me to”.
I picked it up second-hand of course, but it’s actually a great little resource for any native English speaker trying to develop a feel for, and ability to distinguish between the variety of styles and diction in literary French.
In an interesting recent Twitter/X thread, Ben Wilkinson talks about how the ‘uniquity’ of poetry as a form is currently not very obvious. As far as I can tell, this is much more true of Anglophone poetry than contemporary poetry in other languages, but a well-informed comparative discussion would be fascinating.
Strong Words: modern poets on modern poetry, ed. by W. N. Herbert and Matthew Hollis (Bloodaxed, 2000), pp. 266-7.
The poem was first published in the Irish Times on 17th November 2001, with the title ‘Horace and the Thunder’.
I started secondary school in 1996 and learnt the shall/will distinction! It appears rather handily in Blake's "And did those feet" - "I will not cease from Mental Fight, / Nor shall my Sword sleep in my hand"