‘Traduttore traditore,’ the translator is a betrayer. In other words, every translation is an act of treachery against the loved original, a stab in the back. If this Italian proverb is right, the translation I have just offered of the proverb itself must be just one more betrayal. Indeed, the case against me is strong. The Italian phrase gets much of its force from the jingling assonance of the two words, but one finds swiftly that it is no good trying to reproduce this with the weaker assonances available in English: ‘translator traitor’ and the like. Yet by a bald rendering something is achieved: the point or burden of the proverb really can be set out in another language. One is tempted to draw a conclusion of premature simplicity: matter is translatable, manner not. Therefore, jokes, puns (apart from lucky accidents), allusive titles and, above all, poetry will translate less well than, say, motorists’ manuals, left luggage information and realistic novels. It has been said that the natural unit of the ‘transparent’ novel is the event, set in a sequence of other events, while the natural unit of the poem is the word, in association with other words. That is why Tolstoy and Dostoevsky are read all over the world while Pushkin, despite the efforts of John Fennell, Antony Wood. DM. Thomas and Vladimir Nabokov, remains primarily a writer for readers of Russian. The myriad imperfections of rendering in any translation of a novel do not seriously impede what looks like genuine literary enjoyment: we weep for Anna Karenina and tremble at Raskolnikov. Don Quixote found readers everywhere: centuries later Lorca remained trapped in the brilliant liberty of his native Spanish.
The moral of this is that all the poets should, so to speak, be refused passports. We should cease to pretend that they can be translated. The reader of the new Oxford Book of Classical Verse in Translation can only be relieved that the English poets represented in it had never heard of that rule. Jonson, Dryden and the rest, like the bumble-bee who has never understood that his flight is contrary to the laws of physics, produce translations which take to the air successfully because they are unaware of the impossibility of what they attempt. As Virgil wrote, possunt quia posse videntur, meaning (if translation is after all permitted) ‘They can because they think they can.’ One must add, however, that just occasionally, although they think they can, in fact they can’t. Cowley’s Pindar is a case in point. The infant Hercules from the Greek ode becomes a Brobdingnagian monster: ‘The big-limm’ed Babe in his huge Cradle lay.’ This has affinities with the consciously coarse-lined, gigantesque drawings which Giulio Romano did for the Palazzo del Tè in Mantua, but it has little to do with Pindar. Elsewhere, when a certain grossness is needed, the chance is missed. Both Sir Robert Stapylton (1647) and Henry Fielding (1743) have a go at translating Juvenal’s ferocious account of the profligate Empress Messalina working as a prostitute in a Roman brothel and both suppress Juvenal’s reference to gilded nipples (‘papillis ... auratis’). As G.K. Chesterton once remarked, if a thing’s worth doing it’s worth doing badly. The rich variety of translations offered in the Oxford Book score hundreds of misses but thousands of hits.
If the sage, pusillanimous doctrine of abstention had been applied we would never have had from A.E. Housman (and from Horace) ‘The snows arc fled away,’ a poem which discovers in the coming of spring – in the opening leaves and the first warm suns of the year – sheer, terrifying despair. I have named the poem by giving its first line. This is a translation into five words of English of two words in Latin, ‘Diffugere nives’. A good Latinist will feel in the very first syllable, Di-, which is a mere prefix to the verb, ‘have fled’, a marvellously economic reference to the way snow melts variously, first here then there, lingering in shadowed places – roughly the force of the prefix in the English word ‘dis-tribute’. Latin is thought of as a heavy, architectural language, but here we have at once a lightness of reference which English simply cannot match. So, it might be said, Housman ran into the real impossibility of translation before he had completed his rendering of a single word. Housman wrote ‘fled away’, which has in it a late Romantic enjoyment of vacancy (or vagueness) which is not present in ‘diffugere’, but the sense of vanishing is utterly appropriate to the poem taken as a whole.
The truth is that in this case, across the centuries, the poem found its proper translator. It is, I think, the latent bleakness, the quiet irreligion, which found an echo in the atheism of Housman. Horace tells his well-born friend, Torquatus, that, once he has died, nothing can restore his life, not social class, not eloquence – not even his piety: ‘non, Torquate, genus, non te facundia, non te / restituet pietas’. Given the fact that ‘pius Aeneas’, ‘pious Aeneas’, has been named in the poem nine lines earlier, we can infer that Horace is here expressing a kind of dissent, is rejecting Virgil’s Aeneid, the greatest religious poem of Latin antiquity. In the Aeneid the hero, with his exemplary piety, conquers death by descending into the underworld and returning to the light, after which he sets in motion the triumphant progress of Roman history, a history in which he is, as it were, reborn in the form of the Emperor Augustus. In Virgil, piety gets you a long way. In Horace, it gets you nothing; we end in dust and shadow. But all is said from a full heart, not an empty one. Horace is moved by Virgil’s poetry just as he is moved by the coming of spring. All this is profoundly congenial to Housman. The element of homoerotic grief which Horace gives us at the end of the poem must then have had a clinching effect with Housman. It will be said that his actual rendering of the passage is weakly evasive:
And Theseus leaves Pirithous in the chains
The love of comrades cannot take away.
‘Comrades’ has a false heartiness, but a similar heartiness surrounds all those lads in Housman’s Shropshire poems, and one learns to find a special pathos in its evident falsity. Repression distorts, but at the same time it can strangely intensify poetic utterance.
Horace’s poem therefore found an almost ideal translator. Minute analysis could certainly show one failure after another in Housman’s version to convey this or that feature of the Latin. But if we invert the question and ask instead, ‘Is this, then, simply a free-standing English poem, having no connection with Horace’s ode?’ we shall at once be forced to answer, ‘No: it really is a translation.’ The area of overlap is greater, far greater, than the area of divagation. The Latin poem has reappeared in English and it is, once more, a poem. I want to say, wildly, that this is a miracle born of love. Lovers, we all know, betray themselves by a break in the voice. Housman’s lectures on Latin poetry were, it seems, grim stuff, highly technical investigations of minute textual and linguistic problems. On one occasion he seems to have allowed himself to consider ancient poetry as poetry. He chose ‘Diffugere nives’, and the mere reading out of the text caused him to be so overcome with emotion that he was unable to go on with the lecture.
Housman’s translation has found its way into the Oxford Book, but it has done so, so to speak, in the teeth of the gale. The taste of the editors favours earthy, late 20th-century renderings and vigorous, sinewy versions from the 16th, 17th and early 18th centuries (Dryden is everywhere). The 19th century is seen – with some reservations – as the lean time, the period of ‘translation on its knees’, producing reverential, vacuous sonorities.
Few would dispute that the 19th century was a great period of classical scholarship. I suspect that, partly because of the sheer quality of attention paid to ancient texts at this time, the body of texts must have seemed less bland, more charged with uncomfortable tensions, than is generally acknowledged. The readers to whom this applies were of course almost all male. Schoolboys who were brought up within a code of suffocating sexual propriety were given, as culturally authoritative reading matter, the extravagant indecencies of Aristophanes and Martial. Horace, who would have been ‘core curriculum’ because of the purity of his Augustan Latin, is far from pure in content; there are many lines in Horace extolling Stoic integrity, but still more on the Epicurean pleasures, drinking and love (both heterosexual and homosexual). One is sent to the Ancients to be schooled, and the schooling they offer turns out to be brilliantly shocking, almost libertine. Translation necessarily and obviously involves the substitution of one language for another. Less obviously, it also involves in some degree the substitution of the receiving culture for the original culture. Nineteenth-century English mores obliged translators of the period to bowdlerise and sanitise, so that Aristophanes sounds like Gilbert and Sullivan in the metrically exuberant, sexually discreet translation of Benjamin Bickley Rogers. But the under-presence of the ancient freedom is everywhere, for those with ears to hear. The charge of not-quite-perfectly-concealed, intense homoerotic emotion in Housman’s scoutmasterish ‘love of comrades’ is a good example. It might almost be said that the philosophic category of the aesthetic, conceived in the manner of the Late Victorian decadents as either amoral or immoral, is a product of 19th-century Hellenism.
Meanwhile the combination of the editors’ accusation of excessive reverence with their emphatic admiration of Dryden would amaze many Late Victorian classicists. For them it was the other way round. Arnold had taught them to look down on Dryden as a poet (Housman, who lived on until 1936, deeply admired the criticism of Arnold). The trouble with Dryden as translator was precisely that he made the Aeneid frightfully classical, as it could never have felt when Virgil wrote. C.S. Lewis, who carried this form of 19th-century sensibility even further into the 20th century than Housman had, hated Dryden’s rendering of Virgil’s words, ‘rosea cervice refulsit’, ‘She turned and made appear /Her neck refulgent’. He greatly preferred the 16th-century Scots of Gavin Douglas: ‘Her nek schane like unto the rois in May.’ The rose in Virgil’s ‘rosea’ is completely suppressed in the marmoreal pallor of Dryden’s language, but the real offence is that ‘refulgent’. ‘Refulgent’ in English is ponderously classical; ‘refulsit’ to a Roman was an ordinary word: it would have felt much more like ‘shone’. Lewis liked to argue that the poetry of Shelley was much more truly classical than the studiously echoic verse of Dryden, with its parade of knowing allusion. This is a version of the old, deep translator’s paradox set out by the Count in Castiglione’s Il Cortegiano: ‘If we follow them of old time, we shall not follow them’ – meaning that if we try too hard to be like the Ancients the difference between them and us will become inescapable; it will be clear that we are writing imitative poetry and that they were not. Johnson, borrowing a phrase from the Emperor Augustus, said that Dryden found English poetry made of brick and left it marble. No wonder, Lewis might say, he reduced the green, living poetry of Virgil to a petrified magnificence. Each age has its peculiar movement of taste and strong tastes almost always involve, together with a newly focused perception, a silent aversion of the mind from previously acknowledged truths. The Oxford Book has averted its editorial mind from the way the 19th century felt to the 19th century.
But C.S. Lewis was only half-right about Dryden. It is just false to say that Virgil did not feel at all ‘classical’ to the Romans. The Aeneid is written not in native Latin metre but in august Greek hexameters. Moreover ‘refulsit’ is not quite the ordinary word for ‘shone’, which would be ‘fulsit’, but is a mild poeticism. Greek metre depends on quantity, that is the length in time of the syllable; Latin (like English) had stresses, as distinct from quantitative accents, and the occasional clash of stress with quantity produced difficulties – and sometimes metrical pleasures – unknown in Greek. By the time of Virgil the Romans had become accustomed to the Homeric hexameter, but it must still have felt grandly artificial, faintly alien. Horace’s Alcaics, Sapphics and the like, meanwhile, were, because newly adapted, vividly Greek. So it may be that Augustan Latin poetry actually did feel classical, even at its first appearance.
This takes some of the sting out of the Count’s remark. The queer feeling of being in limbo, suspended between two cultures, two languages, may not after all be a necessary treachery produced by historical distance, a truancy from the original; there may be an element of sheer fidelity in this very intermediacy. Given that classical Latin poetry is subtly lit from behind by Greek one begins to wonder about the widely held doctrine that in any good translation we must always feel that we are reading natural English, not ‘translationese’. In fact our literature has been enriched by those obstinate souls who were willing to violate the natural harmonies of the receiving language by importing, bodily, alien structures from Latin. The Oxford Book offers plenty of examples from both sides. Cecil Day-Lewis’s translation, from Virgil’s Georgics,
She never saw,
Poor girl, her death there deep in the grass before
her feet ...
is completely natural English; it gives no discomfort. But Milton’s translation of Horace’s ‘Quis multa gracilis’ is very different. In this ode the poet imagines his lady, Pyrrha, in the arms of some smooth young man and, instead of raging jealously, pities his rival – the lady will be false to him as she has been to the poet:
O how oft shall he
On Faith and changed Gods complain: and Seas
Rough with black winds and storms
Unwonted shall admire:
Who now enjoys thee credulous, all Gold ...
This is not easy because in a sense it is only half-translated. The crunch comes in the last line quoted, which is in real danger of being unintelligible. Because Latin is an inflected language, it is perfectly clear that it is the lover who is credulous and the lady who is golden. In Milton’s English it is not grammatically apparent that ‘credulous’ applies to one party and ‘all Gold’ to the other. In James Michie’s Penguin translation (not in the Oxford Book) all these difficulties are smoothed away:
He’s still credulous though, hugging the prize he thinks
Pure gold ...
No doubt many readers will prefer Michie’s version. To me the Milton remains strangely, obstinately marvellous. Horace planted a special sort of pain in his poem by placing ‘credulous’ next to ‘aurea’; this collocation Milton, at the cost of some trauma to his own native language, has chosen to preserve.
If we translate we domesticate, and to domesticate is to tame. Milton as it were refused to tame the beast called Latinity. I have argued from the presence of an analogous Graecism in the original Latin poetry for a kind of propriety in this half-translation. But what of Greek poetry? No such excuses are available there, it would seem. A.E. Housman’s hilarious parody, ‘Fragment of a Greek Tragedy’, is not included in the Oxford Book, yet it is a mosaic of weirdly literal translations from Aeschylus and Sophocles. Lines like ‘sailing on horseback, or with feet for oars?’ are ludicrous but in a certain sense they are thoroughly Greek. If we want from a translation not just an extension of English literature but some notion of the way images and concepts are really employed in the more distant culture, we shall find in Housman’s ‘Fragment’ a peculiar excitement, behind the confessed absurdity. We all agree, however, that this just will not do as serious poetry. But Milton’s lines in Samson Agonistes:
But who are these? For with joint pace I hear
The tread of many feet steering this way
similarly sacrifices English expectations to Greek habits of imagery and as a result we experience a real modification of our culture from outside. This cannot be done without a shock to the system. I have called this ‘half-translation’, but of course if we switch our aim from producing an end-result which is fully English to the other task of the translator, which is to fetch things from the alien literature and set them before us, so that we can (if appropriate) marvel at their foreignness, we shall have to say that the Miltonic metaphrase is actually fuller, a more complete translation than all those smooth renderings in which the smoothness is achieved through an endless, wholly invisible suppression of features of the original. Just as languages are enriched by calques – secondary senses lifted bodily from other languages, such as ‘dry’ of wine (from French) or ‘numbers’ of verse metre (from Latin) – so literatures are enriched by the occasional well-judged importation of an alien literary idiom; ‘felix qui’ becomes in many English poems not ‘he is lucky who’ but ‘happy the man who’. There is a special literary pleasure in the interlinguistic quality of the phrase but it may well be that when we finally reach the point at which all knowledge of Latin and Greek is lost this pleasure, too, will die. Certainly Kipling understood it perfectly in his brilliant, only half-humorous poem about the proper superiority of literature to science in the curricula of schools: ‘There are whose study is of smells.’
The early translators in this collection are for the most part natural masters of the interlinguistic. But Eliot in the 20th century clearly felt its strange, oblique attraction. The unpunctuated interrogatives in ‘What seas what shores what grey rocks’ (‘Marina’) can be paralleled in Housman’s parodic ‘Fragment’:
Wherefore seeking whom
Whence by what way how purposed art thou come?
The Housman is designedly grotesque, but Eliot’s line is entirely beautiful. Both draw their formal character from Greek tragedy, even though in the case of Eliot it was transmitted through Seneca. Or take these lines from ‘Sweeney Erect’:
Display me Aeolus above
Reviewing the insurgent gales
Which tangle Ariadne’s hair
And swell with haste the perjured sails.
Reading this is like finding a 20th-century artist who can, when he pleases, draw like Correggio. Most of the modern translators excel – and I must allow that they really do excel – in an opposite manner. Andrew Miller does Paul the Silentiary, Tony Harrison does Palladas and Humphrey Clucas does Catullus into a harsh but wholly English English which is true to the fiercely direct sexuality of the originals. We in the 20th century can do what Pope, with all his dexterity in bawdy, could not. It remains slightly disquieting to find that one 20th-century translator thinks he is quoting Horace when he writes ‘nunc pede liberum’ (‘liberum’ makes no sense; it should be ‘libero’, but Robert Lowell needs ‘liberum’ to rhyme with ‘rhythm’ at the end of the next line). Adrian Poole and Jeremy Maule allow themselves a similar donnish sharp intake of breath, in their Introduction, at the notorious howlers of Ezra Pound.
The compilers of the Oxford Book have made an admirable collection for our time. Presumably, with the decay of classical learning, the pieces in this book will be read less and less as versions, more and more as poems ‘in their own right’. It is a cheering thought in the circumstances that so much of the book can in fact be read in this post-classical way, with great enjoyment. Everyone will have his or her own list of regretted omissions. I was sorry to see no Stesichorus, and still sorrier that no space had been found for the extraordinary rendering of Lucretius in Tennyson’s poem of that name. Ben Jonson’s ‘Drink to me only’is not included, perhaps because it is taken not from a Greek poem but from a Greek prose work. On the other hand I welcome the decision to print lots and lots of Martial and hardly any Pindar. The Ancients themselves would be amazed at this. Pindar was thought supreme. He does not translate however, as it seems to me, and Martial does. Which perhaps brings us back to the fear I expressed at the opening, the fear I thought I had dispelled, that poetry is ‘what gets lost in translation’.
When Horace wrote his odes he saw himself as a translator, not of the thoughts and images of Sappho and Alcaeus, but of their metres. English attempts to mirror, with the thumping stresses of our language, the subtle quantitative hexameters of Virgil are, by and large, not a success. The stressed English hexameter is, as Clough knew, a metre for comedy. Robert Bridges’s attempt (yet once more not included here) to put Virgil into quantitative English hexameters – the length in time of the syllables being played, as in Latin, against the natural stress – produced a result so subtle as to be almost inaudible as metre. The title ‘Ibant obscuri’, ‘They went obscure’, was inadvertently prophetic. Robert Louis Stevenson on the other hand tried to do as Horace did – to translate not matter but metre, in his ‘Alcaics to Horatio F. Brown’:
Brave lads in olden musical centuries
Sang, night by night, adorable choruses,
Sat late by ale-house doors in April
Chaunting for joy as the moon was rising ...
These are stressed Alcaics and so they jingle as Horace did not. Also, it will be noticed, Housman’s lads are back again. But it is haunting. It is not Horace, but without Horace it could not have happened. Only one piece of Stevenson in the Oxford Book (my last ‘regretted omission’), but the volume is crammed with good things.
Send Letters To:
London Review of Books,
28 Little Russell Street
London, WC1A 2HN
Please include name, address, and a telephone number.