Translations are like women: when they’re beautiful, they’re not faithful, and when they’re faithful, they’re not beautiful,’ wrote Carl Bertrand in the introduction to his late nineteenth-century French translation of Dante’s Divine Comedy.
Like most aphoristic truths, it exaggerates, but my possibly too faithful translation reveals that in English ‘beautiful’, unlike ‘faithful’, has a more restricted meaning than it has in French. Can a translation be beautiful in English? Is it even noticed?
To be sure, translation is not a flamboyant activity. Its practitioners are not required to dress either fashionably or in rags, to lead protests or disdain society altogether, or of course to drink absinthe late into the night. They are never extreme in their life style, because visibility is not only unnecessary – it is almost impossible to achieve. This is true even in countries where most books – certainly most novels – are in translation. It is truer still in the English-speaking world where translation is rare and mainly restricted to the classic novels of the nineteenth century.
This does not mean that translators are all self-effacing Saint Jeromes drudging away with no other thought than the propagation of great literature into other languages and cultures. On the whole, I suspect that many or indeed most of them resent the failure of readers, critics and indeed writers to acknowledge their presence in a work (with a few exceptions). A translator’s elegant sentence is exactly that, and the elegance may not have been present in the original. Equally translators must take personal responsibility for the clumsy or confusing sentences that issue from their pens or word processors. But then, what profession is there that does not feel undervalued? Translators should perhaps savour the freedom of their métier and its inconspicuousness.
And I know that I will have raised the blood pressure of not a few of my fellow professionals by insisting on calling translation a craft, while they would much prefer the more exalted category of an ‘art’, a palliative for their long hours and poor remuneration, which most certainly should be increased. A craft does not presuppose less skill than an art. Technically it is far more difficult to translate a poem than to write one, and translating a novel requires almost the same technical skills as writing one, but ultimately a translation is a copy.
It is, however, a very strange kind of copy, and this perhaps explains the persistent attempts to make an art of translation. If you copy a painting, you get out your easel, place your canvas on it and proceed to apply paint of the same consistency in the same colours as the original artist, hopefully in the same manner. If you copy a novel into another language, you are using a completely different material – one that renders it indecipherable to most of the speakers of the original language. To say that translation might be like copying an oil painting into watercolour would not do justice to the complexity of the process. It is difficult for monoglots to understand just how different languages are. Students at evening classes will often ask impatiently, ‘Why do they do it like that?’ With a single linguistic template it’s natural to perceive one’s own language as the standard format for human communication. Not only do languages do things differently (syntax and morphology), they also contain surprisingly different vocabulary.
In the sixties, a popular work on sociolinguistics claimed that Inuits have many words for snow, and it became a commonplace. More recently this highly probable factoid has been challenged. Whatever the truth of the matter, the point is well made: languages categorise meanings, in entirely different ways. In one language, there may be greater categorisation in a particular area, while in another, there may be greater use of collective nouns. Nor are abstract concepts free from this kind of difference. Until recently, there was no word for privacy in Italian. Intimità, as in English, was generally associated with human relationships rather than the relationship between individuals and their property. The adjective privato existed (as in proprietà privata, private property), so it was not too difficult for someone to coin the new word, privatezza, although I have never heard anyone make use of it. Finally someone just put out their hand and grabbed the English word, and ‘privacy’ became la privacy (that same person presumably also had to invent the new word’s gender in Italian). We have here the three methods of inventing new words: expanding the semantic field of an existing word, coining a new word from an existing root, and bringing in a loan word. This demonstrates the problem for a translator but not the solution, because a translator cannot indulge in linguistic invention in the way a writer can. Translation is like a jigsaw puzzle in which the pieces from two jigsaws have been mixed up.
Consider the following experiment in which researchers had three cards: one was a picture of a man (A) kicking a ball, another a picture of the same man (A) about to kick a ball and another a picture of a different man (B) kicking a ball. A group of English-speakers and a group of Indonesian-speakers were asked to pair off two of the pictures. Nearly all the English-speakers paired off the two different men (A and B) both kicking the ball, and nearly all the Indonesians paired of the same man (A) kicking the ball and about to kick the ball. These extraordinary results would appear to reflect the different tense usage in the two languages. English has clearly defined tense usage which is adhered to fairly carefully, and generally Indonesian does not use tense, although it does have some marker words that are very occasionally used. Thus the expressions ‘he’ll go’, ‘he’s going’ and ‘he went’ will follow the simplified pattern of ‘he go’. I have already argued that language is all about categorisation and this applies to syntax too. The manner in which a language categorises the world could surely affect the manner in which a speaker categorises the world. This is one of the fundamental ways in which language governs how we think, although the effects of language on our thought are almost certainly more complicated than suggested by this experiment, particularly in the area where language comes closer to wider cultural questions, such as the structure of argumentation, formal relations between social groups, hierarchy and the like.
The aforementioned researchers formed a third group of bilingual English and Indonesian speakers, and the result was a mixture. The tests also showed that bilinguals behaved differently according to whether they were interviewed in English or Indonesian. The results are, I think, quite convincing, but they do not fully convey the subtlety of linguistic difference, which is probably impervious to sociolinguistic methodology.
Given these difficulties, you may well think that precise translation is impossible. And you would be quite correct, but perhaps something much finer can be created: a brilliant approximation and a different perspective on the same work. A scientific paper on dementia can be translated without any loss or gain in the process, but a novel about dementia and how it affects people in a given society and language community cannot, because literature takes, or should take, language towards its limits, possibly subverting it as it goes. This process magnifies the mismatches between languages that I have just described. A copy produced through translation can be exact in a way, but its texture will be different.
A couple of years ago I translated a novel set in Gorbachev’s Russia by a prominent Italian author, Alessandro Barbero. In a way, this work was already a kind of translation: the author, a medieval historian, writes novels by immersing himself in the language and ideas of a particular society. He had never visited the Soviet Union or Russia (nor has he since), but he was able to understand that society by reading its newspapers, magazines and reports on archival material. Central to the novel, which I entitled The Anonymous Novel, was the narrative voice that cleverly reflected popular conceptions and prejudices in the Soviet Union shortly before its fall. As Italian is a more ‘formal’ language than English, in the sense that its uses more subordinate clauses, apposition, clauses in apposition and flexible word order, the same syntax that could appear to be that of a popular voice in Italian, became rather formal in English, undermining the author’s intentions. I resolved the problem by pushing the register of the narrative voice downwards, and effectively rewriting it. I sent a sample to the author, who has good English, and fortunately he immediately understood what I was trying to do and approved it.
I should perhaps be honest and also admit that this was not always the case, and during the translation of that novel of truly Russian dimensions (187,000 words to be exact), we had quite a few stand-up arguments, often over a single word: a phenomenon only understandable to other pedants like the pair of us – monomaniac writer and monomaniac translator. As an interesting aside on the sometimes fraught relationship between writer and translator, I should add that translators sometimes aver that the best author to translate is a dead one, followed by one who knows no English, and if it has to be a writer who knows English, then one who, like Barbero, knows English well.
Actually, the best writer is this last type, as long as he or she also knows how to delegate the task of translation. Moreover, the odd argument is not so bad a thing: translation can and often should be a negotiation; it only becomes abusive when translators are obliged to accept wordings alien to the languages they are translating into. Enough of this digression: the significance of my example is that by distancing myself from a straight translation using the original syntax I was being more faithful to Barbero’s original intentions. By rejecting word-for-word translation we can sometimes produce a ‘better’ translation. In translation, as in so many other fields, particularly in literature, it is not a matter of generalised rules or even guidelines; it is a matter of judgement on a case-by-case basis. And we could argue endlessly over each of those judgements.
The question of translators being forced to make changes alien to the language they’re writing in (under pressures from either writers or editors – who usually have been approached by the irate writer) is one that leads us to the core concept of this essay which, I freely admit, does contain an irresolvable tension: a good translation must follow the syntax and style of the target language but remain as faithful as possible to the culture of the source language. Amongst the best examples of this are two novels by the great Sicilian verist, Giovanni Verga, who also effectively translated one culture into another. I Malavoglia, which was translated into English as The House by the Medlar Tree, and Mastro Don Gesualdo are both set in nineteenth-century Sicily and its protagonists would therefore have been speaking Sicilian, a so-called dialect but really a different Neo-Latin language closely related to Italian. Very occasionally the author feels he has to use a Sicilian word, and declares his hand by putting it in italics, but principally he creates an impression of Sicilian language and culture entirely through the standard Italian of the time, albeit in a strange tone that underlies the foreignness of Sicilians seen through the eyes of mainland and particularly northern nineteenth-century Italians. Both books make ample use of free indirect discourse, and in the case of I Malavoglia, it is a kind of collective free indirect discourse – a choral voice that represents the fishing community in which the family of protagonists live, with all its beliefs and prejudices.
For this reason, I believe that a translation of I Malavoglia would be an extremely difficult task, more difficult even than Mas-tro Don Gesualdo, Verga’s masterpiece in my view, which, as it happens, was also translated by D. H. Lawrence. It would be a bold translator that would resort to Verga’s techniques to reinvent a culture and present it to another language community in a manner that allows them to understand that culture, perhaps for the first time. As the primary task of a good translation is to allow its readers to understand a culture free from stereotype, such boldness should on occasions be encouraged. However, this is not a simple process and the dangers are several, not the least of them being the opposite result – exoticism and condescension to that culture, although Verga avoided that pitfall admirably. Boldness requires very fine judgement, but even when it ends up misrepresenting the original, the literary result can be startlingly successful.
One such case is Edward FitzGerald’s very loose translation of The Rubaiyat of Omar Khayyam, whose distorting effect might, perhaps unfairly, be accused of Orientalism. I have a great fondness for the work, probably because my father quoted it incessantly, particularly the quatrains that justified his drinking habit. We should not be too precious, and FitzGerald was honest about his method. On the other hand, it would be quite reasonable to claim that his Rubaiyat is not a translation, but rather an original work heavily influenced by another. A translator, in my opinion, cannot go quite that far.
It could also be argued that poetry cannot be translated, although it would be more correct to say that some poems cannot be translated. Shakespeare translates wonderfully into Italian, unsurprisingly because the influence of Italian Renaissance literature and theatre is clearly there, although he develops it in an original manner, but Dante does not translate well into English, in spite of some valiant efforts. TS Eliot argued that Dante was at least as good as Shakespeare and you had to learn Italian to read him. Interestingly, Eliot wrote a superb episode for The Waste Land that was based very closely on Dante’s description of Ulysses’ demise close to Mount Purgatory (Canto XXVI of Inferno). Unfortunately the passage was removed on Ezra Pound’s advice. One thing is clear: the boundary between translation and creative reinvention is particularly blurred when it comes to poetry.
I hope that it is now becoming clear that translation is an integral part of a literary culture. In 2009, Nicholson Baker developed his theory of English poetics in The Anthologist. Whether or not you feel that this book works as a novel, he does have some interesting things to say about our poetic tradition. His principal theory is that a language has a natural rhythm, and the ‘four-beat line is the soul of English poetry.’ Throughout its history, foreign influences have taken English poetry in different directions, but always it has returned to that solid and inescapable four-beat line. For Baker this rhythm is not only natural but preferred, and here I differ. These foreign ways that break with the four-beat line are not an imposition but useful poles of attraction that have improved our poetry – and more arguably have provided us with our greatest poets. The influences, be they through translation or a writer’s knowledge of foreign literature, are essential to our literature, as they are to all literatures, which develop through a continual process of cross-fertilisation.
What are the ways that translation facilitates the growth of a literary culture? Firstly, it provides the best possible apprenticeship for prose writers. A translator of literary novels can shift 250,000 words in a year or more; other professions such as journalism might rival that figure, but it works in a fundamentally different way. Literary translation is not against the clock and writer-translators should take their time to twist those sentences around. In so doing they learn the mechanics of the sentence in their native tongue, a skill that serves them when they settle down to write their own work. Secondly, as I have already suggested with poetry, it introduces new forms to the translator and the reader of the translation.
Of course, the influence is more direct in the case of the translator. George Eliot, a translator from German, was one of the few Anglophone writers in the nineteenth century to make extensive use of erlebte Rede, which I have already mentioned. This function has a ripple effect, and some foreign ideas will be more successful than others, perhaps because of pure chance or perhaps because some ideas will discover a particular affinity within the new culture. Thirdly, as suggested by the FitzGerald case, translation can produce benign misinterpretations. After the Second World War, most Italian writers were translators and many translated from English. These people had learned English in difficult conditions under a Fascist regime; very few had been able to travel abroad. The United States was a mythical place, as much a product of their own imaginations as of their readings in and translations from a foreign language. The great thing about cross-fertilisation is that it increases the variety of literatures, while the current situation in which Anglophone writing influences other cultures in all directions but receives little feedback is leading to increasing uniformity on a global scale, ultimately something damaging for our own Anglophone culture itself.
I used to think that British and Ameri-can publishers were the villains of the piece – a problem of supply rather than demand. However, there have been some notable exceptions, which in recent times have included the heroic efforts of And Other Stories and Peirene, two English publishers who specialise in translated contemporary fiction. My own more eclectic foray into publishing in Scotland (through the imprint Vagabond Voices) has produced four works in translation, and I am now very conscious of the difficulties publishers encounter in this field. It is, at least in part, a problem of demand. In Italy, France and Germany, translation is viable in spite of the added costs, because translated works sell in vast quantities – not just translations from English, although they are dominant, particularly at the popular end of the market, but from a wide range of countries including some of Europe’s smaller nations. Exactly why translation habits vary so greatly (20-25% of titles in Italy and France to just above and just below 3% in Britain and the United States respectively) is not a question I can answer with any great precision. It must be in part about a very long season of cultural dominance, first by Britain and then by the United States. In the sixteenth century, there were two publishers in London specialised in Italian books only, in the eighteenth century educated Britons read French; patterns of dominance change, but America’s global reach is unprecedented. There is also something of the original insularity of English-language culture which survives despite its now planetary spread, but Scotland, as the smaller part of this island, has always been more European, something that shows in its architecture and literature. If it is going to make a meaningful contribution to European culture as an independent nation, it must regain that cosmopolitan edge.