Sign up to the newsletter

Translating History

The Watchmaker of Filigree Street

History is a foreign country. Digressing a bit, so is Japan. Haruki Murakami is one of the best-selling writers worldwide — translated. Nobody expects readers who love IQ84 to go off and learn Japanese to read it in the original. On the other hand, lots of people, readers and writers alike, do expect to see historical fiction in the diction of the period. Even though this sometimes doesn’t work much better than offering up Japanese, it isn’t always treated as something that could be translated.

That said, everyone knows it sometimes does. Take the Anglo Saxons. You can’t write raw English from before the Norman Conquest and expect more than ten or twelve philology professors to understand, and maybe some very determined students with dictionaries. It was a different language. Any story written about that period now must come in modern English. Where a real Anglo Saxon would have said ‘hlaford’, a modern writer must say he said ‘lord’.

The trouble is that this doesn’t feel like translation at all, and it isn’t really. If a writer sets out on an original story about Alfred the Great, there’s no existing source text to ‘translate’, only a much more nebulous awareness that if this really were to have happened, it wouldn’t have happened in recognisable English. Because it is only imaginary translation, all the usual questions we ask about ordinary translation hide cunningly in hedges and camouflage themselves as things that look unimportant. They ought to be hauled out, though, because they’re terrifically easy to trip over otherwise. The worst is to do with making Anglo Saxons sound like Anglo Saxons.

Most readers and writers today have an instinctive idea of what Anglo Saxons and Vikings ought to sound like. It’s often quite strict. The people were Germanic and bleak, so they’re matched with Germanic and bleak words; there will be mapmakers rather than cartographers, whale-roads rather than oceans, thanes rather than gentlemen. Someone even told me recently he had been ripped to bits in a creative writing workshop for writing ‘won’t’ rather than ‘will not’ in the speech of a Saxon character. Not everyone has seen real Old English, so it seems odd that we all know what to expect, but you don’t need to have spent hours banging your head against Aelfric’s account of the martyring of St Edmund to have been exposed to this stuff. Anyone who’s watched Lord of the Rings has heard Old English poetry — and since the sword-and-sorcery era in other fantasy novels usually equates roughly to Anglo Saxon England, the tone of those affects expectations of the period too. Every few years or so somebody makes a film of Beowulf. Ultimately they’re all based on scholarly translations of the original texts, of which there are plenty. Tolkien was only a novelist in his free time: he taught Old English at Oxford. But academic translations from historical languages — especially obscure ones — are atrocious things from which to extract any expectations about what the people of period sounded like.

The reason is to do with who those translations are for. The readership is almost always students and scholars, who read them either to take an examination or to write a research paper. That context is crucial, because in essays and papers about Beowulf, the elegance of the modern translation is irrelevant: what’s important is to know what each Old English word actually means. The ‘best’ translations are not usually the most elegant or the most natural, but the most accurate. For the purpose of scholarship, this accuracy is absolutely necessary. For the purpose of getting a natural idea of tone and register — what Old English speakers sounded like — it’s completely counter-productive. 

A good example of what I mean is this prose translation by E. Talbot Donaldson. It’s a rendering of one of the most famous passages of the poem.

Then from the moor under the mist-hills Grendel came walking, wearing God’s anger. The foul ravager thought to catch some one of mankind there in the high hall. Under the clouds he moved until he could see most clearly the wine-hall, treasure-house of men, shining with gold. That was not the first time that he had sought Hrothgar’s home. Never before or since in his life-days did he find harder luck, hardier hall-thanes. 

It sounds very Anglo Saxon, very authentic, and very weird. It’s full of strange double-barrelled words — ‘mist-hills’, ‘wine-hall’, ‘treasure-house’, ‘life-days’ ‘hall-thanes’; none of them are natural English now, but what they reflect are the kennings — poetic compound words — from the Old English. The grammar and syntax is noticeably odd. It hasn’t been naturalised but left almost literal.

That would be very strange indeed if this were a modern foreign language translation. If in a translation class today you translated the first line of IQ84 as ‘The radio of the taxi was flowing an FM classical music programme’ , you would almost certainly get a tap on the shoulder and a suggestion that you might like to have another go. Where Japanese flows or runs music, English plays it. That seems like an obvious change to make in the context of Murakami, but it doesn’t happen so automatically in scholarly translations, and for a good reason. The primary function of a translation of the Donaldson kind, just like any other, is to convey the meaning. But it has a second function that modern foreign language translations don’t: it’s a kind of glossary. It’s meant to be read with one eye on the original. To naturalise anything would be nothing short of obfuscatory.

And anyway, Beowulf is poetry, of an extremely strictly formed kind. It’s lofty. So is a lot of the Old English we have left; the Battle of Maldon, the Wanderer, the Seafarer. The surviving prose is often religious — Bible translations and hagiography feature prominently, not least because a good number of the people who could actually write were ecclesiastical. Quite high-brow stuff, written by poets and holy men, then translated for scholars.

Despite all that, this is still the tone associated with old England, and therefore what readers and writers often expect to see in novels about that period. But the Vikings (Beowulf is about Danes) were hardly poetic. They turned up, stole loads of stuff, and then came back whenever the weather was all right to steal more stuff. The Anglo Saxons were hardly all swooning poet types either. So reading about them in a strange style that comes from old poetry and literal translating is a lot like reading about modern gang-killings as if they were in an epic.

And raising up his gun, Big Doggy said to the maiden, ‘Lo, but if you refuse to grant me the spoils of your hall-raided gold, I shall shoot your face off and bring God-wrath to your kin over the mist-towers of Detroit.’

Completely silly. Equally silly to imagine beserking Vikings in these terms.

That’s a very extreme case of translation and translation questions. Not everybody in history is a Viking though, and not everybody’s real language was so distant from modern English as Old English is. As the Normans Frenchify the language and the date edges toward fourteen hundred, it becomes far more recognisable. It’s easy to think that there’s no reason not to report raw what people would have said verbatim from that time onward, and people who do think so are in good company. Plenty of toweringly good writers have done just that; Peter Ackroyd, William Golding, Rose Tremain, and Eleanor Catton among them.

I think there is quite a substantial reason not to, though. My undergraduate thesis was about Elizabethan child actors, so I do understand early modern prose after a certain amount of mental shoving. A colleague of mine, otherwise an extremely clever man indeed, foolishly wrote his on simulating magnetic monopoles and now struggles with Victorian language. Whoever the reader is, though, their understanding of historical diction is always going to be more imperfect than their understanding of what they use every day. That’s a very obvious thing to say, but I think it’s sometimes much more of a difficulty than it’s made out to be.

A nineteenth century character who calls somebody a ‘damned fool’ sounds appropriately old-fashioned and well-placed in his period. He also sounds posh. The phrase now belongs to a higher register than it did at the time, when it wouldn’t have taken a mansion-dwelling tweed-wearer to say it. One of the problems that pitch-perfect historical diction novels suffer from, then, is that everybody tends to sound rather refined. That can quickly become limiting. What we understand by that sort of diction now isn’t what somebody at the time would have taken away from it. We understand ‘you damned fool’. But we don’t take it in the same visceral way as ‘you fucking moron’. Of course nobody in 1820 would have said that, the one word being largely a medical term and the other not recorded in use as an intensifier until 1864, but that doesn’t mean a comparable force of feeling didn’t exist. ‘Damned’ was a bad word, to the extent that it’s often written d—d, in the same way we sometimes now write f***,  and it’s attributed to rough types as much as anyone else . The register has lifted now. Reporting such speech directly in a historical novel gives a strange and wholly wrong impression of restraint. The character seems as though they have either less emotional range than a modern person, or infinitely better manners, because they seem to be sticking only to one, now rather elevated, register.

That’s unobjectionable if the characters are posh. Actually it’s very good, because it helps give us information about their personalities. If they aren’t, historical diction tends not to suit. Full dialect and slang from more than two hundred years ago tend now to be as impenetrable as what the Saxons spoke. What was standard English has this rose-tinting effect. Instead of trying to wade through shades of historical accuracy, it’s worth trying another approach. Decide what sort of person they are, then report it the way somebody like that would talk now; translate. The best example I know is a short story by Wells Tower, Everything Ravaged, Everything Burned. It’s about rampaging Vikings and not once do any of them say ‘Lo!’

Something is lost if you use modern English in historical fiction, of course it is, but other things are lost if you don’t. If history is a foreign country, historical diction is a foreign language. I don’t think it matters that Haruki Murakami’s word choices must be altered to make sense in English, and by extension I don’t think it matters if a historical character’s grammar becomes modernised for the sake of clarity. The difference between a Victorian and a modern woman in a pocketwatch and corset isn’t that one of them doesn’t say ‘okay’. It’s that they think differently. Even such basic things as the perception of distance and the size of the world are different. If you wanted to get to Japan by sea before the opening of the Suez Canal in 1869, it took about two and a half months, via Hong Kong. In 2006, it took the New Horizons probe two and a half months to pass the orbit of Mars.


Natasha Pulley studied English Literature at Oxford University. After stints working at Waterstones as a bookseller, then at Cambridge University Press as a publishing assistant in the astronomy and maths departments, she did the Creative Writing MA at UEA. She later studied in Tokyo, where she lived on a scholarship from the Daiwa Anglo-Japanese Foundation. She was chosen to be a Writer in Residence at Gladstone's Library and is now associate lecturer at Bath Spa University and panel tutor at the Cambridge University Institute of Continuing Education. Her first novel, The Watchmaker of Filigree Street, was an international bestseller, won a Betty Trask Award and was shortlisted for the Authors' Club Best First Novel Award. The Bedlam Stacks is her second novel. She lives in Bath.