On Self-Translating Icelandic to English.

Alexander Dan Vilhjálmsson discusses how he translated his own novel into English:

I never intended to translate my own book. The way it happened felt almost as if by accident. I self-published a fantasy novel called Hrímland in 2014. I was exhausted after the process, but soon after I decided to try out translating it. Gollancz had opened their submissions for a limited time, and I used that for motivation. It was just supposed to be an experiment to see how on earth I would manage to translate this work. I never imagined it would go anywhere. That book ended up becoming Shadows of the Short Days.

See, I knew this was the kind of book where I couldn’t just hand it off to a translator and get a good result without massive interference from myself. I wanted to find out how I could do it, and perhaps in doing so learn something new about how I wrote in English.

Writing fantasy fiction naturally lends itself to a lot of worldbuilding – and that worldbuilding is done through language. In Icelandic, I had pillaged my country’s archaic vocabulary when coming up with fantastical terms in universe, also looking up old kenningar, poetical words and phrases. I turned these words into species, warrior-castes, sorcerer names. Icelandic also lends itself well to making up new words – a language with a lot of compound nouns is fun that way. So, when I looked how I would tackle the translation, I had to decide what to keep and what to translate into English.

When is something too precious to worldbuilding lost in translation? When is something a bit too untranslatable, too culturally important, or just too damn cool to be turned into English?

A whole lot, according to the glossary at the end of the book. […]

In the Icelandic text, the ravenfolk speak a very strange type of Icelandic (or, well, Hrímlandic). Their corvine vocal cords can mimic human languages effectively, but they care little for the ways of land-bound species. They much prefer to speak in the rough caws of their native skramsl (which is an archaic word for a raven crowing). They speak a faux type of Old Icelandic, like something you see when reading the Sagas of the Icelanders. The Sagas took place in the 9th to 11th centuries during the settlement of Iceland, written a few centuries later. In universe, the náskárar learned the human tongue then, and have since not bothered with updating it much.

So how do you translate someone speaking fake Old Icelandic in English? […] This is where it becomes convenient being the author as well as the translator. I did something a bit strange. I rewrote their dialogue and kept a lot of Icelandic words in there. Like, a lot. They almost speak Icelandic half of the time, although it’s the same kind of strange, faux Old Icelandic. The fusion feels like something from another world, a different time, and the English reader gets a real sense that these bastards really are speaking the old tongue. This was a big part of why I wanted to do the translation myself. I knew I wanted to do something unorthodox with them, as with so many other parts of the world.

Even if it’s just mundane things like not translating landi as moonshine.

I personally think those are pretty dubious decisions that another translator would sensibly have avoided, but hey, it’s his book, and it’s fun to read about. (There are more details at the link, of course.) Thanks, MattF!


  1. “Another translator” (as you put it) is in one way misleading. Yes, he is “translating” the novel into English. But in another sense he, as the author, is rewriting it in English.

    Under traditional theories of translation I think there is likely to be a fundamental distinction here. I suspect, however, that under more modern theories the distinction is moot.

  2. Under traditional theories of translation I think there is likely to be a fundamental distinction here.

    I don’t; plenty of traditional translators have felt free to rewrite the original.

  3. I started reading A Void, Gilbert Adair’s masterful English translation of George Perec’s La Disparition. It’s quite engaging, if violent. The original and the translation don’t use the letter e. Among other things, it means that the numerals one, three, and five cannot be translated directly, nor la, nor , etc.

  4. >They almost speak Icelandic half of the time, although it’s the same kind of strange, faux Old Icelandic.

    An sample sentence would have been helpful here, though I guess the interview format doesn’t allow it. I’m intrigued at how much Icelandic could be left while making it intelligible to an English reader.

  5. PlasticPladdy says:

    Here is a quote from Google Books:

    ‘[TH]érr err-at with honour,’ said Rotsvelgur…’to come hérr, to vor hall, deman’t to pay’t skuld, wit’ forneskju ok ruin the honour of Krxgraak’úrtek’

    [TH] is the thorn and the last word is italicised and looks to me like a “raven call” word and not Icelandic😊

  6. Do you seriously think you are the first young, arrogant galdramaður we’ve had who wants to revolutionize the craft? That other misguided galdramenn before you have not tried to achieve the same lofty results you are hoping for?

    And so on.

  7. PlasticPaddy says:

    @sfr, ryan
    The above quote is not from the “raven speech”. I had a comment with a quote, but it seems to have gone to Spam. Ryan, if you look in the Google books entry for Rotsvelgur, you can find quotes.

  8. I rescued the comment with the quote.

  9. Part of the library charter at Cornell University (which houses the 3rd largest collection of Icelandic/Iceland-related materials) stipulates that there must always be an Icelander graduate student on campus if one applies and that that Icelander must teach Old Norse if there are at least 3 people who are willing to study it.

  10. Thank you for the word “skramsl.”

  11. There’s quite a difference between SFReader’s quote with a single non-English word that is fairly clear even without any other context than the passage given, and PlasticPaddy’s:
    >‘[TH]érr err-at with honour,’ said Rotsvelgur…’to come hérr, to vor hall, deman’t to pay’t skuld, wit’ forneskju ok ruin the honour of Krxgraak’úrtek’

    Is this sentence more intelligible if you’ve read what comes before? Or if you have some other Germanic language than English? When I type foreskju, google helpfully suggests I probably meant foreskin. And believes that Icelandic ok means OK.

    Though I see that og would mean ‘and’. Is ‘t used to indicate a third person pronoun?

    Is skuld cognate with should in the way that ought relates to owe?

  12. PlasticPaddy says:

    I only have seen the Google book version. In this case, the dialogue is in the middle of a scene where a debt has been mentioned, implied to be large, and the person addressed in the quote I have given has proposed to pay the debt by providing some sort of magical spell or artefact (“svartigaldur”).
    Skuld is recognisable from modern German Schuld (Anglo-Saxon scyld) and DWDS confirms that the Germanic form is derived from the corresponding verb “should”. For forneskju see

  13. Thanks, Plastic. Some of that I was able to find. It’s more research than I’d want to do to read a single sentence. I didn’t try too hard. But I still don’t understand therr, err-at, nor quite what deman’t to pay’t means. I was taking the latter to be “demand that he pays his debt,” but I gather from your reply that’s not correct. And I’m just guessing at herr and vor hall.

    It’s interesting to ponder a novel that began in English, and slowly transitioned to more and more Icelandic, introducing terms and constructions slowly, with obvious cognates first, in a way that made some of the sound rules and correspondences obvious, so that more distant cognates would be recognizable by the time they came in. How Icelandic could it be by the end while still carrying most reasonably smart readers along? Perhaps that’s exactly what’s going on here, and the sentence would be completely intelligible if you’d read this far.

  14. Plastic Paddy’s sample indicates that the faux Old Icelandic uses more-or-less standard modern English grammar. Vilhjálmsson tells us there’s a glossary, so I would think a committed reader would be able to read proficiently after a few pages. More difficult than A Clockwork Orange, but easier than Chaucer. The question is, will it attract a readership willing to put in the effort?

  15. Peter Maydell says:

    I see the author also mentions in one of the comments on that piece that the editing process of the English version involved so many changes that he ended up re-translating it back into Icelandic to publish the Icelandic version, which is a nice little loop…

  16. Yes, it reminds me of Nabokov writing Conclusive Evidence in English, reworking it in Russian as Drugie berega [Other shores], then reworking that in English as Speak, Memory.

  17. an-sky’s “tsvishn tsvey veltn” (better known as “der dibek” / “The Dybbuk”), which is having its centenary this week, is another one like that! i last checked the philology some years ago, but at that point the consensus was that there was no sense trying to call either the earliest extant russian or yiddish version “original”, since an-sky rewrote them pretty intricately against each other…

    what’s clear is that the first performance was in yiddish, by the vilner trupe / Vilna Troupe, followed after a few years of dybbuk-mania by vakhtankov’s hebrew-language* production with Habima.

    in any case, there’s a centenary production on monday (af yidish); like many productions regardless of language, it’s a new adaptation – i’m excited to see it!

    * i don’t know whether the translation was into literary ashkenazi hebrew or ivrit/israeli.

  18. Interesting! I saw a performance many years ago and was duly impressed (in English, though, I’m afraid).

Speak Your Mind