We’ve discussed translation many times, but it’s a subject I never tire of, and I liked very much Val Vinokur’s take on “translating marked voices,” quoted at this XIX век post:
So here’s one of my insights from having translated Chamoiseau and from having translated skaz in general (which is to say, kind of orally inflected speech). A lot of translators, when they’re confronted with this, go for a super-folksy kind of, you know, Cockney or whatever… you know, like film noir kind of thing. The interesting thing about this kind of orally inflected speech, especially when it’s written but even when it’s oral… The person who gets this right more often is David Milch. The way that he writes dialogue, which sounds totally fanciful, but what he realizes is that people who are effectively self-educated, they will overreach, they will actually make their syntax more complicated, right, or more Biblical, because often the thing that they know is the Bible. So it will often sound more archaic, or some combination of the folksy and the archaic, because they’re reaching for something. So it doesn’t sound, you know, quite like the cliché of something low. So I try to keep that awkwardness, that it sounds both idiomatic and idiosyncratic at the same time, really. And also because, in particular, that’s how Babel’s skaz works, right? His skaz is very much Babel’s skaz. You know, his Cossacks talk like, you know, if Babel were a Cossack. They sound beautiful. And weird. And I feel like this is a mistake that a lot of translators make: they say, “well, I know how regular people sound like, so I’m going to… this is what it means to take license and make this sound folksy.” But there’s no one-to-one equivalence with that, so you have to kind of invent this idiom that shouldn’t sound exactly like something you’ve already heard before.
The way that this came up when we were translating Chamoiseau is, what do we do with the Creole, what do we do with the Creolized French? It’s like, “oh, let’s make them sound like Jamaicans!” And it’s like, “no!” Right? So we drew on, sort of, all of the dialects that we knew—including Yiddish!—to kind of come up with this idea of, what would Creolized French sound like in English, and it has to sound like something you’ve never heard before, because you’ve never heard it before.
That’s from a reading and discussion he gave at NYU in 2018; you can watch the whole thing at the link, and there’s another good quote there from Boris Dralyuk, taking a different point of view. Skaz is defined at Wikipedia and in the second paragraph of this 2009 post; it’s one of my favorite things about Russian fiction.
I don’t know how Chamoiseau deploys creolized French in his books but creolized English doesn’t actually strike me as an Obviously Bad way to render it in translation if there’s a translator who can do that competently. Maybe not anything too specifically-Jamaican (vs. the way people talk more generically throughout the Formerly British West Indies in Antigua or St. Kitts or what have you) and certainly not anything specifically Rastafarian, but still. Why should it evoke Yiddish? Why make people sound Generically Exotic when they are in fact exotic, if at all, in particularistic and geographically-rooted ways?
I don’t actually think it’s possible to discuss the pluses and minuses of his approach without having an example of it to hand, preferably an extended one.
I mostly agree with JWB on this, though I think I’d find even Generic West Indian English patois for Creole French would give me the same kind of wrong-kind-of-exotic feeling as Yiddishisms would, though not as acutely, to be sure.
Admittedly I don’t know what the poor translator is to do here. IIRC, Dorothy Sayers in her Dante version has Arnaut Daniel speaking Scots. You go “Yeah, I see what you’re trying to do there, Dot, but really?”
On Babel’s Cossacks: I think this is a subset of a broader artistic problem. Real Life is very often (thank God) quite boring; if you are going to write an interesting more-or-less realist novel, you generally cheat by selecting only the unusually interesting bits, but if you’re really good, like Babel, you can (also) smuggle in interest into the dull bits, but in such a way that your reader is lulled into not noticing that you’ve seriously cranked down the realism.
Dorothy Sayers in her Dante version has Arnaut Daniel speaking Scots
Almost nobody does that kind of thing anymore in English translation. And thank God, imo, although the half-hearted folksy approach with odd gimme and gettin’ is still widespread and I find it almost as cringey. I think the two examples of different strategies that I’ve been happiest with as a reader are Oonagh Stransky’s recent version of The House on Via Gemito by Domenico Starnone and Michael Hofmann’s of Berlin Alexanderplatz. She leaves in a lot of the dialect whenever it’s possible to figure it out from context, whereas he… I don’t know exactly what’s in the stew, to be honest. But it’s terrific.
I don’t know exactly what’s in the stew, to be honest. But it’s terrific.
This is the reaction I would be hoping for if I were a translator.
Real Life is very often (thank God) quite boring; if you are going to write an interesting more-or-less realist novel, you generally cheat by selecting only the unusually interesting bits, but if you’re really good, like Babel, you can (also) smuggle in interest into the dull bits, but in such a way that your reader is lulled into not noticing that you’ve seriously cranked down the realism.
If Real Life were exciting, art would be boring. I think that’s what is meant by “I don’t know art, but I know what I like”. All the world’s a pond, where all the people go to float their boats.
Realism is an art form. It’s not real, except as an art form.
Like translators of ancient Greek plays sometimes tried to make the Spartans sound Scottish to represent their Dorian accent.
It’s a personal view, your mileage may vary, but I always thought that was silly. Too much cognitive discomfort. Spartans were not Scottish.
Better to just let that detail disappear, I thought. Remember that important principle of writing: your readers will never know what you’ve left out.
The theme here is, maybe sometimes you need to accept that a translation can’t do everything. For example, I think this is a warning if you’re thinking of trying a rhyming translation of a long rhyming original (like Sayers’ Dante).
Better to just let that detail disappear, I thought. Remember that important principle of writing: your readers will never know what you’ve left out.
I agree.
Did any of those translators of Lysistrata translate the Spartans’ dialect into “Doric”?
The translation we read in high school, by the American Dudley Fitts, aimed at a Southern (American) accent.
Better to just let that detail disappear, I thought. Remember that important principle of writing: your readers will never know what you’ve left out.
The theme here is, maybe sometimes you need to accept that a translation can’t do everything. For example, I think this is a warning if you’re thinking of trying a rhyming translation of a long rhyming original (like Sayers’ Dante).
In the case of dialect comedy, I’m not sure at all that I agree with letting it disappear.
I agree that a translation of literature can’t do everything (maybe with rare exceptions), but even for a long original, I don’t think rhyme is necessarily the thing to dispense with. For some readers, I think, a rhymed poem with some minor discrepancies in imagery (for instance) is closer to “dynamic equivalence” than an unrhymed poem that gets as close as possible to the original imagery. (Of course not all readers will agree. One participant in the workshop I’m in found the “formality” of one of my translations distracting.)
And readers may know what you left out, if they can see the original and pronounce it, or if they know it by reputation as with Dante, or if you tell them in your introduction, as translators often do.
I feel like we talked about something like this before – something to do with Italian dialect.
My feeling is that it can work for something that’s not very fixed geographically. I’ve got a Norwegian Harry Potter book somewhere and Hagrid definitely speaks quite differently to most of the characters, which didn’t bother me – Hogwarts is nominally Scotland, but it’s really just Somewhere for the stories to be set.
It doesn’t work if you’re translating a story that’s set very specifically in e.g. a particular Italian town (where people are not Scottish, or whatever accent you’re trying to use instead).
I don’t really know where Dante falls on that spectrum, though.
I feel like we talked about something like this before – something to do with Italian dialect.
Yup, back in 2014.
And I stand by what I said here.
I’m not sure how I feel about the Ancient Greek plays, either – I wasn’t immediately against it, but possibly staging a play that way in Scotland with naturally Scots-speaking or Scots-accented actors would come over differently than if you tried to do the same thing somewhere else.
“For example, I think this is a warning if you’re thinking of trying a rhyming translation of a long rhyming original (like Sayers’ Dante).”
To clarify, I thought the style of Sayers’ Dante was tediously mock-archaic in a way that’s quite untrue to the original.
Maybe that would have been her choice anyway. But trying to reproduce the rhyme scheme is relevant because you’ll be naturally tempted to use too much rare or archaic language as you scrape the bottom of the barrel in the desperate struggle to find rhymes.
I would summarise Sayers’ Dante as a courageous attempt to do the impossible.
In the rhyming translations I’ve seen (of songs, mostly – no idea about epics), the opposite happens: the “translation” strays very far from the original in order to rhyme.
…in content, I mean: translators can end up inventing almost entire stanzas.
Why should it evoke Yiddish?
i didn’t take vinokur to be saying that the result would specifically evoke yiddish (or yinglish / yiddish-inflected englishes), but making a fairly conventional nod to yiddish as a (supposedly Very Different) language that’s often talked about alongside caribbean creoles as a fellow “fusion language” or/and “minor literature”.
i can easily understand the temptation to translate from a french-creole to an english-creole, but unless the translator is actually a fluent everyday speaker of the relevant lects – as opposed to a speaker of standardized english and french who can follow along – i think the result is almost inevitably a fairly appalling kind of “dialect” caricature. a brathwaite translation of chamoiseau, though? that would be a dream!
Are there examples of someone translating a different speech variety by inventing a wholly imaginary variety of the target language? Not having an association in the reader’s mind would be an obvious advantage, and an obvious disadvantage.
For some readers, I think, a rhymed poem with some minor discrepancies in imagery (for instance) is closer to “dynamic equivalence” than an unrhymed poem that gets as close as possible to the original imagery.
I absolutely agree (says someone in the middle of translating a 434-line poem in AABBABA that’s driving me out of my mind). I see no reason to privilege content over form as a matter of course when you’re dealing with poetry. Sometimes the form is the content, it depends on the line. I mean, poets make choices all the time that are based on the desire for a particular sound, and that doesn’t necessarily mean they’re settling for mediocre imagery—even though you can often see, if you look closely, which choice came first, and in my opinion that’s the choice that should be also be privileged by the translator. Which you certainly can’t do if you’re throwing out rhyme altogether. Of course in something long and from a language with many more rhyming options than English, you’re going to have to settle for a lot of slant rhymes and a lot of simple assonances, but if you can come up with something good for the really key words, you’re still saving an important aspect of the poem. And if it comes out too forced or too semantically distant, sometimes you have to go back and try just a little bit harder, as Janis Joplin said.
Are there examples of someone translating a different speech variety by inventing a wholly imaginary variety of the target language?
Lewis Grassic Gibbon’s famous introduction to Sunset Song seems tangentially related to this, although it’s obviously not translation in the sense of starting with someone else’s work.
—
If the great Dutch language disappeared from literary usage and a Dutchman wrote in German a story of the Lekside peasants, one may hazard he would ask and receive a certain latitude and forbearance in his usage of German. He might import into his pages some score or so untranslatable words and idioms–untranslatable except in their context and setting; he might mould in some fashion his German to the rhythms and cadence of the kindred speech that his peasants speak. Beyond that, in fairness to his hosts, he hardly could go–to seek effect by a spray of apostrophes would be both impertinence and mistranslation.
The courtesy that the hypothetical Dutchman might receive from German a Scot may invoke from the great English tongue.
…which English lets you get away with. German, for example, does not.
Like Eminem rhyming “orange juice” with “four-inch screws”. It’s all in the enunciation.
and the rhoticity! a friend of mine from melbourne, who’s non-rhotic in both yiddish and english, told me recently about being startled to be told by a colleague from the u.s. that one of her rhymes (i think in an english translation from yiddish, though it could’ve been the other way around) was regionally-specific.
I was reminded yesterday, reading A Thurber Carnival, that Whittier, in “Barbara Frietchie” (does anyone read that anymore? Does anyone read Whittier anymore?) rhymes “staff” with “scarf,” which puzzled me as a youngster. But of course, in his day any pronunciation common to London and Boston was Simply English.
Like Eminem rhyming “orange juice” with “four-inch screws”. It’s all in the enunciation.
Yeah, and that’s actually a rhyme that works pretty well on paper compared to a lot of rap. Many hip-hop rhymes look terrible on the page but sound great (if the artist is good, of course) because the words have been so skillfully bent into place. Alas, though, you can’t assume the average reader is going to help your poem out by doing that in their head.
and the rhoticity!
Is that such a big deal, though? I think readers tend to be at least vaguely aware of such basic differences in accent. I get more worried about stress variation creating metrical atrocities, because that’s not as much on anyone’s radar unless they’ve spent a lot of time around other varieties of English.
@dm
Nachtgeräusche. (C.F. Meyer)
Melde mir die Nachtgeräusche, Muse,
Die ans Ohr des Schlummerlosen fluthen!
Erst das traute Wachtgebell der Hunde,
Dann der abgezählte Schlag der Stunde,
Dann ein Fischer–Zwiegespräch am Ufer,
Dann? Nichts weiter als der ungewisse
Geisterlaut der ungebrochnen Stille,
Wie das Athmen eines jungen Busens,
Wie das Murmeln eines tiefen Brunnens,
Wie das Schlagen eines dumpfen Ruders,
Dann des Schlummers leise leise Tritte. (1)
(1)In der Ausgabe von 1892: „Dann der ungehörte Tritt des Schlummers.“
https://de.m.wikisource.org/wiki/Nachtger%C3%A4usche
Well, if awe and ore are homophones for some ([oː]) but [ɑː] vs. [ɻʷː] for others…
(This does happen in German to a lesser extent, with things like final fortition or the very few stress variations like nachher.)
Oh, interesting. But maybe that’s exactly why I’ve never heard of C. F. Meyer.
But maybe that’s exactly why I’ve never heard of C. F. Meyer.
I guess it’s more a generational thing. His poems were still part of the anthologies and the textbook canon when my parents were at school, which I know because I was the kind of weird kid who read their parents’ textbooks for fun, but already in my time he suffered from the culling of classical 19th century poets to make place for more modern stuff and from the general decrease of reading and learning poetry by heart at school in the 70s.
So… slant rhymes & assonance in German: frantically avoided by Goethe & Schiller; amply experimented with in the 19th century; accidentally dropped in the memory hole in the mid-late 20th? I can see that.
One of the most famous Swiss writers of the 19th century (both poetry and prose)?
Erst das traute Wachtgebell der Hunde,
Shouldn’t that rather be das ferne Wachtgebell der Hunde?
already in my time he suffered from the culling of classical 19th century poets
C.F. Meyer has 20 poems in my 2023 edition of Der Ewige Brunnen, a standard German anthology of 12 centuries of German poetry. Ingeborg Bachmann only has 10, Rilke 14, and even Schiller just 15 poems, so someone must be still be reading Meyer.
@Trond
I see what you are doing with night noises.
@dm
I told you once you did not miss much by not having read “Wilhelm Meister”. I am not sure I would say the same about Conrad Ferdinand Meyer’s poetry. Maybe try “Stapfen” or “Römischer Brunnen” (I think there are several versions of both; he liked to rewrite his poems).
Technically I didn’t know there were Swiss writers in the 19th century (poetry or prose). Now, that’s not terribly representative of anything – but chances are good that I only know the most widely known writers. (And Adalbert Stifter, because he’s impossible to escape in Upper Austria.)
Technically I didn’t know there were Swiss writers in the 19th century (poetry or prose).
You imagine that the Swiss were all involved in making cheese, so had no time to write ? Technically I’m sure you mean “I don’t know any Swiss writers of the 19C”.
Not even Gottfried Keller. Or (with one foot in the 19C) Robert Walser.
Oh well. I don’t know anything about Swiss dinosaurs.
@Vanya: Good to know that he’s still anthologized! It was just that I didn’t come across him in our school readers, and only knew him from my extracurricular reading of the old anthologies in my parents’ and grandparents’ book collections, so I thought he was more or less dropped from the school poetry canon everywhere (perhaps excluding Switzerland).
Everyone knows *one* Swiss writer of the 19th century, don’t they? (Johanna Spyri!)
@Jen: I think lots of people are familiar with Heidi without knowing the name of its author
a shout out
https://www.tumblr.com/udderliteracy-blog/28728120306/the-final-paragraph-of-the-walk-by-robert-walser
for Robert Walser…
To clarify, I thought the style of Sayers’ Dante was tediously mock-archaic in a way that’s quite untrue to the original.
Maybe that would have been her choice anyway. But trying to reproduce the rhyme scheme is relevant because you’ll be naturally tempted to use too much rare or archaic language as you scrape the bottom of the barrel in the desperate struggle to find rhymes.
I would summarise Sayers’ Dante as a courageous attempt to do the impossible.
I took a look at her introduction to the Divine Comedy here, where she discusses features of her translation on pp. 55–65. She agrees that it’s impossible to convey the poetry into English, especially where Dante is “sheerly beautiful”. However, she seems satisfied with her decisions to use terza rima and “thee and thou”, and to try to “steer a discreet middle course between Wardour Street and Hollywood”. Readers may disagree, and I’m certainly not going to defend her decision to use Scots for Provencal.
@Biscia: I’ve been going even further in translating Antonio Machado and almost always used exact rhyme where he does, since I feel that for many readers it has a very different effect from approximate rhyme. Also, for Machado, that preserves the distinction between his poems or parts of poems with assonance and those with exact rhymes (though some might argue that was just a matter of convenience for him). But the longest piece I’ve attempted with exact rhymes is a mere nine abab quatrains, nowhere near as difficult as what you’re doing or as terza rima.
And yes, “try just a little bit harder” is very helpful. To some extent, this is just a matter of what one has patience for. But I’m not doing this for a living or on deadlines. If you are, I’m very impressed.
I confess I’m quite fond of Laurence Binyon’s Dante, even though he violates various of my principles of good translating (forces exact rhymes with archaic diction, that sort of thing).
And looking up the name Binyon, I discover it’s Welsh: “ap Einion.”
Everyone knows *one* Swiss writer of the 19th century, don’t they? (Johanna Spyri!)
Just as every American child knows (knew?) one French novelist: Jules Verne. (Or “Joolz Vurn.”)
the rhoticity! / Is that such a big deal, though?
it depends on how rigorous you’re being with rhyme, and who your audience is. i think my melbourne friend would be entirely justified in (for example only; i don’t know that she’s done this) translating things that only rhyme in her own non-rhotic australian yiddish into things that only rhyme in non-rhotic australian english, whether or not either would rhyme for rhotic me. but if she’s aiming to convey precise rhyming to a mixed-rhoticity audience in either language, i think she’d do better to choose words that work either way. and closer to (my) home, katz’s deli’s famous “send a salami to your boy in the army” sign gets a lot of its punch and appeal from the local rootedness that its non-rhoticity (and a bit of specific vowel quality) conveys.
the problem comes up more with vowels in yiddish, since there’s (notoriously) so much variation between lects (/u/ vs /i/, /a/ vs /o/, /ɛ/ vs /ej/, /ej/ vs /aj/, etc). so it’s quite common for a song or poem’s rhymes to only work in the writer’s own lect, which drops a problem in any translator’s lap. when i translate, i usually don’t try to reproduce rhyme and meter – as much because i don’t have the patience for it as anything else. so i generally pick works to translate that don’t need that, like roza gutman’s free verse, with her luscious alliterations and assonances, as opposed to fradl shtok’s equally lush but tightly structured sonnets and ballads.
@hat
Dante, Paradiso xvi 121-126:
Già era ‘l Caponsacco nel mercato
disceso giù da Fiesole, e già era
buon cittadino Giuda e Infangato.
Io dirò cosa incredibile e vera:
nel picciol cerchio s’entrava per porta
che si nomava da quei de la Pera.
Binyon:
Already to the market-place below
Had Caponsacco come down from the hill:
Giuda, Infangato, yet had worth enow.
I’ll tell a thing true, though incredible:
One entered the small circuit by a gate
That took its name from the Pear’s blazon still.
Prose summary (without filler)
Caponsacco had already descended to the market,
where Giuda and Infangato were, from Fiesole.
i will tell you something incredible and true:
one entered the small circle by a gate
that was called after those of the Pear.
Problems
1. Enow is COW; below is THROW.
2. Incredible and still is unfortunate.
3. Placenames matter. Bring back Fiesole!
4. Dante does alliteration–per porta Pera
5. The words used are simple and repetitive (see prose above)
6. Caponsacco (“Caponbag”) and Infangato (Stick-in-the-Mud”) are real Florentine patrician families.
PP (suggested):
Caponsacco was already at the square,
Descended from Fiesole another day.
Good Giuda and Infangato too were there.
You won’t believe what I am going to say
To reach this tiny circle a great gate,
Named for the noble Peras was the way.
I know, I know, he’s indefensible, and yet I like him.
Yes, so technically I didn’t have any evidence any existed, aside from sheer probability of course.
However:
Argh, yes. And I even know Gritli and Heidis Kinder (not that I’ve read them).
@Hat: my principles of good translating (forces exact rhymes with archaic diction, that sort of thing).
But what if the original uses words or grammatical structures that were already archaic or poetic? (I’m not saying that’s true of Dante, since I don’t know italian of any century.)
@PlasticPaddy: Good stuff! Now do the preceding rhyme with “square” and the following rhymes with “great”, and then… and pretty soon you’ll be done with the whole canto.
But what if the original uses words or grammatical structures that were already archaic or poetic?
I don’t care so much about that as about how well it works in English — what the result is like.
@Jerry Friedman
Thanks. More from Sayers’ introduction to Dante:
“When Dante chooses to be sheerly beautiful, he writes not like a man but like an angel … there remains Dante the poet, who walks equal with Homer and Aeschylus and Virgil and Shakespeare, and whose shoe’s latchet none but the very greatest is worthy to unloose. ”
I’m not a fan of that sort of deification. There is no perfection in this vale of tears.** Even Homer nodded.
I’ve just reread the first few cantos of Inferno looking for places where the poet was clearly stuck for a rhyme or a filler word and eventually thought “life is short, I need to move on, second best will have to do.”
1.46-48: superfluous “parea”, twice in quick succession (the scene is otherwise presented as reporting real experience, not as something that ‘seemed’ to be something as though it was a dream)
2.141: awkward pluperfect tense in a context that’s mostly in the simple past – compare 1.136
3.14-15: “convien” is a rather weak word for a meaning that translates better as “must” (Binyon, Sayers, Kirkpatrick) (?? assuming that its meaning is similar to the modern one: ‘be convenient, worthwhile, useful” ??)
3.66: low value filler words: “ch’eran ivi”
4.7: low-value filler words: “vero e che”
4.79: inappropriate passive verb: “Intanto voce fu per me udita”
Of course one such example in every hundred lines is still an amazingly good batting average. But I find it comforting to think that Dante sometimes sweated over a line as you or I might.
Incidentally, it had never occurred to me that a final ‘r’ in non-rhotic speech is like ‘certain male spiders’ (p.57 footnote in the link you gave to the introduction to Sayers’ Inferno)
** To be fair, she wasn’t claiming he was *always* perfect
assuming that its meaning is similar to the modern one: ‘be convenient, worthwhile, useful”
That’s a very dangerous assumption about a seven-century-old variety of the language. In general, seeking out possible errors/weaknesses in a great work of literature strikes me as a mug’s game, though of course if it makes you feel better, more power to you; obviously, as you say, there is no perfection in human things, and I don’t think people using that kind of rhetoric actually believe that every word and phrase is perfect (they’re giving way to the same impulse as a fond parent saying their baby is perfect), but it seems to me the main problem with discourse today is not inappropriate positivity but people approaching everything from a prosecutor’s viewpoint, looking for some error or incorrect perspective so they can dismiss it (“I stopped reading when I got to…”). Not saying that’s what you’re doing, just explaining where I’m coming from.
@Hat: Thanks, that makes me feel better about my resistance to translating with the kind of structures that some poets used a century or more ago, though I know some readers will have the opposite preference.
Yeah, the longer I live the more strongly I feel that there are no absolute answers in these realms, just things that work for you or not. Some things work for me but not for others, some things the reverse, it’s all good. Variety is good.
Homeric epithets were often just used to fill out the meter. How often exactly is the tricky question for a translator.
@Julian: Thanks, I’d skipped that footnote on spiders, which almost makes up for Sayers’s failure to mention varieties of English from outside Britain and Ireland.
@Hat: Next time I submit something that requires a translator’s statement for publication, I may say that I’m trying to provide the appropriate strokes for certain folks. Or not.
I’ve always admired Tolkien’s use of a variety of English accents and registers to represent various cultures of Middle-earth, although this was not translation except in the internal history of the setting.
His under-rated parodists Henry Beard and Douglas Kenney (Bored of the Rings) extended this to varieties of English outside England.