Bathrobe is delving into translation studies, and he’s found another good link to send me: The Art and Craft of Translation, by Elborg Forster (from Johns Hopkins Magazine, way back in February 2001). As with the Johanna Hanink piece I posted recently, it’s full of good things, so I’ll just quote a few bits to pique your interest:
First, a few words by way of characterizing myself: although my first language was German, I am now a translator with many years of experience translating texts in such fields as history, anthropology, sociology, and history of science and medicine from French and German into English. Having stopped using German in my daily life 45 years ago, I now find it rather difficult to translate into German. This has to do with lack of practice, of course, but also with the fact that there are many areas of life and letters with which I was not familiar as a 20-year-old. Indeed I sometimes feel–no doubt erroneously–that I learned everything I know in the medium of English: politics, French history, child-rearing, cooking, life and death, gardening, healthcare, automobiles …. In any case, whenever these days I translate something into German, I send it to my verbally highly gifted sister in Germany, who usually finds a few anglicisms and some expressions “that we don’t use any more,” for of course the German language itself has evolved over the last 45 years. […]
My greatest adventure as a translator was a collection of letters of Liselotte von der Pfalz, the 17th-century German princess living at the court of Louis XIV. It did not take me long to realize that her very lively and expressive German had become somewhat archaic after she had spoken mostly French for many years (a situation I intimately understand!). I therefore chose a somewhat old-fashioned American idiom, often using expressions marked ant. or F. in the dictionaries. I also realized that “Madame,” as she was called at the French court, varied the level of her style depending on the subject of the letter, her level of familiarity with the recipient, and especially that person’s position in the hierarchy of court society.
Even in letters to close relatives, Madame used the address “Your Grace” (Ew. Liebden) as a matter of course; it was obligatory for persons of a certain rank and gives modern readers a feel for the distance that separates us from 17th-century court society. It seemed necessary to make a point of this distance, for much of Liselotte’s writing calls for such colloquial translation that we might take her for our contemporary. Yet that would be unfortunate, for it would prevent the reader from realizing that in many respects this woman was way ahead of her time.
Translating Liselotte thus sometimes called for a simple vocabulary (“rumors flying,” “go after,” “do away with,” “a lot of useless information”) and a straightforward sentence structure (essentially run-on sentences) to convey easy familiarity. But at other times I had to search out inflated terms to render the painfully constrained formality of a letter, so that I used such expressions as “favor me with a letter or any word,” “filial trust,” “pay my respects in person,” “bestow,” “hard put to give credence,” “at length…” and a host of others. And I certainly had no right to cut up the endless sentences.
Sometimes authors are not consistent in their level of style, and then the translator gets into trouble with copy editors. In one rather formal German study of Weimar Germany published in the 1970s, the author suddenly–and effectively, I thought–used the word “aufgehübscht,” which I literally translated as “prettified,” but the copy editor felt that this kind of expression “did not belong into serious academic discourse.” Unable to persuade him that it might, I dropped it, to my regret.[…]
To begin with, linguistic communities have different historical memories, which are rendered in a kind of shorthand but must be spelled out in another language. Thus I once gave a word-for-word rendition of “a portrait of the King of Rome,” whereupon my copy editor suggested that for simplicity’s sake I make it “the pope.” Seeing that even this educated person had missed a cultural allusion that would be obvious to every French reader, I therefore wrote: “Napoleon’s son, who bore the title King of Rome.” A translator from American English, of course, would have to add similar glosses to “crossing the Delaware” or “the man from Independence”–and these examples only refer to historical allusions. Every language is full of cultural concepts that require paraphrasing and sometimes a complete transposition.
Thanks, Bathrobe!
Stories about the counter sense that machine translations have produced are legend. My two favorite–and possibly apocryphal– examples are these: When programmers fed the phrase “Out of sight out of mind” into an English-Russian program and then, to check the accuracy of the translation, back into a Russian-English one, what they obtained was “Blind idiot.” And for “The spirit is willing but the flesh is weak,” the same operation yielded, “The vodka is good but the meat is rotten.”
While machine translations have indeed been known for egregious mistakes (much more so in 2001, or even 2011, than now, when the neural-network methods made the obvious howlers much less frequent), as noted already by Language Log in 2006, the mistranslation of those two specific proverbs dates back to at least 1903.
In fact I can push it at least three months earlier, to October 1902:
Among the stories about errors in English the Los Angeles Times tells the following: “Many ludicrous mistakes are made by foreigners in grasping the meaning of some of our common English Expressions. A young German attending the State University translated ‘The spirit is willing, but the flesh is weak,’ into “The ghost is willing, but the meat is not able.” And a Filipino youth fairly sat the class in an uproar by the statement that ‘Out of sight, out of mind’ meant ‘The invisible is insane.'[”]
– The Richmond Dispatch, October 4, 1902 (page 4, near bottom of second column)
Alas, I was unable to find the implied original in the online archives of the LA Times. It would have been interesting to find out how they attributed the story (and whether it was, in fact, surrounded by a bunch of similar stories).
Nice antedate! It amazes me that people continue to pass on these old chestnuts as though they were fresh and meaningful.
Huh, I didn’t even know that word. And I did know the long-forgotten abbreviation Ew. for Euer (with w for cosmetic 16th-century reasons).
And I did know the long-forgotten abbreviation Ew. for Euer (with w for cosmetic 16th-century reasons).
I wondered if that’s what it meant! What were the reasons?
@dm
https://de.m.wikipedia.org/wiki/Kurialien
“Ihre [Der Kurialien] Anwendung war schwierig, weil es für sie nur grobe und keine allgemein verbindlichen Regeln gab, sodass in der Praxis immer die Gefahr bestand, einen protokollarischen Missgriff zu begehen.”
So even those who knew “Ew. Liebden”, like Madame were in danger of misapplying it. But at least none of her correspondents were likely to demand Satisfaktion from Madame…
Elborg Forster is quite a prolific translator: Books by Elborg Forster on Goodreads.
I also suggest ‘tarted up’ as a translation of aufgehübscht.
Random use of ew for eu: if you can’t use more letters, use wider letters, I guess.
Much too pejorative.
The Kurialien article says Euer Liebden was the address used by princes to other princes of the same or lesser rank, and ladies were always addressed above their station.
The reasons for using W for U/V were twofold.
First, at the time, there was no distinction between u and v. Both were regarded as a single letter with typographic variation, with v usually used at the beginning of the word, and u in other positions. Both also stood for consonantal and vowel values. In German, Polish, Latvian and Lithuanian w was increasingly used for the consonant, but also for the doubled U. In Croatian of the time we also find traces of this laissez faire approach to spelling eg. WKOVICH for the surname Vuković.
The second is that in the handwriting of the time, it was difficult to recognise letters composed only of minims. So U and V were sometimes replaced by W to facilitate recognition of the letter. Similarly I and J could be replaced by Y. In English this is preserved in surnames such as Wylde or Smythe.
Standardised spelling was something that was a by-product of 18th century rationalism. In the Habsburg lands, it came in with Maria Theresa’s reforms in the second half of the 18th century. These reforms affected all the Volksschule in the monarchy – German, Hungarian, Czech and Croatian – that is, the reforms brought a standardised approach to spelling in all those languages and government-issued spelling books were introduced for school use.
Finally in abbreviations, archaic spelling conventions tend to be preserved for longer. For example Latin C for Gaius, Spanish Vd for Usted, & for “et”.
Definitely too pejorative. There’s no indication what aufgehübscht was being applied to. If not a person, “spruced up” might work. Otherwise “spruced up” would be appropriate for a man rather than a woman.
in the handwriting of the time, it was difficult to recognise letters composed only of minims
I believe that this is also at the back of the English use of o for u in son, some, love etc.
Definitely too pejorative.
“Sexed up” is right out, then?
I believe that this is also at the back of the English use of o for u
Indeed it is, as well as the dot on the i. Turkish ı was a bad idea typographically: writing mınımum in either italic or rounded handwriting is nothing but a series of zigzags or bumps.
Talking of which, January invoked the spectre of Sütterlin just now in the Linear A thread:
http://languagehat.com/solving-linear-a/#comment-3831969
https://en.wikipedia.org/wiki/S%C3%BCtterlin
Says Wikipedia: “For most people outside Germany (as well as younger Germans), Sütterlin is nearly illegible.”
Nearly?
“Gussied up” seems the closest English equivalent to me, but even that has a bit of snide to it that the German lacks.
According to the fantastic Upward & Dawkinson’s The History ofEnglish Spelling, for the same reasons we now have O instead of U in wolf, and OO in wood and wool.
The v/u thing was present in English up to the mid-17th c. The rule was initially and medially regardless of whether a consonant or vowel is meant. Lass says: “dictionaries however continue for a long time to alphabetise together: Johnson 1755 has an entry only for , and the lemma vizier is followed immediately by ulcer, unzoned is followed by vocabulary.”
I was taught about her reforms early and often, and yet this part never came up once. 😮
Yes. Hübsch = pretty, so “prettified” really is the only good option.
Well, not if you keep your ∩ and your ∪ distinct… which most people don’t.
The University of Vienna did that into the 1970s if I’m not confusing that with alphabetising I & J together, but both were kept for much longer than elsewhere.
I just deciphered the whole sample before noticing it’s transcribed in the footnote, but it took a while. I couldn’t make sense of the first word till I finally got to Bezeichnung and found out that the L with the little extra loop is a B! And that’s with the following background knowledge: e looks like n for (presumably) some reason, only narrower; ſ is used, but the two vertical loops that you’d expect to be ſ are h instead; u is distinguished from n by a breve on top.
Sütterlin — try reading Danish church records from before 1820 or so, then. The first computer-generated sample in Wikipedia is exceptionally confusing because all the strokes are straight and the letters mercilessly connected — actual handwriting can be crabbed and imperfect, but you get different forms of letters depending on context which helps a lot, and also cap height is much more than the x-height + 20% of the sample so you can easily spot the nouns.
In one rather formal German study of Weimar Germany published in the 1970s, the author suddenly–and effectively, I thought–used the word “aufgehübscht,” which I literally translated as “prettified,” but the copy editor felt that this kind of expression “did not belong into serious academic discourse.”
This is a prissy copy editor talking rubbish. For example, the accepted American translation of a passage by Walter Gropius in „Wer hat Recht? Traditionelle Baukunst oder Bauen in neuen Formen”, 1926 (i.e. during the Weimar Republic), namely ohne romantische beschönigungen und verspieltheiten is “without romantic prettification and cuteness”.
The German is five or six lines from the bottom of p.1 here. The American translation with “prettification”, near the bottom of p.1 here.
One odd thing about the document is that neither the German nor the English uses capital letters. That may be a Bauhaus typographic affectation.
Re Maria Theresa, I’m referring primarily to the “Ratio educationis totiusque literariae per regnum Hungariae et Provincias eidem adnexas” from 1777 for Hungarian and Croatian as well as Slavonian.
The English court hand and chancery hands, both used in law courts, are also descended from Gothic hands, and are as bad as Sütterlin.
writing “mınımum” in either italic or rounded handwriting is nothing but a series of zigzags or bumps.
As well as being very much er… prettier, an advantage of italic over copperplate handwriting is that gaps can be made between letters. This problem comes up with italic only when people like Hitler & Trump, to show how cross and forceful they are, draw everything as connected zigzags.
Christ how i used to hate those machine translation anecdotes – JFOM’s comment really drives home the point that they are not in any sense about machine translation (or any other kind of translation), but rather an updated iteration of an old xenophobic joke about how foreigners talk funny, like something that would have been forwarded to your email by an uncle in 1997 with the subject matter FW Fw Fw….etc.. In all fairness, however, the 1902 story in the LA Times concerns a relatively plausible utterance – flesh/meat and ghost/spirit are more or less identical in meaning, though not in usage, and this is exactly the kind of mistake second language speakers tend to make. It’s still a story about how stupid foreigners are, but nevertheless a fairly gentle and good-natured one.
By contrast, the present-day version with Russian and vodka is not a plausible sentence, machine or not; дух does not secondarily mean “alcohol” the way ‘spirit’ does so there is no way to get from spirit to vodka in the given expression. “Out of sight” is not the same “invisible” even to very small children. However in the logic of ethnic stereotype humor these makes perfect sense, since the stereotypical Russian is always drunk and not very talkative, so of course their computer would make translations that were terse and add mentions of vodka. (Incidentally there exist earlier versions of the joke using Russian but ending up with whiskey as the mistranslation of spirit – though based on the same faulty translation logic, it comes across as much less hostile and mean-spirited).
An element of the story that always irritated me intensely for less than rational reasons was the software being tested using backtranslation; how was the software written without the developers knowing Russian, I shouted at no one in particular (in my head). But of course they have to use it here because 1) there’s no joke without it, and 2) it sets up the idea that the accuracy of a translation is verified by translating it and translating back, and then comparing to the original (something that the intended audience likely tacitly believes already.) Given all that It’s not the worst set up for a joke, but it does require some recognizable connection between the two sentences, ie that the second is readily understood to be a poor translation of the first, ideally in some absurd and/or hilarious way. Alas they came up with neither. Well ok, they inadvertently came up with a good self-parody of people who distrust technology and/or hate foreigners.
Thematically, the change from human to machine translator also introduces into a lame old joke about foreign people the notion that automation isn’t all it’s cracked up to be, that machines can’t think like humans, that they’re prone to these kinds of extreme errors etc. These are all valid points but it’s really a shame that presented with such a rich satirical target the best we could do is recycle ethnic humor from two centuries ago
Nemanja, the xenophobic connotations didn’t strike me at all because I first saw that one before I had any notions about Russians. Still, having seen in for the umpteenth time, I find it contrived, lame and… suspect. It’s too pat and seems to me would only be told by someone who doesn’t know Russian (or any foreign language possibly).
The one by the Filipino student seems just possibly true… maybe. It’s the kind of mistake a student might make, not necessarily as a howler but as a puzzled guess at the meaning. Or it could just be someone making a funny joke about how curious English is. After all, it’s funnier to say “a foreign student got this one wrong! (haha)” than “isn’t this a curious saying!”