translation Archives - max gladstone

Posts Tagged ‘translation’

Translation is Weird

If I had a time machine and perfect language skills and were bound by some geas to use them only to answer weird literary questions, one of the first things I’d do would be go back to the Tang dynasty and ask Li Bai’s opinion about pronouns.

Here’s a great Li Bai poem, called 静夜思, which renders as “Silent Night Thoughts,” but the poem’s so iconic that if you ask Google to translate a page with that poem title, it’ll just read “Nostalgia”—the poem stands in for the whole experience. Anyway, here goes.

床前明月光，疑是地上霜。
举头望明月，低头思故乡。

For those of you who don’t do Chinese, here’s a simple, bad translation, courtesy of me:

Bright moon shines beside the bed
Like frost on soil
I raise my head and watch the moon
I lower my head and think of home.

Now, to continue this essay I’m about to do the thing you should never do, which is offer a character-by-character reading of a Chinese poem—thereby falling into the old Ezra Pound “Chinese is a language free of grammar, it consists of beautiful pure images!” trap. I’m doing this because anyone reading my blog at least speaks English, and while most English-speaking readers can look at, say, a Spanish poem and extract a little meaning, since the languages share common roots, they can’t often do the same with Chinese poetry. So, understand that there is grammar at work here, even though my character-by-character rendering will obscure that. Okay? Okay.

床前明月光，疑是地上霜。

chuang2 qian2 ming2 yue4 guang1, yi2 shi4 di4 shang shuang1.

bed-before-bright-moon-shine, as-if earth-(on top of)-frost.

举头望明月，低头思故乡。

ju3 tou2 wang4 ming2 yue4, di1 tou2 si4 gu4 xiang1

raise-head-watch-bright-moon, lower-head-think-old-country.

As you can see, my super-lightweight translation falls short. For example, to preserve English syntax I switched the image-order in the first two lines; I haven’t come up with a way to land the lines on “moonlight” and “frost-shimmer” respectively that doesn’t make the English read stilted. (That’s not to say such a rendering doesn’t exist.)

But I did greater violence to Li Bai’s original—or did I?—when I inserted ‘I’ and ‘my’ into the second couplet. The original poem does not, so far as I can tell, indicate that the speaker is the person raising his or her head. Nor, of course, does the original language indicate the gender of the speaker. (Fun fax: spoken Chinese doesn’t gender the third-person pronoun, and written Chinese didn’t gender the third person pronoun until IIRC the Westernization and modernization pushes of the late 19th and early 20th centuries. Hooray! Wait. [Though to be ‘fair’ you could read this as erasure, too.] ) To create an English version of Silent Night Thoughts, the writer has to decide: is the poet speaking? Is the poet describing someone else? If so, what’s that person’s gender? (He raises his head? She raises her head?) Is the poet addressing the reader? (“You raise your head”?)

And this is where I want my time machine. In English, this indeterminacy seems deliberate. You’ll rarely write a person-indeterminate English sentence without meaning to. But in Chinese poetry, that’s a straightforward task. In fact, formal restrictions can require it. So: would Li Bai’s readers have read an ‘I’ into the poem? A ‘he’? A ‘she’? A ‘you’? Would the indeterminacy operate for them the way it seems to operate in English, allowing the reader to flow freely through the poem, choosing to see it from the subject’s point of view or from an outsider’s, or from the point of view of a person in the Old Country thinking of his exiled lover? Or is this an artifact of differences in language construction? Would a contemporary reader even have recognized this indeterminacy?

And yes, the author is dead, and yes, Sapir-Whorf doesn’t work, but—how dead is the author really? Translating from a language I learned far too late to experience natively, I find myself asking all the time: is this what the author wanted to say? (Or, is this what the author’s intended audience would have read, which seems like the same question seen from the other angle…) And how false is Sapir-Whorf, when the translation process is nothing but wrestling with thoughts that are trivial to frame in certain languages, and nigh-impossible in others?

I am all but certain monographs exist on this subject. I haven’t read them; I don’t know what they would say. But whatever they do record, all the people to whom this poem was first delivered are over a thousand years dead—any experience of the poem they didn’t write down, we’re at a loss to reconstruct. And, depending on their linguistic background, it may not have even occurred to them to think about this issue.

I don’t know.

Hence, the time machine.

Hackeysack, Boozeomancy, and Virtual Languages

Hobbits don’t speak English.

We read their speech in English, sure, but the language in which Bilbo et. al. talk Longbottom Leaf bears little relationship to English as spoken in Tolkien’s day, let alone ours. We read a representation of that speech in English—that’s to say, a translation. And though I wouldn’t put it past Tolkien to have a detailed grammar of Western Common salted away in his papers, he wrote his manuscripts for the most part in English. So we have pipeweed and second cousins twice removed on the mother’s side, and birthday parties and country gentlemen turning the ripe old age of eleventy-one.

Translation, done right, is brilliant and difficult, but when done even a little wrong it can break the meaning and cultural associations of the source text. My favorite example: there’s an old Chinese sport / pastime that features prominently in Ming Dynasty fiction, which most modern translations render as “football.” Now, think about Ming Dynasty football. Have a vision in your head of what that would look like? Does it feature Ming Dynasty Pele or Peyton Manning?

Yeah, well, you’re both wrong. The term translated “Football” here refers to a game in which folks stand in a circle and attempt to pass a leather ball from one to another without using their hands. We’re talking, basically, about hackeysack as played by 15th-century Chinese gentlemen. I don’t know why American translators shrink from calling a hackeysack a hackeysack—except maybe that (a) it stinks of modernity (in much the same reason you can’t name a character in historical fiction about 11th century England “Tiffany” even though people back then were named Tiffany), and (b) it summons up weird cultural associations, mostly of skinny dreadlocked barefoot prep school boys kicking the sack with weed smoke heavy on the air and Widespread Panic playing in the background. (I guess that might be Mumford & Sons these days? I AM NOT COOL.) Now, I think those cultural associations are informative and interesting, but I’m not a professional translator and apparently there’s been a consensus of translation—but the consensus means uninformed readers of translations that describe the sport as “football” will have a picture of what’s happening in the story that’s as vivid as it is incorrect.

Language is weird. And it gets weirder in subcreated or “secondary world” fantasy, in which, ostensibly, neither English nor any of the hundreds of tongues it’s mugged for grammar and vocabulary exist. Do you like your secondary-world steampunk gentlemen to wear purple ascots? Then you’d better take care that your world has a Royal Ascot Club, because that’s where the word comes from. Anyone ever eat a sandwich? Where does that word come from? What do your characters drink? Wine comes from the French, rivverrun roundabout from Latin. Whiskey springs from a Gaelic source word, lager is German, vodka’s Russian, aqua vitae is Latin rendering of the meaning of the Gaelic, aquavit has similar origins but refers to something else entirely.

And of course, the physical correlates of all these linguistic artifacts have their own cultural significance! The ascot has the social connotations it does because of accidents of history—and the same’s true of spats, the necktie, golf, swing music, slam poetry, minstrels, druids, scotch, pinstripes, sagging pants, the zoot suit, the miniskirt, blue jeans, sequins. We could try to shuffle the significance of these symbols, but it’s rare to pull this off without utterly confusing the reader. We could try to invent new symbols whole cloth, but that way lies three-page descriptions of the significance of various characters’ modes of dress. Which is great if that’s the kind of book you want to write! But it’s a particular kind of book, meant for a particular audience.

The closer we get to a modern setting, the more we have to deal with modern words and concepts and frameworks: Dumpster’s a brand name, as are Kleenex and Xerox and Polaroid. Jazz is jazz because history. It’s easy to claim we see these things as complex and contingent because the modern world is complex and contingent, but I wonder if, say, 14th century France didn’t seem every bit as complex and contingent to people who lived there. There probably would be fewer brand names, sure, but it’s not as if fashion and prejudice are original to the 20th century.

There are many ways to deal with this in writing fantasy, and they’re all right when used well. One’s to use new language for old stuff with old connotations. That’s cool, but occasionally confusing. One’s to use new language for new stuff with new connotations. That works too, though it’s so easy to mess up by creating a world that’s too simple and too complex at once. (Readers may not be amused if, once they learn the seventeen new words you’ve asked them to, they realize your culture is a stripped-down analogue of Western European medieval feudalism. Then again, they may! Certain writers can make drying paint interesting. If you can get away with this, I doff hat and ascot alike.) One’s to use old language for old stuff with new connotations, a nice trick—one of my favorite examples that works is the position-swap of haute and rest stop cuisine in Samuel R Delany’s Babel-17, in which coq au vin is simple spacer fare, while burgers with French fries and ketchup are the height of elegance. It works because it’s funny, but even such a sharp writer as Delany has to spend half a scene in a very tight book highlighting the change. Another path is just to use existing words for existing stuff with existing connotations where it works, because readers know what a suit is, and they know what a cocktail is, and you can waste a disgusting amount of time trying to explain that a Fantasy Dark & Stormy is, you know, a Dark & Stormy—time that would be better spent building character, developing conflict, accelerating tempo, deepening tension.

I do a bit of all the above, and certainly there are other methods; the final one I listed is a favorite when I’m feeling cheeky or want to cheat in a slip of smooth exposition-free characterization, but it has weaknesses. Sometimes a Gin Mule will just throw people out of a story. And I don’t mean by, you know, a mule made of gin. Though that would be theoretically possible in a fantasy novel that contained, say, boozeomancers.

Hm. Boozeomancers. (*Makes note.*)

Anyway! To my mind this is one of the core fascinations of writing secondary-world fantasy: the creation of a working language and system of social connotations distinct from our own yet within our own, a sort of linguistic virtual machine. Tolkien walked these lines very well. He knew just when his characters should say “Namarie,” and when “I ain’t been dropping no eaves, sir, honest!” Sometimes the challenge feels an awe-inspiring. Sometimes it feels like kickboxing in a straightjacket: inherently limiting and on its face pointlessly difficult. But if you can pull it off, you’ll look so damn cool.