So for my thesis I’ve spent a lot of time wading around in the original text of these 1600-1850 stories, but premodern Japanese isn’t exactly my native language, I read slow, and so I also leaned pretty heavily on extant English translations to help me find the places I ought to focus my attention on.
They’re kind of terrible sometimes.
I’m reluctant to call anybody out by name, because I know how much work goes into translation. It’s not a lucrative field, the people who translated these books did it for love not money and I’m greatly beholden to them for it — their translations made my research possible. Not to mention how much easier it is to nitpick a few points than it is to translate a dozen, or a hundred, or a thousand pages of neo-classical Japanese. And I also know that if I do wind up going into academia, it’s not going to endear me to potential future colleagues to be on record shredding their shit.
But on the other hand, you’re accountable for what you publish, and some of these errors are pretty egregious.
Point the First: Grammar is fucking important
Okay, so Japanese is a null-subject language, which means that if the grammatical subject is clear from context, you can omit it. Basically, where we’d use a pronoun in English (because you know who’s being referred to), they just drop it altogether.
English: I went home.
Japanese: Went home.
English speakers see this and flip the fuck out. Particularly in translation, because when you have a null-subject sentence without a clearly-defined actor, as a translator it is painful, on a physical level, to just make one up. Translators are understandably wary of inserting anything into the translation that wasn’t in the original, something that they couldn’t necessarily justify if challenged on it, and so the two strategies tend to be:
(1) Make it really vague. “Someone once wrote…” “They say that…”
(2) Make it passive. “It is said that…”
I object to both pretty strenuously. Preparing to address that in my paper, I wrote:
In the face of a context-less null subject, often the English-speaking translator’s impulse is to render the verb as a passive (“it is not known”)
I got the draft back from Professor Lady with the comment “That is not a correct translation.”
DON’T TELL ME, TELL THEM.
My hand to god, this is not just me being pedantic, not when the null-subject is “I” and some sloppy translator has just erased the first-person narrator that any native speaker would identify as such.
Or if you want something more meaningful, Jay Rubin points to the inscription on the atomic bomb memorial at Hiroshima:
And its passivized English translation, “Rest in peace, for this mistake will not be repeated.” When it actually reads “because [SOMEONE] will not repeat the mistake,” which throws you face-to-face with the question of WHOSE fault it was, in a way that a passive sentence lets you sidestep. There’s a reason politicians use passive sentences when they need to apologize for something.
Let’s just say that MISTAKES WERE MADE in these translations, a lot.
“Mmm, yes, this is definitely a third-person narrative,” Professor Dude 1 murmured, looking very professorial as he peered at Ueda Akinari’s “Shiramine” over the rims of his glasses. “You can tell by the use of mi-mahoshi and the quotative particle, a first-person narrative would have used mitakute or something along those lines.”
“Mmmm,” I said.
“No. It’s just that Zolbrod translated it as first-person.”
Professor Dude 1 scowled. “…Zolbrod couldn’t translate his way out of a paper bag.”
Point the Second: Honorifics are fucking important.
I’m not just saying that because honorifics are what I wrote my thesis on, I wrote my thesis on them because they’re fucking important.
In fact, they’re often the vile enablers that make null-subjects possible.
Take the verb “say,” for example — default is iu. Honorific forms are notamau and ōsu. Humilific form is mōsu. All four of them mean “say,” and the distinction gets entirely flattened in translation.
And because they don’t need to be translated differently, a lot of second-language learners of Japanese just map them all to the same mental space. So then when they’re reading Japanese and come across any of the four, they automatically think, say, without registering which it was.
But it matters because sometimes it’s your only clue as to who the fuck is talking. Prime example occurs in Richardson’s translation of the Asai Ryoi story “Flying Kato,” in which Kato, a sneak thief, is having a conversation with Uesugi Kenshin.
It’s a conversation. Alternating inquit tags to iu (plain) and to notamau (honorific). Richardson translated the whole damn thing as Kato’s monologue.
(Richardson… manages to mistranslate honorifics almost every time they appear. >_>)
“How did he mess this up??” I demanded, appalled. “He is better at Japanese than me. He translated the entirety of Otogi-boko, which I could not have done. He understands so many things that I don’t. How could he have missed something so simple?”
“Well,” Professor Dude 1 said, unruffled. “Richardson did learn Japanese from the CIA.”
“I was talking with Royall Tyler once,” remarked Professor Dude 2, meditatively. (Royall Tyler being the latest person to tackle a translation of the gargantuan Tale of Genji.) “He said it wasn’t until the ‘Wakana’ chapter that he felt he’d finally grasped Murasaki Shikibu’s use of honorifics.”
“I asked if he’d, ah, gone back and fixed the earlier chapters, then…? He said no.”
With a pencil, I strike through a line of the Richardson translation and write in the margin: “This is a causative, not an honorific.”
Halfway down the page, I strike through another line and write: “This is an honorific, not a causative.”
Point the Third: Werds are pretty important too
Perhaps it’s not quite so damning a sin as outright mistranslation, but gawd, how some people have a tin fucking ear for language.
No sooner did he open the door of the sleeping chamber, than a demon thrust its head out at the priest. The projecting extremity was so huge that it filled the doorway, gleaming even whiter than newly fallen snow, with eyes like mirrors and horns like the bare boughs of a tree.
“Projecting extremity”? Really, Zolbrod? Really?
Or like this line from the cinematic opening of Kyokutei Bakin’s Hakkenden, as the main character is fleeing a doomed battle and turns back, Orpheus-like, just in time to see his father fall:
He stopped his pacing horse and, when he turned to look back, he heard the noise of war and the sound of arrows. Knowing the castle was about to fall, he saw the light of a fierce fire burning the sky. (Widmer)
“Sound” of arrows. Yes, I suppose screaming is a sound. “Toki no koe,” by the way, is not just “noise of war,” it is specifically the thing you holler as you run into battle — the voice you give at the appropriate time, as it were. Not to mention the really terrible ordering of the whole sequence, and the bizarrely juxtaposed participle. I submit for your consideration:
He reigned in his horse and looked back, toward the battle cries and the screams of arrows. He could see the castle about to fall, and the light from its roaring flames setting the sky itself ablaze.
This is also the translation that gave us “They were defeated refugees with nothing left. Master and servant alike were extremely hungry and tired.”
I think the words you’re looking for are exhausted and starving.
Hungry, you say? Were they hara ga hetta? Onaka ga suita? A bit PEKO PEKO, perhaps?
Or were they — as it says in the damned Japanese — ueru, aka, literally starving.
This is like the “sound of arrows” versus “screams of arrows” thing again — why on earth would you pick the phrase that is both further from the original and seriously fucking weak beer? It’s the worst of both worlds.
One more from this translation, because I can’t– I just can’t:
“He ceremonially picked up the piece of dirt three times and inserted it into his breast pocket.”
into his breast pocket.
I can’t decide which is worse — the weirdly clinical “inserted” instead of something like “tucked” or even “put,” or that “breast pocket” makes it sound like he’s wearing a sports jacket around 15th-century Japan.
(Then again, an anachronistic translation kind of suits the spirit of Bakin, who has his 15th-century Japanese dudes shooting each other with guns.)
Oh man, I thought, comparing the Japanese text of Hakkenden chapter 25 to Donald Keene’s translation of it, He didn’t do the whole chapter, he just selected a part from the middle.
It’s okay though! Because I remember seeing someone else’s version of the same chapter in a different anthology. Yup, there it is, “Shino and Hamaji,” translated by Chris Drake.
…which, curiously enough, begins at the same point as the Keene translation.
…and also ends at the same point.
You even read, bro?
To be sure, not everyone is terrible. Paul Gordon Schalow is very good at spotting the subjectivity cues of a first-person narrator even in the absence of first-person pronouns. (Although he does make null subjects overly vague sometimes.) Barry Jackman eschews false passives with the fervor of a convert. Anthony Chambers can make Ueda Akinari’s creepy-cool ghost stories reach across the centuries to raise the hair on your arms.
And really, I can’t wait until I finish my thesis (< 2 weeks!) and have the time to start on translations of my own. I've decided that it's more valuable, in the long run, to contribute to making Japanese literature more accessible than it is to write a paper that only a handful of people in the world will be interested in, even if the latter offers more in the way of short-term gains. (I mean, we all hope that our turgid academic esoterica will be ground-breaking, but how often is it, really?)
So let’s finish the thesis and then do this thing. Let’s put my name on some things, put ‘em out there, and then everyone I’ve flung spitballs at is welcome to come take a swing at me. I’m looking forward to it.