I’ve been working on an ongoing translation project for the past four months. It’s being released in the USA under the title Unbeatable Banzuke on the G4 cable station.
The show was called ç‹è‚‰ç•ªä»˜ (kinniku banzukeâ€””Muscle Ranking”) in Japanese, and aired about ten years ago.
Yesterday, I caught just a few minutes of a segment I had translated. From what little I saw, the production company hasn’t tampered much with my translation (as edited by my editor at the translation agency). The American version is kind of weird. They’ve got an American doing completely new voiceover, and his pronunciation of Japanese words is as bad as anyone who doesn’t know a lick of Japanese. The show closes with more completely new content in the form of a signoff by a Japanese-speaking announcer named Kei Kato, who was not a part of the original show. I’m not exactly sure what the point of this extra “local color” is. I’m also a little puzzled that they’d want the local color, but stick with such stridently Americanized pronunciation for Japanese words.
They also seem to have deleted all the original telops, including the many advertising the prize money for each event. This is reasonable, but since the contestants frequently make reference to the money they stand to win, I’m guessing they’ve probably edited those parts out. I’ll need to watch more to find out.
I’m translating segments of a Japanese TV show right now. It’s very different from my usual work, and not what I consider a strong suit, but the client seems happy with my work right now, so I’ll take it.
This particular show is of the “physical challenge reality TV” variety. It’s sort of like the show Ninja Warrior that’s currently on U.S. cable, but much sillier and with recognizable Japanese TV celebrities as commentators and sometimes as competitors. The commentators are clearly trying to call the proceedings the way a sports announcer would, and as I go along, I’m trying to imagine how it would sound if Bob Costas were calling the games with my translation. But some of this stuff doesn’t translate. It’s not so much that I don’t know the words, or don’t know what the speakers mean by them (although that happens here and there), it’s just that they’re saying things that would never be said in the same situation in an English-speaking contest.
I just ran flat-faced into a perfect example. The game in question has the contestants trying to sit atop a gigantic ball and navigate it through an obstacle course. At one point, one of the contestants gets stuck in a hole and is rocking unsteadily and impotently, trying to get out. The commentator says “ã¾ã‚‹ã§ç¾ä»£äººã®æ—¥å¸¸ã®ä¸å®‰å®šæ„Ÿã‚’ãƒ“ã‚¸ãƒ¥ã‚¢ãƒ«åŒ–ã—ãŸã‹ã®ã‚ˆã†ã ,” which I have translated somewhat loosely as “Itâ€™s as if the malaise of modern life has been made tangible in his plight.”
And there’s the thing. No American sports announcer, no matter how literate, would ever say anything remotely like that in this situation. I’m content with the translation, but it’s undeniably weird to an American audience. Then again, the rest of the show is kind of weird.
I clicked through a link from a gadget site to a machine-translated press release for a new car-stereo head unit. I noticed that when my cursor hovered over a block of text, one of those floating mock-windows that are so popular in web2.0 appeared. It permits readers to enter their own translation for that sentence or chunk of text.
This is interesting, and something I hadn’t noticed before. It raises all kinds of interesting questions. Most obviously, how do they vet these reader-submitted translations? But it’s fascinating as a machine-translation paradigm. There are two general approaches to MT: one is basically lexical and grammatical analysis and substitution: diagramming sentences, dictionary lookup, etc. The other is “corpus based”, that is, having a huge body of phrase pairs, where one can be substituted for the other. And there is a hybrid between the two, that uses the corpus-based approach, but with some added smarts that permits a given phrase to serve as a pattern for novel phrases not found in the corpus (this is also pretty much how computer-assisted translation, or CAT, works). I wonder how these crowdsourced submissions work back into the MT backendâ€”if they’re used strictly in a corpus-based translation layer, or if they get extrapolated into patterns. I’m skeptical that they’re getting a significant number of submissions through this system, but if they did, the range of writing styles, language ability, and so on that would be feeding into the system would seem to make it incredibly complicated. And perhaps a huge jump forward in improvement over older MT systemsâ€¦but perhaps a huge clusterfuck of unharmonized spammy nonsense.
Paul Graham always writes interesting articles (though I can’t figure out for the life of me why he hosts them as a Yahoo store), but I don’t track him very closely, so when I ran across a link to his somewhat old Writing, Briefly, I read it eagerly.
And noticed with interest that it has been translated into a number of languages, including a Japanese version (which I can read), and a Spanish version (which I can kinda fake). His advice â€œuse simple, germanic wordsâ€ may be good (though I’d phrase it as â€œAnglo-Saxon wordsâ€), and as a translation issue, it certainly stands out.
The Spanish translator preserved it but struck it out: â€œusa palabras simples,
The Japanese translator included it without comment: â€œç°¡å˜ã§ãƒ‰ã‚¤ãƒ„èªžçš„ãªå˜èªžã‚’ä½¿ã„ãªã•ã„ã€‚â€
Now, the funny thing here is that there’s a pretty good equivalent to Anglo-Saxon vocabulary for Japaneseâ€”å¤§å’Œè¨€è‘‰. I’m not how perfectly the two accord, or whether avoiding æ¼¢èªž would be as important to a Japanese version of Paul Graham as avoiding Latinate words apparently is to the English-speaking Paul Graham. Regardless, though, the translator kept that in there. Apparently the translator is relying on the reader to keep in mind that this is a translation of an English text for English audiences, and to understand what Germanic vocabulary means in terms of English style. Going the other way, I would never make that assumptionâ€”an English audience would be completely lost if I presented them with the phrase â€œyamato kotobaâ€ in a text translated from Japanese. But then again, it might be jarring to them if I adapted the concept to â€œAnglo-Saxon vocabularyâ€ if they knew that my piece was a translation. One could dodge this by simply saying â€œuse native vocabulary.â€
The case with Spanish is knotty in its own way. I know Spanish has its share of loanwords, but it doesn’t have the overwhelming influence of French and Latin that English has (and even if it did, it would be harder to tell them apart), or of Chinese that Japanese has. So the call to use native vocabulary is redundant. The way the translator chose to deal with it here is interestingâ€”it transparently acknowledges that the text is a translation, and that in this case, the idea doesn’t quite fit in the translation.
I started a web page that I called the Honyaku Home Page back in 1995 (for whatever reason, the wayback machine only shows iterations going back to 1997â€”still pretty old for a web page). Over the years it grew and transmogrified. For a while, I ran it using a crude homebrewed CMS written in Hypercard.
When I upgraded my Mac to OS X, Hypercard became a non-option; my hacked-up CMS had been very awkward to use for a long time anyhow. Eventually I transferred most of the site’s functions to Movable Type, using some jury-rigged templates. That too became unsatisfactory, and for a very long time, I looked around for other options.
I found one in Drupal, though when I first encountered it, it was somewhat primitive. Over time, Drupal has developed, and I committed to using it. After a false-start, I hired a developer to customize a module for me, bought a domain name (which I should have done a long time ago anyhow), and soft-launched the new site.
Even though most of the ducks have been in a row for some time, I’ve been reluctant to have a public launch. Drupal has numerous add-on modules these days, many of which would no doubt make the new site more useful, but the double-edged sword is that every new module creates new administrative tasks, and some of the spiffier features would require even more futzing around. I was stuck in option paralysis. The best is the enemy of good enough.
Today, I finally said â€œfuck it,â€ decided to launch with a plainer site, and announce it. Check it out. honyakuhome.org.
Language Hat posted links to some translations of Jabberwocky, which has been a pet interest of mine ever since I saw the French and German translations in GEB. Someone, years ago, sent me a bunch of Japanese translations of Jabberwocky, and they’ve been languishing on my hard drive ever since. Now seems like a good time to get them out there.
I am posting these with minimal formatting because I’m lazy. Headings are translator names. Notes and credits are as I received them. See after the jump.
Post-move, I’ve been cleaning out some old papers, and found this. I’ve decided to type it up and post it online for the benefit of future generations. This was originally typed up (and orchestrated) by Chris Poole. Although I’ve tried very hard to reproduce this in exactly the same form as Chris typed it up, it’s quite possible that I’ve introduced a few typos.
I don’t remember exactly which one of these I translated, but it was somewhere in the late 30s/early 40s.
At the closing luncheon of IJET-4 an exercise in consecutive translating was conducted, drawing on the expertise of the assembled translators and interpreters. A simple phrase in English was chosen as the starting point and a Japanese speaker was asked to translate it. This in turn was translated back into English, and then back into Japanese again and so on. People were asked to translate into their own language and were given sixty seconds to do so. No one saw anything but the previous version, and were therefore unaware of the subtle changes that were taking place.
It should be noted here that some difficulty was encountered due to people’s handwriting, but as the participants became aware of the overall objective, a guarantee of anonymity seemed to become more important. In deference to these numerous requests I therefore present the results typed up, with annotation where appropriate.
- Bridges between cultures are built on foundations of tolerance.
- Patience, indeed, is the foundation of bridges between cultures.
- æ–‡åŒ–ã®ã‹ã‘æ©‹ã«ãªã‚‹ã®ã¯ã€å¿è€ã—ã‹ã‚ã‚Šã¾ã›ã‚“ã€‚ â€œFoundationâ€ component of metaphor disappears.
- The only cultural bridge is forbearance å¿è€ alternatively translated as â€œtolerance,â€ â€œpatienceâ€ and â€œforbearanceâ€. The latter perhaps confusing the translator, who finds refuge in an ambiguous use of the word ç†è§£ which then of course becomes â€œunderstandingâ€. A very durable concept which lasts until 21.
- æ–‡åŒ–ã¯ä»–ã‚’ç†è§£ã™ã‚‹ã“ã¨ã§çµã°ã‚Œã‚‹ã€‚ã€€â€œBridgeâ€ metaphor disappears via çµã³ and â€œlinkâ€.
- Cultures are linked by understanding others.
- ä»–ã®äººãŸã¡ã‚’ç†è§£ã™ã‚‹ã“ã¨ã«ã‚ˆã‚Šæ–‡åŒ–äº¤æµãŒãªã•ã‚Œã‚‹ã€‚ â€œPeopleâ€ are introduced through the ambiguity of ä»–.
- Cultural exchange is done by evaluating other people.
- International understanding begins with an understanding of foreign people.
- International understanding begins with an understanding of foreign people.
- International understanding begins with the act of understading foreigners.
- Understand first that behavior is to understand the behavior of foreigners. Statement becomes rather incoherent imperative due to confusing layout of 16.
- å¤–å›½äººã®è¡Œå‹•ã§ã‚ã‚‹ã¨ã¾ãšç†è§£ã™ã‚‹ã“ã¨ã€‚ Does not read 17 as imperative.
- To understand from the outset that this is the way foreigners behave. Seems to become conditional clause here.
- å¤–å›½ã®æ–¹ã¯ã“ã†ã„ã†ãµã†ã«è¡Œå‹•ã™ã‚‹ã‚‚ã®ã ã¨åˆã‚ã‹ã‚‰ç†è§£ã™ã‚‹ã“ã¨ã€‚
- You must understand that this is how foreigners behave. Back to the imperative.
- å¤–å›½ã®æ–¹ã¯ã“ã†ãªã•ã„ã¾ã™ã€‚ Then back again to descriptive statement.
- This is the way foreigners would do it. â€œwould do itâ€ if what? Do what?
- ã“ã‚Œã¯å¤–å›½äººãŒã‚ˆãã™ã‚‹ã‚„ã‚Šæ–¹ã§ã™ã€‚ Solves above problem, but introduces question of frequency.
- This is what foreigners often do.
- å¤–å›½ãŒã©ã‚“ã‚ã“ã¨ã‚’ã‚ˆãè¡Œã„ã¾ã™ã‹ï¼Ÿ Inexplicably becomes question. Also omits äºº, leaving sentence to mean â€œwhat sort of things do foreign countries often do?â€
- What kind of things do they like to do in foreign countries? In order to make sense of the above, invents identity/ies, not necessarily native to the countries, who now have a choice about â€œwhat they doâ€.
- ãã®äººãŸã¡ã¯ï¼ˆã‹ã‚Œã‚‰ã¯ï¼‰å¤–å›½ã«ã„ã£ãŸã¨ãã©ã‚“ãªã“ã¨ã‚’ã—ãŸã„ã®ã§ã—ã‚‡ã†ã‹ã€‚ Good, if cumbersome, translation that makes it plain that â€œtheyâ€ are visitors.
- What do you think they might want to do when they go overseas?
- å¤ªã‚Šã™ãŽãŸã‚‰ã©ã†å¯¾å‡¦ã™ã‚‹ã¨æ€ã„ã¾ã™ã‹ã€‚ Handwriting problem. Misreads â€œoverseasâ€ as â€œovereatsâ€.
- If you are too fat, how do you handle the problem? Introduces value judgment on obesity.
- å¤ªã‚Šã™ãŽã¦ã„ãŸã‚‰ã€ã©ã†ãã‚Œã«å¯¾å¿œã—ã¾ã™ã‹ã€‚ Female translator said she would rather not translate something like this. I emphasised that it was only a game so she obliged (but didn’t see obesity as a problem).
- If you were too fat, what would you do?
- ãµã¨ã‚ŠéŽãŽã¦ã„ãŸã‚‰è²´æ–¹ã¯ã©ã†ãªã•ã„ã¾ã™ã‹ã€‚ Renders â€œyouâ€ as è²´æ–¹
- What will the lord do when he gets too fat? Mistakes è²´æ–¹ for è²´æ— and renders it as â€œlordâ€.
- ç¥žã¯è‚¥ã‚Šã™ãŽãŸã‚‰ã©ã†ã™ã‚‹ã‹ï¼Ÿ Reads â€œlordâ€ as â€œGodâ€.
- What do you do if God is too fat? 37, 39, 43, 47 all manage without a personal pronoun in Japanese. Personal pronounds cause problems on both occasions they appear in 34 and 40.
- What would you do if god was too fat?
- ç¥žãŒè‚¥æº€éŽå¤šã ã£ãŸã‚‰è²´æ–¹ã¯â€¦ Bases vague, open-ended question on condition that God were too fat.
- If God were too fat, what would you be? Good logical translation that deduces remainder of question.
- è‚¥ã‚Šã™ãŽã®ç¥žæ§˜ãŒã„ã£ãŸã‚‰ã©ã†æ€ã„ã¾ã™ã‹ï¼Ÿ Raises question of attitude rather than â€œbeingâ€.
- If there is an overweight God, what do you think?
- What would you think of a fat God.
- å¤ªã£ãŸç¥žæ§˜ã‚’ã©ã†æ€ã‚‚ã„ã¾ã™ã‹ã€‚ Rumoured fat God lives!
- What do you think of the fat God.
- ç¥žæ§˜å¤ªã£ãŸã§ã—ã‚‡ã†ï¼Ÿ Renders simple question as traditional Japanese greeting addressed to God.
- You look well God! Good translation.
- ã‚„ã‚ã€å…ƒæ°—ãã†ã˜ã‚ƒãªã„ã‹ï¼ Supreme being departs as â€œGodâ€ is read simply as exclamatory component of greeting.
- Hello my lover. You’m be lookin’ fine today (Devonshire) Very ably translated into equivalent dialect.
I’ve been a freelancer since 1989, and in that sense, I get a new job every time a new document comes down the pipe. I am still a freelancer, still doing the same thing, but I still feel like I’ve got a new job.
When I started translating as a freelancer, things were kind of thin. Gradually, my client base and workload picked up, and by 2000, I was making a pretty nice income.
Then came 2001: my income was less than half what it had been in 2000: the Japanese economy–already bad–seemed to get worse, and we all know what happened in the U.S. economy. 2002 was slightly worse; 2003 about the same.
December 2003 and January 2004 were alarmingly quiet, and I could no longer pretend that I was riding out a lean spell and things would pick up in their own time. Either I needed to get more translation work, or I had to get a “real” job (though the idea gave me hives). In the past, I had occasionally made efforts to get new work by cold-calling, but I found the process uniformly unproductive and had given up. Time to try again. I contacted a lot of translation agencies. Most of them thanked me for my resume (if that much) and that was the end of it. Some had me do translation tests. This turned into a more labor-intensive way for me to wind up buried in a rolodex somewhere. But I contacted one company that was different. They sent me a very demanding translation test–a long passage (as trials go) of very challenging material. I slacked a little on finishing it, but eventually did so. And eventually heard back that they like my work and want to add me to their stable. And that they can, it seems, completely saturate my pipeline. And they pay pretty well (especially for Americans). In short, if I choose to, I can pretty much work full-time and exclusively for them. I feel both relieved that a long and difficult period seems to be ending, and anxious that I might screw up.
Jenny and I have discussed before the danger as freelancers of turning one’s client ecology into a monoculture, but right now, it beats the hell out of a xericulture.