Hearts and Hearths


Saturday I met with Tatsumi Takayuki, an old friend from Honyaku Benkyokai days in Tokyo and author of Full Metal Apache. We actually met to talk about the articles being written for the upcoming Rampo Reader, but talked about quite a number of things on the way: the possible content of Speculative Japan volume 2, a number of novels that should be considered for publication in the future, upcoming Poe Conferences in Japan and the US, important new books in the field by Sari Kawana and Mark Silver, and more. And then we touched on the deliberate mangling of Japanese works by translators.

This is a constant problem in literary translation, because the translator (and the editor) have to make tough choices about how much alteration is ethical... how much of a source work can you change before it stops being translation and becomes creative writing?

I think the first time I encountered this was when the first Perry Rhodan novels came out in English, decades ago. I remember that one of key characters in the early books in the series was an elderly scientist named Khrest... and read later that in the original German his name was spelled Crest. Is that a reasonable alteration? I'd say it was, because the Crest spelling would strike entirely the wrong chord in American readers (the publisher was Ace Books, which meant they were only interested in the US reader). The translators chose a different spelling that preserved the sound of the original name but neatly eliminated the (highly misleading) suggestion that he had something to do with oral hygiene.

So does that mean it's OK to change names? Apparently, perceptions of whether or not it is appropriate to change a name have changed over the years. In Mark Silver's excellent Purloined Letters, he discusses a translation of Émile Gaboriau's The Lerouge Affair (1866) by Kuroiwa Ruikō (黒岩涙香,1862-1913), saying

In deference to his Japanese audience, Ruikō feels compelled not only to transliterate [proper names] into the Japanese sound system but to assign them ateji, or Chinese characters that approximate their sounds. As he says in his preface, "Things like place names and names of persons, since they differ from those commonly used in our country (wa ga kuni), are especially difficult to remember." Thus, he explains, Commarin, the legitimate son, will be referred to as "Komori." Noel becomes Minoru, Claudine because Oden, Clair become Kuretake...and so forth.

His use of ateji is perfectly reasonable, I think, because that was an accepted way to represent foreign words at the time in Japan. But was he justified in replacing, for example, Commarin with Komori? Commarin could have also been written, for example, Komarin and appropriate ateji selected. The jump from Clair to Kuretake is even more severe.

I would say this is unacceptable in modern translation, that Kuroiwa left the realm of translation and entered into a creative enterprise drawing on work in another language.

In fact, this is exactly what he did: he changed storylines and endings to suit his personal taste, or what he believed the reading public would most enjoy. He is recognized as an excellent writer of the time, and unquestionably did much to bring Western culture into a modernizing Japan, but by our standards I'm afraid he just took too many liberties with his pen.

Unfortunately, the practice continues. Publishing companies are in business to sell books, for the most part, not bring literature into other languages. Their loyalty is to selling, not to translation accuracy. For example, Rebecca Copeland, the translator of Kirino Natsuo's Grotesque, commented in the July 2008 issue of the SWET Newsletter

From the outset the editor was tasked with shortening the book. Minor characters were eliminated and scenes were cut, all in an effort to streamline the novel. I hated to see the deletions, having spent time with the chartacters that were cut. But I also understood that a novel like Grotesque would tax most American readers. The length wasn't the biggest obstacle; rather it was that the work is a "concept novel." It is not plot-driven.

I'm facing this problem myself right now, in my own translation of a short story by Tanaka Kōtarō (田中貢太郎) for our kaiki anthology. The story concerns a strange encounter in pre-Meiji Japan, and during the course of the encounter the main character sees a weird and unsettling face peering out from under a kamado, a Japanese stove generally consisting of a stone enclosure with a place on top to put a pot or teakettle. In fact, this kamado is also in the title of the story, and while it really has little to do with the story flow (it could as easily be a part of the house, for example) I feel it has to be preserved in the translation. But how?

Stove or oven, as the wikipedia article suggests, are simple answers, but to modern readers they suggest big lumps of steel, which is certainly not the right image. The word hearth came to mind, but a hearth is set into the floor, and there is no place for a face to peer out under from. Fireplace? Far too Western, especially for this particular setting. Now I'm thinking of using something like a "log rack" located under a hanging teapot, for example.

I don't have a good answer yet, and welcome suggestions, but it's proving quite difficult to come up with a clean way of explaining what this is without derailing the reader, or adding a boring footnote somewhere...


Kurodahan Press

Kurodahan Press
2305-9 Yunomae Machi
Kuma-gun, Kumamoto
868-0600 JAPAN