In the February 2002 issue of the monthly journal Eureka, Fusae Kawazoe gives a rundown of translations of Murasaki Shikibu’s “The Tale of Genji” — not only into foreign languages, but into modern Japanese as well. In doing so, the noted Genji scholar reminds us of the profound influence of Arthur Waley’s English translation, which was published in six installments from 1925 to 1933.
Subsequent French, Swedish, Dutch, German, Italian and Hungarian translations of “Genji” were based on Waley’s version. The process of secondhand translation is dubious at best, but in this case the results were remarkable. European readers came to rank the 11th-century Japanese author along with Miguel do Cervantes, Honore de Balzac, Jane Austen, Giovanni Boccaccio and Marcel Proust.
Not everyone was enamored of Waley’s translation, however. It provoked the poet Akiko Yosano (1878-1942), the one translator of “Genji” into modern Japanese who counted at the time, into making a new Japanese translation of the novel. Yosano, who by then had made two “Genji” translations (one partial and published in 1912; the other complete but lost in the Great Kanto Earthquake of 1923), apparently felt betrayed by the extravagant praise Japanese intellectuals gave Waley’s work. Hakucho Masamune, for one, announced that Waley “revived the original that appeared to be dead.” For the modern Japanese reader, Murasaki’s language is hard to decipher, said the foremost advocate of Naturalism of the day, whereas Waley’s translation is easy to comprehend.
Yosano, who had read “Genji” since her early teens, didn’t like this. “You do not read literature only for its meaning but for its unique language,” she wrote. ” ‘The Tale of Genji’ does not exist separate from the beauty of Murasaki Shikibu’s language.” She was implying, dangerously, that “Genji” could properly be appreciated only in Japanese.
Yosano, in any case, set to work on a third translation and completed it while battling deteriorating health. She wanted to demonstrate, Kawazoe suggests, that a translation into modern, colloquial Japanese would be superior to Waley’s English translation.
Waley’s work also provoked a reaction from novelist Junichiro Tanizaki (1886-1964). “Mr. Waley’s English translation of ‘Genji’ is reputed to be the one great translation in recent years,” the novelist wrote in his guide to prose, “Bunsho Tokuhon” (1934). But the praise may be misguided because Waley has greatly “supplemented” (“oginau”) or amplified the original, he said. To make this point, Tanizaki cited a passage from Waley and translated it into Japanese to show that Waley used twice as many words as Murasaki’s original.
You might say that if Yosano’s nationalistic reaction was rather strange, so was Tanizaki’s analysis. If you back-translate into the original language, you may well end up expending more words. Also, in his third (and for him, definitive) translation of “Genji,” Tanizaki used one-third more words in restating what Murasaki said. If modern Japanese requires more words than classical Japanese in saying the same thing, English can certainly require more words than Japanese in doing the same, can’t it?
The aim of Tanizaki’s analysis was even more questionable. He wanted to advance the claim that the Japanese people are laconic, Westerners loquacious. The only other example of Western verbosity he cited was U.S. novelist Theodore Dreiser.
And yet, where Tanizaki saw Waley exemplifying the Western proclivity to say things that the Japanese would rather leave unsaid, Edward Seidensticker simply saw a translator’s willfulness. In publishing his translation of “Genji,” in 1976, he said that Waley’s “amplifying and embroidering,” which is “continuous,” impels one to conclude that “Murasaki Shikibu has the worst of it all the way.” This criticism was notable because it came from a scholar who professed his indebtedness to Waley to be so deep that preparing a new translation “felt like sacrilege.”
So, let us look at the passage Tanizaki quoted to make his case. It occurs at the start of the “Suma” chapter. Prince Genji, the protagonist of the story, decides to exile himself to a remote coastal village because of the troubles he has created for himself in the Capital (Kyoto). Here’s how Waley imagined Murasaki would have written in English:
“There was Suma. It might not be such a bad place to choose. There had indeed once been some houses there; but it was now a long way to the nearest village and the coast wore a very deserted aspect. Apart from a few fishermen’s huts there was not anywhere a sign of life. This did not matter, for a thickly populated, noisy place was not at all what he wanted; but even Suma was a terribly long way from the Capital, and the prospect of being separated from all those whose society he liked best was not at all inviting. His life hitherto had been one long series of disasters. As for the future, it did not bear thinking of! Clearly the world held in store for him nothing but disappointment and vexation.”
Elegant and smooth Waley certainly is. But he is also extravagant. How much so becomes immediately clear when you put this side by side with Seidensticker’s rendition.
“He thought of the Suma coast. People of worth had once lived there, he was told, but now it was deserted save for the huts of fishermen, and even they were few. The alternative was worse, to go on living this public life, so to speak, with people streaming in and out of his house. Yet he would have to leave, and affairs at court would continue to be much on his mind if he did leave. This irresolution was making life difficult for his people. Unsettling thoughts of the past and the future chased one another through his mind.”
As Tanizaki noted, Murasaki doesn’t say anything like “It (Suma) might not be such a bad place to choose,” and Seidensticker doesn’t, either. When it comes to something like “the prospect of being separated from all those whose society he liked best was not at all inviting,” the only thing the original says is “furusato obotsukanakarubeki” (“the hometown would worry him”). Even though Seidensticker’s translation of this sentence, “affairs at court would continue to be much on his mind if he did leave,” may not be too “economical of words” (his characterization of Murasaki’s style), altogether he uses a quarter less words than Waley in re-creating in English what Murasaki describes.
Little wonder Tanizaki pointedly noted that Waley may add “seimitsu” (“precision”), but in so doing loses “anji” (“suggestiveness”).
So, how does our third English translator, Royall Tyler, handle the same passage?
“There was Suma, yes, but while someone had lived there long ago, he gathered that the place was now extremely isolated and that there was hardly a fisherman’s hut to be seen there — not that he can have wished to live among milling crowds. On the other hand, merely being away from the City would make him worry about home. His mind was in undignified confusion. He reflected at length on what was past and what was yet to come, and the effort brought many sorrows to mind.”
Here, let me note the changes in “Genji” scholarship over the past century.
When Waley decided to translate the novel, the scholarship wasn’t extensive. I have tried to find out what annotated texts he may have used, but learned only that whatever texts he may have had were unlikely to have been fully annotated. This may not have mattered much to a linguistic genius who famously observed: “since the classical (Japanese) language has an easy grammar and limited vocabulary, a few months should suffice for the mastering of it.” Still, the truth is, even dedicated Japanese scholars of classical Japanese find “Genji” opaque.
By the time Seidensticker started his work, a fully annotated text with interlinear notes indicating who is saying what was out, and he could use several other texts and consult translations, including those by Yosano and Tanizaki. Whether what he saw in Waley provided the impetus or not isn’t clear, but Tanizaki started translating “Genji” into modern Japanese in 1935 and in the end produced three different versions (published in 1941, 1954 and 1960, respectively). And Seidensticker was a Tanizaki translator.
“Genji” scholarship has greatly expanded in Japan and elsewhere since Seidensticker’s days, producing highly specialized studies such as the one by my friend Doris Bargen, “A Woman’s Weapon” (University of Hawaii, 1997), which focuses on spirit possession in the tale. Among the theories advanced has been an argument — initially elaborated by Japanese scholars, then taken up by some of their American counterparts — that since “Genji” is mostly written in the present tense, an attempt should be made to re-create the effect in English (and other languages) as well.
It may be in reflection of these developments that Tyler acknowledges the existence of the Waley and Seidensticker translations, but doesn’t talk about either the way Seidensticker spoke of Waley’s. On the other hand, he makes it clear that he differs from Seidensticker in his assessment of the basic subject-matter: Murasaki’s prose. Where Seidensticker speaks of a language “brisker and more laconic” than Waley’s English suggests, Tyler speaks of “the tale’s many long sentences” and “its evenness of flow.” And each, naturally, tries to re-create his view in translation.
So, in the opening part of the passage referred to earlier, Waley uses five sentences, Seidensticker three and Tyler just one, which is what Murasaki does, though this observation assumes that such comparative parsing in such different languages has any value. Despite his best effort, Tyler often fails in this particular endeavor. But his effort is commendable. As may be expected, Tyler uses even less embroidery than Seidensticker — in this passage, 88 vs. 100 words.
What is curious in this light is Tyler’s decision to apply the 5-7-5-7-7 syllable pattern to his translation of the tanka in “Genji.” The novel is full of them — 795, by Tyler’s count. As he notes, however, Japanese is a polysyllabic language, so applying the syllabic count in English translation means embroidery. Indeed, Tyler’s verse translations are often as extravagant as Waley’s prose translation. Some prefer nettles, as Seidensticker put it in translating the title of one of Tanizaki’s stories. Tyler’s decision nonetheless strikes me as a strange error. The contrast between verse and prose is conspicuous.
Is Tyler’s translation superior to those by Waley and Seidensticker? In its overall accuracy, yes. Will it supersede them? Probably not. I say this because on the question of faithfulness to the original alone, a translator faces many choices.
Take the word “himegimi,” which occurs a few sentences after the passage we’ve seen, in the Japanese original. It would normally be translated as “princess.” Yosano, who usually makes explicit the person suggested or referred to but unspecified, translates it as “wakai fujin” (“[Genji's] young wife”). Waley names the person — Murasaki. Tanizaki, who attempts to be as rigorously faithful to the original as he can, uses the word as is and in a footnote tells the reader that Murasaki is meant. Seidensticker follows the example of Waley. Tyler translates the word as “his darling.”
Which is better? Which do you prefer?