by Hiroaki Sato
Zabuzabu to kodomo ga aruku kawa no naka
Splash-splashing a child walks in the river
智慧もなく行く水もなき川の景
Chie mo naku ikumizu mo naki kawa no kei
Neither with wisdom nor with flowing water the riverscape
What kind of Zen did Kōi perceive in these haiku of Takayanagi’s? Devoted to Dōgen, the founder of Sōtō Zen in Japan, Kōi is known to have particularly prized the Zen master’s word, “At this moment only does my life exist.” Does that explain it?
Hearn, Bickerton, Hubbell: Translation and Definition
Let us begin by citing five different English translations of a single haiku—the most famous frog-pond one by the best-known haiku poet, Matsuo Bashō:
1.
Old pond—frogs jumped in—sound of water
2.
A lonely pond in age-old stillness sleeps . . .
Apart, unstirred by sound or motion . . . till
Suddenly into it a lithe frog leaps.
3.
Into the calm old pond
A frog plunged—then the splash.
4.
Breaking the silence
Of an ancient pond,
A frog jumped into water—
A deep resonance.
5.
An old pond
A frog jumping
Sound of water
Does this variety surprise you? Probably not. You have read some foreign writers and know they have been translated by different people. You may have compared some of them and marveled at how distinct they can be from each other. These translations of Bashō’s haiku are just five of the one hundred forty versions I once collected in my book One Hundred Frogs: From Renga to Haiku to English, but these may be enough to show how various people can come up with such a wide range of results. And the differences between them bring up some interesting questions.
For one, suppose the first translation, done in one line, says essentially what the original does. How, then, should we account for the explanatory translations of the second and fourth examples?
Second, I’ve arranged this selection chronologically—a span of more than eighty years separates the first from the last. The opening one, by Lafcadio Hearn, was published in 1898; the fifth one, by my late poetry teacher Lindley Williams Hubbell, was published in 1983. Can we then say that the so-called art of translation makes progress?
Third, as you must have noticed, more than one frog is suggested in the first translation but not in the others. Why?
Finally, how can these translations of a single poem have four different line formations? Is haiku such an amorphous verseform that the translator can cast it at liberty into any number of lines?
To state the obvious, the act of translation is greatly influenced by the translator’s knowledge and sensibility. With a poem, the translation is shaped by what the translator deems poetic at the time he does it. This may explain Curtis Hidden Page’s translation in 1923 that begins, “A lonely pond in age-old stillness sleeps . . .”
Page, a professor of English and French at Dartmouth in the first half of the twentieth century, compiled some anthologies of British and American poetry and translated a number of French works. Though he was the author and translator of Japanese Poetry: An Historical Essay with Two Hundred and Thirty Translations and a member of the Japan Society, he most likely did not know Japanese well; but even if he did, he must have felt free to recast Japanese poems into rhyming English verse, adding whatever words and images he thought necessary in the course of doing so. Yes, by then, Ezra Pound’s Imagist movement, which was inspired by haiku, had come and gone, and T. S. Eliot had published The Waste Land, but such work may well have been too avant-garde for the professor.
Conversely, you might say that Yuasa Nobuyuki’s quatrain, which begins by “Breaking the silence” and ends in “A deep resonance,” was based on Yuasa’s anxiety that the haiku, if translated straightforwardly, would be unintelligible to English readers of poetry. Yuasa, who published his translation in 1966, was a Japanese scholar of English whose notion of what makes an English poem tick seemed a bit outdated by then. The likelihood of that may be discerned from his application of the quatrain to a verseform that, in its standard configuration, consists of three syllabic units.
Does translation make progress? It doesn’t seem to. The versions by Page and Yuasa, prepared more than forty years apart, suggest, at the very least, that a good command of the languages involved and a good knowledge of poetic traditions do not necessarily make progressive or better translations.
What about the question of the number of frogs? Does the original haiku describe one frog, as four of the five translators say it does, or more than one frog, as Hearn says it does?
In 1999, a Japanese play about Hearn was staged at the Japan Society in New York. In a memorable subplot, one of Hearn’s Japanese students asks insistently, “Professor Hearn, may I dare point out, sir, that in your translation you say ‘frogs,’ but in this instance there could be only one frog?” Is the student correct? Or, more appropriately, is the playwright who seized on that notion correct?
The question arises because the Japanese language does not have the equivalent of the plural-making s as it exists in English. Another haiku that provokes perennial singular-versus-plural debate occurs also in the Bashō canon. I translated it in From the Country of Eight Islands as: “On dead branches crows remain perched at autumn’s end.” In the most recent translation I’ve seen of this haiku—in Haruo Shirane’s Traces of Dreams: Landscape, Cultural Memory, and the Poetry of Bashō—only a single bird on a single branch is signified.
With this particular haiku, though, the matter is complicated by the existence of several pictorial depictions of it. The earliest painting, with the haiku written in Bashō’s own hand, shows twenty crows flapping about in the air and at least seven crows in various lively postures perched in a tree that has shed its leaves. Another painting, also inscribed with the haiku in Bashō’s own hand, depicts but one crow perched on a branch of a tree, a dejected one at that. Considering the gregarious nature of crows, I prefer what appears to be the initial pictorial conception Bashō approved. Yet that doesn’t negate the validity of later interpretations—by Bashō, by his contemporaries, and by later readers.*
Was Hearn wrong about frogs, as one of his students asserts in the play? Or, as the Japanese critic Matsumoto Michihiro knowingly said in reference to my collection of 140 translations of this haiku, did Hearn commit a mistake in what he termed “cultural pattern recognition”? By this, Matsumoto meant that in one’s native culture one automatically recognizes how a certain thing signifies a certain other thing. In this instance, someone brought up in Japan should instinctively, and correctly, see that Bashō was describing a single frog. Or so Matsumoto thinks.
Well, then, did Hearn make a cultural mistake? Not really—especially if you believe an early account of how the haiku came into being.
One spring day, the story goes, Bashō and a couple of his friends were sitting around in his hut when “the sound of frogs dropping into the water [was] not frequent,” Bashō was moved to compose the haiku. Here again, kawazu, the original word for “frog,” doesn’t indicate whether there was one or more than one. But the genesis story strongly suggests more than one, if not a great many.
This very natural picture changed greatly as a Zen interpretation took over. As you know, in Zen, or in a certain branch of Zen at any rate, you are supposed to attain enlightenment at the caw of a crow, the guok—so Roger Tory Peterson heard the call—of a night heron, a flash of lightning, or the plop of a frog as it hops into the water. In any such instance, the impetus must be singular, and so it follows there could be only one frog.
In sum, not only linguistic ambiguities—and the singular-or-plural question is only one such example—but interpretativ
e differences sway what you do in translation.
How about the number of lines? How is it that five different translators have come up with four different lineations?
On this question, I must make it clear that the lineation of poems in translation is my hobby horse, and that I regard the haiku as basically a one-line poem and translate it accordingly—an exceedingly unpopular view among academics. Among such scholars, Ueda Makoto made his objection most strongly. In his book Bashō and His Interpreters, he wrote that although “premodern Japanese poets had no notion of lineation as a poetic device,” to “insist that a [haiku] should be a one-line poem in English because the original Japanese poet had no sense of lineation is tantamount to insisting that no English grammatical article, such as ‘a’ or ‘the,’ should be used in translating Japanese sentences because the Japanese language includes no concept of articles.”
Here, I’d simply point out my argument that even as the greatest majority of Japanese haiku writers regard the haiku as a one-line poem, some break up their pieces into lines, and that this situation compels the translator to make a choice: ignore the original lineation as printed or try to reproduce it. I have opted for the latter. I’d also say that modern prosodic analysis can be applied retroactively to classical or premodern poetry.
All this may be a negligible stir in the saucepan, with no kitchen-shaking consequences. Still, the haiku is one of the few cases, I venture, where the doings of translators—or the majority of them anyway—have helped shape the view of the verseform in foreign countries. For this reason, I wish to devote much of the rest of this chapter to this topic.
I assume that most of you have composed haiku, either in high school, junior high school, or earlier, or on the Internet. If you have, the definition that your teacher gave you in class or the rules you’ve followed on the Internet must go somewhat like this: Haiku is a “lyrical Japanese verse form . . . tending to emphasize nature . . . and the times of year, and consisting . . . of seventeen syllables arranged in three lines containing five, seven, and five syllables, respectively.”
Of course, the haiku that circulate on the Internet tend to be of the satirical variety, such as “spam” haiku, but most people interested in haiku would agree with the definition that I’ve just cited, though in a somewhat abbreviated form. It happens to be the most recent American definition I’ve seen, and it appears in Douglas Hofstadter’s highly opinionated book, to put it mildly, Le Ton beau de Marot: In Praise of the Music of Language—although I must say it isn’t entirely clear whether Hofstadter is describing Japanese or English haiku. If he thinks he’s describing Japanese haiku, he evidently hasn’t seen, let alone read, any of the haiku the majority of Japanese writers compose.
The Haiku Society of America, which formed a committee to work out definitions, was more circumspect and made a distinction between the trunk and its branches. After two years of deliberations, the committee came up with the following definitions:
1. An unrhymed Japanese poem recording the essence of a moment keenly perceived, in which Nature is linked to human nature. It usually consists of seventeen onji.
2. A foreign adaptation of 1, usually written in three lines totaling fewer than seventeen syllables.
Even this cautious description of Japanese haiku, as far as I’m concerned, leaves much to be desired. Why “a moment keenly perceived”? Why in a moment so perceived must “Nature” be “linked to human nature”? Above all, why introduce the pedantic and utterly wrongheaded term onji where the word “syllable” will do just fine?
My quibbles aside, the Haiku Society does not say, as Hofstadter does, that the Japanese haiku consists of “seventeen syllables arranged in three lines [emphasis mine] containing five, seven, and five syllables, respectively.” This is important. The best definition of the traditional haiku I’ve seen says that it is a one-line poem with two main descriptive elements that can’t be divided into lines. The “traditional haiku” means one that is composed of 5, 7, 5, or a total of seventeen syllables.
More recently, the avant-garde haiku poet Natsuishi Ban’ya has offered the following definition:
The haiku . . . is composed of two, three, or four short phrases (shōsetsu). However, the length of each such short phrase cannot be regulated or restricted.
Here, Natsuishi no longer talks about 5, 7-syllable units, but instead refers to the much vaguer “two, three, or four short phrases,” adding that “the length of each such phrase cannot be regulated or restricted.” This draws attention to a branch of modern haiku which is often ignored—not only by foreign students of haiku but by the majority of Japanese themselves. This branch, which started in the early part of the twentieth century, is made up of pieces that ignore 5- and 7-syllable patterns and are shorter than the standard total of seventeen syllables—or much longer.
I have translated a substantial selection of one prominent poet in that branch, Ozaki Hōsai, in Right under the big sky, I don’t wear a hat, with Cor van den Heuvel’s preface and Kyoko Selden’s full introduction. The title of the book itself is a translation of a haiku made up of fifteen syllables. One of Hōsai’s shortest pieces consists of nine syllables, and it reads:
咳をしても一人
Seki o shite mo hitori
I cough and am still alone
Ogiwara Seisensui, Hōsai’s friend and publisher, tended to write haiku much longer than the standard seventeen syllables, so much so that one editor, in apparent exasperation, decided to ignore the ones he regarded as excessively long. Here’s one of Seisensui’s haiku, which consists of twenty-six syllables, with punctuation:
クオレを読み、浪の音、なおも読めと云うのを読み終り
Kuore o yomi, nami no oto, naomo yome to iu no o yomiowari
Read Cuore, the sound of waves, read even more he said but I finished reading it
Cuore (Heart) is a novel by the Italian writer Edmondo de Amicis about a nine-year-old boy living during Italian unification. It became popular throughout the world, or so I understand. Here, the use of punctuation reminds us: At the very start of his career as a haiku poet, Seisensui broke up some of his pieces into two lines, which he said he did under the influence of Goethe and other German writers. But he soon stopped doing so and recast them into one line; he never resorted to multilinear forms again, though he used punctuation, on rare occasion, as he does here.
Natsuishi doesn’t mention lines or lineation in his definition, just as the Haiku Society did not in its definition. His silence on lines is especially notable because he is a student of the famous lineator Takayanagi Shigenobu, who wrote pieces like:
遠望の Embō no In a distant view
重き omoki a heavy
曙 akebono daybreak
戦きはじむ ononoki hajimu begins to tremble
The syllabic formation of this example is 5-3-4-7. Takayanagi lineated all his haiku except for the poems in his first and last books where everything’s written in one line. His lineated haiku range from two to more than ten lines, though the majority consist of four lines.
The haiku lineator I’ve most recently translated is Kamiyama Himeyo. She has composed pieces such as:
子 shi beginning
宮 kyū with
から kara the
はじまる hajimaru womb
円形 enkei a circu-
の 欠 落 no ketsu raku lar defi ciency
Kamiyama’s haiku, which mostly fall into 5-7-5 syllables in the aggregate, range from one to a dozen lines; when she also provides them with roman transliterations, they become even more fragmented. The question is: If you break up monolinear haiku into lines, what do you do with multiline haiku? P
ut them in one line, perhaps.
Natsuishi seldom, if ever, breaks up his haiku into lines. Conversely, even the members of the majority who cast their pieces in one line in print break them into lines when they present them in their own handwriting aesthetically, mostly in ink and brush, as on decorative papers or sheets called shikishi and tanzaku.
How then has translation helped create a new multiline form? The answer is almost tautological: With the majority of translators casting their translations in three lines, the view has taken hold that the haiku is a three-line poem. It has also created the sense, among Japanese haiku writers and observers, that haiku need to be broken up into lines in translation. A reverse intercultural influence, you may say, as happens often between Japan and the West.
The matter was not always like that.
As far as I know, Hearn’s initial response to the haiku form, an example of which we’ve seen at the beginning of this chapter, was the most natural: With no fuss, he took it to be a one-line affair:
Old pond—frogs jumped in—sound of water
Hearn’s contemporary, the scholar Basil Hall Chamberlain, unlike the world-wandering journalist that Hearn was, paid more attention to the English prosody prevailing at the time. Where he could, he turned haiku into a couplet, with each line in iambic tetrameter. Some who followed him applied the heroic couple to the haiku form. But William Maxwell Bickerton, a New Zealander who moved to Japan soon after graduating from Victoria College, published an account of Kobayashi Issa in the magazine Transactions of the Asiatic Society of Japan in 1932, titled “Issa’s Life and Poetry,” in which he reverted to Hearn’s approach, so to speak, and translated Issa’s haiku in one line. Here are two examples:
Ware to kite asobe ya! Oya no nai suzume.
Oh, won’t some orphan sparrow come and play with me.
Mata muda ni kuchi aku tori no mamako kana.
Once more in vain the stepchild bird opens its beak.
Did Bickerton’s approach raise eyebrows among his fellow scholars? I don’t know.† Anyway, in time and somewhere along the way, the notion that haiku cannot be translated in one line became a matter beyond dispute. By the time Burton Watson and I published From the Country of Eight Islands: An Anthology of Japanese Poetry, it had become so ingrained that William LaFleur felt compelled to observe, in reference to my one-line translations: