[an error occurred while processing this directive]

The Translator's Voice: An Interview with Hiroaki Sato

Hiroaki Sato

By Nicholas J. Teele

Hiroaki Sato (b. 1942) is unique among translators of poetry in that he translates from his native language, Japanese, into a foreign language, English. In order to do this successfully, Mr. Sato has seen the advantage of adding to his own considerable genius for the sounds and nuances of the English language by enlisting the assistance of such very able American translators and poets as Burton Watson and Michael O'Brien, who further refine his translations. In resolutely trying to transpose into English as much of the tone and meaning of the original poem as possible, Hiroaki Sato has consciously selected those poems which most clearly lend themselves to translation. Through such a process, Mr. Sato cleverly avoids the traps of a translator who tries to tackle poems that are overly difficult to translate, and at the same time steers away from the need for excessive notes to explain poems to readers unfamiliar with Japanese culture.The results are to be seen in the nine volumes of translations of Japanese poetry Hiroaki Sato has published since 1973, beginning with Poems of Princess Shikishi (d. 1201).His most recent book is Chieko & Other Poems of Takamura Kotaro (d. 1956), published this spring by the University Press of Hawaii.Concerning his work, Gary Snyder has said, "Hiroaki Sato is perhaps the finest translator of contemporary Japanese poetry into American English."

Before getting into the interview, I'd like to take a moment to congratulate you on the anthology of Japanese poetry you have just finished with Burton Watson and J. Thomas Rimer. What is the scope of the anthology?

Our anthology covers the whole history of Japanese poetry, from the Kojiki, compiled in 712, to modern poets. In its emphasis on individual poets it does not depart radically from the traditional approach, but it does differ from, say, the Penguin Book of Japanese Verse, in the number of poems translated for each of the major poets. For example, you will read 60 tanka by Saigyo (1118-90), 90 hokku by Buson (1716-83), and 100 haiku by Ozaki Hosai (1885-1926). It also has six range sequences in complete translation--for the first time, I believe--and some long poems by modern poets, such as "Sharks" by Kaneko Mitsuharu (1896-1975). It is called Japanese Poetry: From the Country of Eight Islands and was published by Doubleday Anchor Books in January of 1981.

How do you select a poet to translate? Do you do it on the basis of personal preference, on the fame or representativeness of the poet, or on the reviews a poet has received?

Preference and fame combined. first, the matter of fame: fame, if anything, means that the person's work has been looked at by a certain number of people and judged to be worthy; that is, to me, a hint, a direction. Once, the Shantih editor asked me to submit some translations from the Japanese for the magazine's special issue on women, so I picked about twenty names, mostly unknown to me, from the Gendaishi Techo directory and wrote to them asking for their books. They kindly complied.Unfortunately, the forty-odd books I received in a few weeks were mostly private editions--though quite expensively printed--and I could not find more than several poems I thought good enough to be translated. Like many other translators, I'd be happy to "discover" a few unknown poets, but that hasn't happened yet.

My preferences are inevitably my prejudices, my quirks. There's one famous poet I feel very reluctant to translate, because he has written a poem for children about hitokui dojin, which more or less is like saying cannibal nigger. "I'm the first to admit to being a racist, but anyone who writes such a poem and then reprints it in various editions as he has done is, I conclude, unaware of anything. Then there's a poet whose pieces I like and translated--until I met the person. Immediately, whatever quality I may have found attractive in her poems paled and vanished.

From a less personal viewpoint, I find puzzling and disheartening those who continue to write haiku and tanka in the same old traditional patterns and using pseudo-literary language. One reason that Ozaki Hosai is the most recent haiku poet included in our anthology is that he broke the traditional pattern and wrote in the language he used in his daily life.

I don't know if this is something different, but one decisive thing in selecting poets for translation has simply been a matter of volume.Often, if a poet doesn't have 2, 10, or 100 pages of poems that are translatable or worth translating, I have dropped him or her for that reason alone.

Once you've selected the poet, do you translate everything by that person, or do you have some process for selecting the poems?

I have never translated everything by one poet. I translated everything in Hagiwara's first and second books of poems for my Howling at the Moon, but of course he wrote more than twice that. Actually, once I pick a poet, the process of selecting his or her poems is mechanical. If a poem looks relatively good but contains a word or phrase that's incomprehensible or whose apparent incomprehensibility does not appear to warrant the effort for further research, I usually drop it. One recent example is a poem by Ono Tozaburo (b. 1903) that is always picked for an anthology. In it, he uses one foreign word I couldn't figure out. So I wrote him. Though ill, he kindly replied and explained what he meant by that word, but he, too, didn't know the spelling of the word. So I had to drop the poem. With unannotated classical tanka and hokku, the process becomes far more mechanical: I simply skip those which I don't understand. So it's fairly easy to make a selection out of the 400 hokku by this poet or the 800 hokku by that poet.Then I sit down and translate the selection. But since some in the selection don't come out well in English, the selection will be narrowed. Then, the friends who go over my translations will object to his piece and that, so it will be narrowed further.

Let me flatter myself a bit, though, by bragging about our anthology.I translated a bout 350 tanka out of the total of about 400 attributed to Princess Shikishi, but I'll be using only a quarter of that in the anthology. Another such example is Buson; I translated about 200 hokku by him, but as I said earlier, I'll be using only about 90 of them.

The fact that you're not only working from your native language into English but that you're also doing it extremely well makes me wonder what sort of methodology you have in translating poetry.

I have been extraordinarily lucky in having friends who are competent, meticulous, and willing to spend an inordinate amount of time going over my translations. Michael O'Brien is one of them; he helped me with Miyazawa Kenji, the first one in Chicago Review Press series of modern Japanese poets. I'm pleased to note that Michael is praised by Christopher Middleton for his "perfect translations" (Translation Review, No. 3, 1979). Robert Fagan is another. He has helped me with the three other poets in the same series and, more recently, with the classical poetry in our anthology. Burton Watson is another person who has helped me. They are all different from one another. Michael is intuitive. Robert is spartan in his approach to English grammar but at the same time wiling to experiment with it. Without him I'm not sure I could have translated Yoshioka Minoru's poems, for example. Burton Watson, the only person among the three who knows Japanese, is generous. As I myself have learned the hard way, not many translators like other people's translations from the same language, the same material. Burton Watson goes out of his way to look at what you have done from your viewpoint. That alone makes him extraordinary.

About your question on methodology, I don't know that I have one, but I try to be faithful to the original. One reason is that I translate out of my native language into an acquired one. When I often don't quite know, say, whether I can or can't drop "the" or "a," how can I tamper with what the original poet says? To do so would doubly distort the result. Another reason I try to be faithful to the original is that I translate a large number of poets. I don't think I should impose my preferred way of writing uniformly on all of them, though, of course, my success or failure in bringing out different voices is another matter. I also think I've come to believe in that approach through my job. I work at a Japanese government agency in New York where I translate, into and out of English and Japanese, letters, reports, speeches, marketing research specifications, and so forth. The majority of them can't be translated faithfully. Take letters, for example. A typical Japanese letter would begin, "The refreshing season, autumn, is here with us again. May we take this occasion to congratulate you on the increasingly healthy prosperity of your firm. Assuming that you will allow us rudely to change the subject, we were somewhat disconcerted to discov er the other day that you appear to have overlooked the payment of the amount we billed on your firm last year, etc." It would be nice if some American law attorneys and vice presidents wrote in a similar manner, but of course they don't. So, when I translate letters and research specs, as they're called, I must pick the points to be made and recast them, unless I want to look charming or incompetent. There is, however, certain writing that requires faithful translation. One such example may be what they call a "sensitive" document spelling out, say, a bureaucratic consensus on an important policy. Translating a poem is, I have come to believe, like translating such a document.

I try to be faithful to the original lineation, too. I translate tanka and haiku as they are printed, which is to say, mostly in one line, because modern poets have attempted to lineate them. If you translate regular tanka in five lines, how do you translate a group of tanka by Miyazawa Kenji, for example, written in varying numbers of lines though still retaining the 5-7-5-7-7 syllable count? I don't think my approach is a solution to all the knotty problems, but I'm hoping my translations in our anthology will at least invite some people to take a fresh look at the translation of tanka and haiku.

What British or American poetic models or influences do you feel have been and are important factors in the translations you have done?

I'm not sure that I can pinpoint any such model or influence. In using English, I am grateful that I met Mr. Lindly Williams Hubbell in college and took his course on poetry writing, among others. Burton Watson said of him, in a recent letter, "I am always amazed at how much he knows, and how wonderfully lucidly he write about what he knows." Yet, when one of us students asked him, "How many languages do you know?" Mr. Hubbell said, "Half of English." He is a source of unswerving encouragement and at the same time a constant reminder that whatever I do in English isn't worth a penny.

As for theories of translation, by the time I read Nabokov's short essay on translating Eugene Onegin, for example, I'd been translating for some time, and all that essay did was to reinforce my views. A recent example of that is what Kimon Friar said in Photiades' interview with him in the Translation Review.

More generally, I think my attitude toward language was shaped by what I was taught in school. In Japan--I say "in Japan," because I've never studied at any educational institution in the United States--we are told from the very beginning, and repeatedly, that different writers write differently. So, from the outset I did not believe that poems, simply because they are poems, must be translated in one particular way--say, ptiy. What is poetry to one person may not be so to another.In retrospect, that attitude was strengthened by working on twentieth-century poets first (especially by Miyazawa Kenji, who wrote in a variety of voices, as the first major poet to work on), and then turning to a group of modern poets who are quite discrete in style. My attitude might be different if I had begun by working on classical poems.

Are there any special considerations that are necessary when translating Japanese poetry?

In a translation I first count on common sense: a poem I can understand in reasonable terms should be understandable in English, especially in the kind of English I can manage. I also count on the reader's sophistication, though sophistication may be too big a word. anyone who picks up and reads a translation from the Japanese must be conscious of that fact--that unless otherwise noted, so to speak, he is reading about Japan, not France. He must, and I think he usually does, have that sentiment so memorably expressed by Frank O'Hara, buying "an ugly New York Writing to see what the poets in Ghana are doing these days."

Oyama Sumio, the surviving friend of the haiku poet Taneda Santoka (1882-1940), expressed pleasure when I told him I was translating Santoka and wanted to see my translations. So I sent them to him when they were printed in a magazine.Some time later he wrote to me, "The Japanese kane is so different from the Western bell that I couldn't help laughing when I saw your translations." Mr. Oyama isn't an English reader--he just happens to read English, but any English reader who has that attitude ought to forget about translations, at least my translations.

In such places as your essays in St. Andrews Review, "Seven Japanese Poets Since the War," and "A Lecture: Translation," you've mentioned that some Japanese poets are especially difficult to translate and that some are just about impossible. Why?

How difficult a poet is to translate depends, I think, to a large degrees on the translator's attitude toward translation, his attitude toward poetry, and his competence. If he believes, for example, that the existence in Japanese of several active pronouns for the first person singular alone makes any English translation of a Japanese poem far less effective than the original, then, naturally, most Japanese poets are difficult to translate, some impossible. If he has a preconception or an exclusive idea about poetry, a good many Japanese poets should also prove difficult, although that must hold true of poets of any other country. It's often said that a translation of a poem must read like a poem, but different people have different ideas of what a poem is or what it should be like. The translator's competence, I think, requires hardly any explanation; I would just add that I include in that word the ability to recognize what is linguistically in the original and carry it over in translation.

Having listed such generalities, it still isn't difficult to say which poets are difficult to translate. Here, let me give three poets: Takiguchi Shuzo (1903-79), Yoskioka Minoru (b. 1919), and Kato Ikuya (b. 1929). In the case of the surrealist Takiguchi, the relation of a given modifier, be it a word or a clause, to the modified is made deliberately ambiguous. Yoskioka's poetry has a similar, deliberate ambiguity, but it was while corresponding with him on his poems that I learned about his special line arrangement. In reference to the opening lines of his poem, "Bokka" (Pastorale), I asked him whether the first line is supposed to be an entity on its own, or to modify the second line, or, along with the second line, it is to be considered part of an inverted sentence:

haguruma ga obitadashiku chite yuku
kami no tenohira yori
harukana tokoro nami agaru

Wheels fall innumerably
From the palms of God,
Further, waves rise.

His reply was: "'kami no tenohira' strongly modifies the first line, but it also modifies the third line. If it's difficult to bring off this effect, you may regard it as the modifier of the first line. Incidentally, most of my poems are made in such a way that each line modifies the foregoing and following lines." The arrangement itself, as you know, resembles that of the renga, linked poem, and I gather some French poets have written in a similar manner. Since I translated enough of his poems to make a book, I'd be the last to say that particular line arrangement makes him an impossible poet to translate, but it certainly makes him difficult.

The punster Kato, to my knowledge, is one person whom I wouldn't hesitate to describe as "impossible to translate." I would not even dare to quote him in transliteration, because the possibility of more than one reading is supposed to be part of his punning. The question is: Should any attempt be made to translate him?My answer is: No. Knowing that there are poets like that in Japan as well should be adequate.

Classical Japanese poetry is extremely rhythmical, alternating waves of 5-7-5-7-7 syllables, and with alternating consonant and vowel sounds.Of course, in modern Japanese poetry, rhythm is also very important. How do you catch the rhythm and put it into English?

Hagiwara's conclusion nearly sixty years ago, that rhythm in Japanese free verse is that which emerges when you project your thought accurately, may apply here. If I manage to translate the original fairly accurately, I'd like to think I also manage to convey its rhythm. As you know, poets and scholars debated and experimented with syllabic variations around the turn of the century, but in the end none of the alternative patterns of 4, 6, 8 syllables or whatever took hold.We are still left with the good old 5- and 7-syllable patterns, and few poets in the past fifty years appear to have used them in any conspicuous manner except those writing haiku and tanka. Since I began by translating the poets of the period when the use of syllabic patterns was largely abandoned, I could, though again in retrospect, ignore the question of rhythm. By the time I began work on a large group of classical poems written in the strict 5- and 7-syllable pattern a few years ago for our anthology, my idea of translation had become more or less fixed, so I did not take a special approach in translating tanka, hokku, and other poems in syllabic patterns except in lineation.

Every poet fills his words with power, but Japanese poets have drawn on the large number of homonyms in their language to make their poetry especially intense and multi-layered. How do you handle "punning" in translation?

I have passed my negative verdict on Kato Ikuya, but I assume you are speaking of classical poetry, such as the tanka of the imperial anthology, the Kokinshu of the early tenth century. When I face a poem with a pun, I will translate the poem only if the pun is not crucial to it but merely a part of the poem. For example, I have translated Princess Shikishi's tanka, yama fukai haru tomo shiranu matsu no to ni taedae kakaru yuki no to ni taedae kakaru yuki no tama-mizu (Shinkokinshu, No. 3) "Deep in the mountains, the pine door isn't aware of spring--on it, intermittently dripping, beads of melting snow," because ignoring the pun on matsu (pine, to wait) doesn't seem to seriously impair the effect of the poem. I have recently written a review of a book criticizing its author for translating every meaning of a pun within the poem, but that is because he also gives extensive notes on each poem. I don't actually know if there's any good solution to this problem.

Whom would you put on a list of Japanese poets you don't plan to translate but think should be done?

If you're thinking of modern poets who ought to get separate volumes, the foremost among them is probably Kaneko Mitsuharu. Fortunately I hear that James Morita is going to publish a book on him soon. Another such poet is Ishihara Yoshiro (1915-77), who spent several years in Russian concentration camps. Though he is rather extensively translated, he definitely deserves a volume that will include some of his prose pieces. Several of his essays on his life in the concentration camps and after his return to Japan are spartan and moving. Nishiwaki Junzaburo (b. 1894) is another. Though I translated more than seventy of his poems, when I told him of my plan to get them published in book form, he said no, he was translating his poems into English himself. I certainly do not plan to translate any more of his work, but someone should be able to come up with new translations and persuade him to see that they are good enough to be considered separately from whatever he's doing.

When you forget about separate volumes, there must be quite a number of poets who ought to be done, particularly among the postwar poets, but I'm sure most of them are being translated at this very minute, if the recent special issues of The International Poetry Review and Loon are any indication.

What are some of the common types of mistakes you see Western translators of Japanese poetry making? What do you think causes these mistakes?

A sizable number, or probably a great majority, of translators of Japanese poetry are people like me--Japanese who translate into English with the help of English or American people, or who provide them with their own translations for transformation. The lapses in the translations done through either arrangement are, I think, mostly traceable to the original translators, including me--their inability to come up with better translations for the English translators to improve on. That situation will no doubt be corrected as competent translators who work into their native language, such as Dennis Spackman, become the majority in the field. Translators of classical Japanese poetry seem to be mostly English and American professors, and the outstanding feature of their translations is overtranslating, which I regard as a kind of error. Here, I hop Burton Watson, you, and other people who favor a leaner approach will gain more influence.

You've written that you consider Miyazawa Kenji to be the greatest poet of the 20th century. What is it that makes him so great? What do you think distinguishes him from and yet endears him to the Western world?Would you, in closing show us a poem by him that you especially like?

Miyazawa moves you to sorrow, to laugh, chuckle, marvel--he makes you live the things he describes, and he describes quite a variety of things from quite a variety of viewpoints. Takamura may have a stronger force of character, Hagiwara a deeper sensibility, but neither matches Miyazawa in the sweeping ability to absorb things observed and present them in a sympathetic light.He is closest to reality and yet most imaginative. I'm glad you've asked me to quote a poem of his. It's rather long, but here's "The Breeze Comes Filling the Valley":

from the south, and from the southwest,
the breeze comes filling the valley,
dries my shirt soaked with sweat,
cools my hot forehead and eyelids.
Stirring the field of rice stalks that have risen,
shaking the dark raindrops from each blade,
the breeze comes filling the valley.
As a result of all kinds of hardship,
the July rice, bifurcating,
foretold a fruitful autumn,
but by mid-August
twelve red daybreaks
and six days of ninety percent humidity
made the stalks weak and long,
and though they put on ears and flowers
the fierce rain yesterday
felled them one after another.
Here, in the driving sheets of rain,
a fog, cold as if mourning,
covered the fallen rice.
Having suffered all of the bad conditions,
few of them which we thought we'd have,
they showed the worst result we'd expected,
but then,
when we thought all the odds were against their rising,
because of the slight differences in seedling preparation
and in the use of superphosphate,
all the stalks are up today.
And I had expected this,
and to tell you of this recovery soon
I looked for you,
but you avoided me.
The rain grew harder
until it flooded this ground.
There was no sign of clearing.
Finally, like a crazy man
I ran out in the rain,
telephoned the weather bureau,
went from village to village, asking for you
until, hoarse,
in the terrible lightning,
I went home late at night.
But in the end I did not sleep,
And, look,
this morning the east, the golden rose, opens,
the clouds, the beacons, rise one after another,
the high voltage wires roar,
the stagnant fog runs in the distance.
The rice stalks have risen at last.
They are living things,
precision machines.
All stand erect.
At their tips, which waited patiently in the rain,
tiny white flowers glisten
and above the quiet amber puddles reflecting the sun
red dragonflies glide.
Ah, we must dance like children,
dance we must, and that is not enough.
If they fall again,
they will rise again.
If, as they have,
they can stand humidity like this,
every village is certain to get
five bushels a quarter acre.
From the horizon buried beneath a forest,
from the row of dead volcanos shining blue,
the wind comes across the rice paddies,
makes the chestnut leaves glitter.
Now, the fresh evaporation,
the transparent movement of sap.
Ah, in the middle of this plain,
in the middle of these rice paddies rustling as powerfully as if they were reeds,
we must dance, clapping our hands, like the innocent gods of the past,
dance we must, and this is not enough.

The interviewer would like to thank the Office of Graduate Studies of the University of Texas at Austin for a great, in part for this interview, which made it possible to go to New York City and meet with Mr. Sato.

Translation Review, Volume 10, 1982.

[an error occurred while processing this directive]