Kojima - Written on water: five hundred poems from the Manyōshū

There is a school of thought that claims poetry cannot be translated. There is no way, states this argument, that the meaning, subtlety, depth, beauty, spirit, and essence of the original language can be transmuted into another language. The only possibility with poetry translation is approximation and compromise, and the result is little more than a wan shadow of the original.

Should the result emulate the formal conventions of the original language or should it take on the formal conventions of the new language? In other words, should a Japanese poem in translation try to approximate the sense and feeling of the original language or should it be translated into an English poem? To what degree should the translation hold its own as poetry, or should a rendering of meaning be the main aim?

With something as fluid and alive as language, and with poetry in particular, there is no simple right answer to such questions. Somehow an ideal mode or method of translation must be chosen, but anything attempted can only be partial. Even with poetry as close to contemporary English as that of, say, Rainer Maria Rilke, the translation problems are hardly small, but with a genre of poetry that follows linguistic conventions totally different from any Indo-European language, and one generated in the ancient past by a society that eludes comprehension by the West even today, the problems of translation take on entirely new nuances of the word "difficult."

Impossible? Probably. But even so, should any attempt to translate such poetry be avoided? That would mean that anyone who does not read Japanese and who is not conversant with eighth-century Japanese aristocratic society would be denied any indication of what the poetry of that period is about. Particularly, if the poetry in question is considered among mankind's great artistic legacies, should there not be some means of conveying, however partial and incomplete, an idea of what the original says?

Assuming that the answer is yes, assuming that the intrinsic worth of the Man'ysh makes its translation into English and any other European language an important contribution, how should the translation be done?

Should a scholar, who is not a poet and, with rare exceptions, does not possess the poet's gift of language, attempt to translate poetry of any kind (in the process adding to mankind s burden of footnotes)? Should a language specialista translatorwho is not a scholar attempt such translation?

How knotty the translation of this material is was made clear when a check of other English Man'ysh translations revealed that other translators' interpretations often were totally different from Mr. Kojima's and from each other's, some poems even being given opposite meanings. This should not be surprising, perhaps, because, as one commentator has pointed out, by the late Heian period, the language in which the Man'ysh was written had already become partially incomprehensible, and the following 800 years or so have complicated the task of interpreting the original poetry.

Without pretension and without scholar's pomp and posing, Mr. Kojima selected 500 poems out of the total of about 4, 500 and has given us a charming book. The translator of this little volume undertook a task rather like skiing down Mt. Blancthere are more reasons for not doing it than for doing it. Ignoring all the obstacles and the arguments for not attempting to translate the poetry of a society that might as well be Martian, he jumped into the task with verve and caring and a lifetime's experience as a translator of Japanese into English. This editor never met the translator, for Mr. Kojima passed away some years before this publication project was initiated. As a result, queries and problems could not be discussed, but had to be handled as English-language problems, and the few translation queries that occurred were given to another linguist to solve. Much gratitude goes to Ruth McCreery for her patient and supportive help on many levels.