These poems, taken from Selected Translations 1948–1968, Asian Figures, Selected Translations 1968–1978, and Sun at Midnight, represent my attempts to make poetry in English out of poems originally written in Asian languages, over a period of more than three decades. I cannot remember when I first encountered Asian poetry, but it was in translation, of course, because I know no Asian languages. When I was still a child, I found, in the Harvard Classics, Edward FitzGerald’s version of the Rubáiyát of Omar Khayyám—the original language not, according to some definitions, strictly Asian, and the rendering, I am told, a distant approximation, but I have remained fond of it ever since, as a great piece of Victorian poetry. By the time I was sixteen or so I had found Arthur Waley’s Chinese translations, and then Pound, and was captivated by them both. Their relations to the forms and the life of the originals I will never be able to assess. But from the originals, by means and with aspirations that were, in certain respects, quite new, they made something new in English and they revealed a whole new range of possibility for poetry in English. Poetry in our language has never been the same since, and all of us are indebted to Waley and Pound whether we recognize and acknowledge it or not. Their work suggested, among other things, that the relation between translation and the original was more complicated and less definite than had often been assumed. But in fact the notion of what translation really was or could be had been undergoing change all through the nineteenth century, partly as a result of efforts to bring over into English a growing range and variety of originals. The assumptions inherent in the word “translation” had shifted radically since the early eighteenth century.
When Pope set out to translate Homer, almost everything (as it appears to us) was known beforehand. He knew who most of his immediate readers would be: they had subscribed for the translations. They, in turn, knew—or thought they knew—who Homer was, and they knew the text, in the original. Both the subscribers and the translator took it for granted that the proper form for heroic verse, in English, must be the heroic couplet. Pope’s work was expected to display the wit, elegance, and brilliance with which he could render a generally accepted notion of the Homeric poems in a familiar English verse form.
Since the eighteenth century, and especially since the beginning of modernism, more and more translations have been undertaken with the clear purpose of introducing readers (most of them, of course, unknown to the translators) to works they could not read in the original, by authors they might very well never have heard of, from cultures, traditions, and forms with which they had no acquaintance. The contrast with Pope’s situation is [End Page 95] completed by the phenomenon, which has appeared with growing frequency in the past half century, of poet-translators who do not, themselves, know the language from which they are making their versions and who must rely, for their grasp of the originals, on the knowledge and work of others.
New—or different—assumptions mean different risks. New assumptions about the meaning of the word “translation,” whether or not they are defined, imply different aspects of the basic risk of all translation, however that is conceived. Which is no risk at all, in terms of the most common cliché on the subject: that all translation is impossible. We seem to need it, just the same, insofar as we need literature at all. In our time, an individual or social literary culture without it is unthinkable. What is it that we think we need? We begin with the idea that it is the original—which means our (as scholars, potential translators, or readers) relative conception of the original. At the outset, the notion is probably not consciously involved with any thought of the available means of translation. The “original” may even figure as something that might exist in more forms than one, just as it can be...