@inproceedings{oai:kokubunken.repo.nii.ac.jp:00002616,
 author = {Carter, Steven D and CARTER, Steven D},
 book = {国際日本文学研究集会会議録, PROCEEDINGS OF INTERNATIONAL CONFERENCE ON JAPANESE LITERATURE},
 issue = {23},
 month = {Mar},
 note = {pdf, Many problems face anyone attempting to translate poetry from one language into another. Some of these specific to the languages in question. When it comes to translating traditional Japanese poetry into English, for instance, the translator nearly always must deal with thorny problems of pronoun reference and the challenges of syntactic compression, to name just two.
There are also some more universal problems. One of these is just how far to go in preserving the "otherness" of the text. Obviously, the appeal of reading the poetry of another culture is partly in its otherness, which one therefore wants to preserve as much as possible. But deciding just how to do so is no easy task.
I have confronted this problem myself over the past twenty years or so in attempting to introduce the poetry of late medieval Japan to an English speaking audience. My first experiences involved the genre of renga, or linked verse; more recently I have been concentrating on uta, the classical thirty-one syllable form that was the major poetic genre of the Japanese poetic tradition from the time of Kokinshu until 1600 or so. At first I thought that translating uta would be an easier task. The verse form was not as short, there was no bothersome "linking" to deal with, and most importantly, there was no need to deal with the tension between completeness and incompleteness that is the essence of linked verse.
Needless to say, I was wrong in my assumptions. For in translating medieval uta I learned quickly that the genre poses its own thorny problems, and that--as common sense should have told me--some of the traits of linked verse as a genre were shared by uta, another fact that I should have expected of genres in the same historical situation. Indeed, I discovered that a certain tension between completeness and incompleteness-- in the form of an almost explicit intertextuality--may be even more conspicuous in late medieval uta than it is in linked verse.
One can talk about this tension in various ways. For this presentation, I will attempt to do so through a discussion of the practice of daiei, or composition on pre-established topics. Nearly all of the uta of the late medieval age were in fact composed on such topics. Indeed, so central was the practice that poets of the time seem not to have reflected much on the assumptions behind it. For modem readers, however, the practice seems profoundly unusual, presenting us with a potent revelation of the otherness of the form. But can the significance of daiei be communicated through translation? It will be my task to explore that topic, using some poems of Tonna (1289-1372), a true master of daiei, as points of reference.},
 pages = {1--22},
 publisher = {国文学研究資料館},
 title = {講演　題詠の翻訳　―頓阿の歌をめぐって―},
 year = {2000},
 yomi = {カーター, スティーブン}
}