What may be gained from mistranslations of Japanese haiku?
Translations into Western languages of Japanese haiku have been done for well over a century. The shortness of these poems with their seemingly simple structure has made the form attractive to translators with little or no knowledge of the original language, and in both old and newer publications of translated haiku, mistranslations abound. In translation studies, the focus is usually on the strategies followed to reach a, in some sense, successful translation. In this paper, however, I will investigate a few examples, in English and Swedish, that may rather be regarded as failures and consider why these have turned out in this way. Following George Steiner, I will suggest that translation is a hermeneutic process that is open ended. Mistranslation will thus not be understood as a complete failure, but rather as a first tentative step towards understanding. In that way we may analyze the structural and contextual aspects of the (mis)translation and show how adjustments of these may help us to get beyond what we may regard as common sense and get closer to the very different world of the poem. This may also help us to reflect over the way in which conventions and the cultural context function to define the borderline between understanding and misunderstanding.In addition I will discuss the common theory in haiku translation, maintained by David Barnhill, Steven D. Carter and others, insisting on that the words of a poem, even when translated, should be rendered in the same order as in the original. It will be shown how this theory often lead to what may be called "pidgin translations" which may have influenced the development of Western haiku into a poem that is slightly different from the form practiced in Japan.