Issues of Translation
There are nowadays many more translations available of poetry from other languages into English, than have been available before. We exist in a time when inter-cultural communication is at a high point of interest, so the demand for new translations is ongoing. So, there are often many translation to choose from.
Given that happy fact, I think the translations that one prefers, beyond the quest for simple accuracy in translation, reveal far more about one's one own personality than one might imagine.
I think the translator's aesthetic matching the reader's expectations and personal aesthetic have a lot to do with this. It remains a subjective terrain, because even the choice to read an annotated literal transliteration is a choice some personality types will immediately go for, while it will baffle and bore others. One reason we keep getting so many news translations is because people get impatient and/or dissatisfied with existing ones, and want to do a better job than anyone else has done before, and also want to express what is being translated through themselves. None of these are bad or wrong motivations for making a new translation.
Ultimately, if you have a very specific idea of what a translation should be and do, you'd better do it yourself, because nothing else is likely to satisfy you. And this too reveals something about oneself, as well.
My current favorite translations of the Tao Te Ching are: 1. the Gia-Fu Feng/Jane English translation, an illustrated classic wherein the photos do as much if not more to get the ideas across as do the words, and part of the imagery is Feng's fresh calligraphy of the Chinese original, as well as English's remarkable photos; and 2. the Ursula LeGuin translation, in which she admits she knows no Chinese, but studied every other translation available, went over the original Chinese character by character in a scholarly manner, then proceeded to write what came forward from her poet's mind; the result is limpid, easy to grasp, as fluid as the water the Tao mentions flowing downhill, and austerely beautiful.
The Corman and Susumu version of Basho's Oko no hosomichi is a pretty good example of balanced notes; each segment of the original is presented on a spread, with artwork, and the original Japanese characters on the verso and the English on the recto. But for sheer page-turning readability, I personally don't think anyone has done better (so far) than Sam Hamill's translations of Basho; they pull me in, make me see the places Basho is describing, and make want me to keep reading.
I often prefer the detailed-notes translations; but I like the notes to be at the end of the book, where they don't interfere with the reading of the poetry. I like to read the footnotes, but I don't like them to interrupt the poetic experience.
There are two main camps in the argument over translation: those that prefer translations to retain as much as possible of the original, and those that prefer to create a new aesthetic experience in the new language; the former tend to be very literal in translation, and the latter tend to create "versions based on" the original, rather than a perfect emulation.
There is something to be said for this latter viewpoint: making a new work in the new language that is readable and able to stand on its own, as poetry. Robert Hass and Stephen Mitchell both make this point, with regard to translating Rilke: if the English version also doesn't work as a poem, on its own, you have not done the original any service, since, while you may have brought the words over accurately and literally, if you sacrificed the spirit of the poem mid-way, so what? Who wants to read a dry, academic, literal translation in English of something that in its original language was inspirational and soaring, renowned for its poetic spirit?
I have seen, again and again, translators who are bilingual argue for a certain stance in translation that can only be called non-translation: no attempt to accomodate the reader in the new language is made, and precision is emphasized over tone and spirit. There also tends to be something a superior attitude about credentials among some in this camp, as they are indeed bilingual and you're not; what this ignores is that languages are dynamic, and even native speakers don't always agree on meaning, interpretation, or aesthetics.
I own numerous translations of the Tao Te Ching and Oko no hosomichi (since I've already mentioned these two examples) that fall into this mode: and none of them make me want to turn the page and keep reading. They have some usefulness as resources for looking up details in the original, and providing minute nuances of interpretations. But they're barely readable as literature. Translation needs to be literature, in the new language as well as the old. Any college grad with a good dictionary and some practice can do a basic literal translation; but most of these literal translations of poems will not become poems in the new language. At least, not in my experience.
So, perhaps it defaults to personal preference. Having a variety of options is a good thing, because people are different, with different preferences. But that's also why I say that your aesthetic choices in translation reveal more about you than you might think, and possibly reveal more about you than they do about the original text.
For example, it has been argued that one reason to do literal translations is a quest for an experience of literature (description of experience) that is outside the box of one's birth language and culture. The contact with foreignness is designed to break us out of the habits of thinking we fall into, all unquestioned.
But this quest for "foreignness," in this context, may be indistinguishable from exoticism (the frisson of meeting the Other) or, perhaps, Orientalism. (cf. Edward Said, Orientalism; and David Maybury-Lewis, The Shock of the Other) At it's worst, it's cultural imperialism; at its most banal, it's postmodern tourism.
It strikes me that what this quest is seeking, is something far Other than one's own, familiar life. It also reminds me that the reason translation is possible at all is that, at root level, we are all human. We share more experiences in common with each other, even across cultural differences, than we don't share: we share life, love, sorrow, the pains of living and the joys of growing and becoming fully human, fully conscious and alive. In other words, all those things that are the topics of poetry. This leads me to think of Jung's concept of the collective unconscious, wherein we all share a pool of archetypes and patterns that rise from within, no matter who we are, no matter where or when we are born. They are our birthright as humans; as is language itself.
If, on the other hand, one is looking for those slight differences between us, in our infinite variety—the elegant way of describing a particular emotion or state of being that might be more refined in one language than another, due to cultural history and circumstances, the geomorphology of metaphor (images ties to places), the genius loci—well, then you might find what you want. Maybe. With luck and fortitude. (I've studied Japanese culture and artistic products, off and on, when I wasn't studying, say, Navajo, for 35 years; yet I'm still a complete beginner.)
Going about translation in a literal way is still never going to be telepathic; there will still be interpretations and grey areas that will obscure and frustrate the desire for total transliterated clarity. The problem with studying a language (and it takes a few years to grow a new persona that thinks in the second or third language) is that the more fluent you become in it, the more you realize that the new language is just as ambiguous, open to interpretation, and labile as your birth language. it's built in to language(s) to be imprecise, because human nature is imprecise and messy. (Attempts to build precise artificial languages notwithstanding; I could point out that even machine and computer languages are open to personal style, and the qualities of elegance and sloppiness. The problem is that humans still write the code. Code very much reflects the personality of the writer. Engineering is still as much an art as it is a science.)
So, if the translator's quest is a quest for absolute meaning and total clarity, we're doomed to repetitive frustration. Such things don't exist—especially, one might add, in poetry. Great poetry across different languages has more in common with great poetry in other languages, than does great poetry with lesser poetry in the same language, or even by the same writer. One of the things that make great poems so memorable is that they have rich layers of meaning and interpretation embedded in them, so one might return to them again and again and still be refreshed by their waters. Taking that into a new language does require that an artist does it, an artist capable of merging, however imperfectly, the two conceptual worlds of the disparate languages, and still make the result into something readable. The best translators of poetry are themselves poets.
For example: have you ever encountered Rilke's late poems written in French? Towards the end of his life, he began writing in French, opening up whole new worlds in his poetry. The best English version of these, in my opinion, is the translation by Arthur Poulin. At one point in my life, I was fluent in French, although I haven't kept it up; but the Poulin translations, published en face, with the French verso and the English recto, bring over both the sense and the music. Poulin has also translated much more of Rilke, but this is where he shines. For the rest of Rilke, in my opinion no has done better, so far, than Stephen Mitchell—for that same reason, he brings across the sense as well as the music. His translations are less slavishly literal than some; but few deliver as rich an experience of Rilke, in English.
At one point in college I had studied enough German to develop the hubris of thinking I could translate one or two favorite Rilke poems myself. I discovered that I could certainly read him in the original, but there was so much there, so many layers and richness and sinewy power, that, yes indeed, I was reduced to interpreting and choosing between possible meanings—and so I ended up with only adequate translations that never fully satisfied me. I am grateful for the learning experience of making the attempt, nonetheless. Perhaps making the attempt, and failing, is something every poet ought to do. It's a good way to learn about some of our limits.
There is no perfect mind to mind translation, without the use of telepathy. Making it as good as it can be, and as musical as a poem can be, even in the new language±that's where the artistry comes in.