“To what extent can a translation be referentially ‘unfaithful’?”
As a history major, translation for me has always been a simple, straightforward affair. Since history writing is mostly for an academic audience, substance is valued over style, and so the translator strives to make as few changes to the source material as is possible, though alterations of grammar and syntax have to be made in order for the translation to remain intelligible. The purpose of this approach is to make sure the translation is as faithful and unbiased (and hopefully uncontroversial) as possible.
Since the beginning of this course, however, I have had occasion to translate works for which style is just as, if not more important than substance, forcing me along the way to rethink my ideas regarding “faithfulness” to the source material. Especially in fiction, the words that make up a sentence, or a line of a poem, can mean far more than the sum of their parts. Take Dante, and his Divine Comedy, for example: the Divine Comedy is the most important work of Italian literature not just because of the story it tells, but because of its meter, rhyme scheme, word choice, etc. Most translations of the Divine Comedy into English, however, abandon the original’s meter and rhyme scheme in order to preserve its narrative. Though this is faithful to the narrative of the original, it is unfaithful in the sense that the reader, once finished with such a translation, has no appreciation of Dante’s poetic abilities, for most of the original’s features have been stripped out.
A better, more specific example of this is Ungaretti’s poem Mattina. The poem, literally translated, reads “I illuminate myself / of the immense.” This translation, though it conveys the poem’s meaning, completely fails to communicate the poem’s value and intent which, in this instance, are far more important than the poem’s meaning. The entire value of Mattina lies in the perfect way all of its sounds blend together. The literal translation—which most would probably assume on principle is the most “faithful”—is therefore extremely unfaithful to the source material. This is a perfect example of how, by not making alterations, and by sticking to the exact letter of what we are translating, we actually create something less faithful than what we would have created had we made more alterations. A more faithful translation would be something with a lot of repeated sounds, with words which flow together perfectly, so much so that they leave an impression on the reader. A translation which achieves this, even if it has nothing to do with the morning, or with illumination, would be more faithful than the literal translation given above.
Though most assume faithfulness in translation is a simple concept—which sometimes, as in academic writing, it can be—the reality is often much more muddy, especially when we try to be “faithful” in communicating both the literal meaning of the source material, and the more subtle senses and feelings that the source material provokes within the reader. A perfect example of the paradox of faithfulness is Ungaretti’s poem Mattina, for which a literal translation, which most would assume is by default the most faithful, is actually the least faithful translation option available to us. In this case, a more ambitious translation would be the more faithful, because the literal meaning of the words is unimportant compared to the feelings the words give the reader.