Language Learning and Translation

I had a glance at Eco's Experiences in Translation today. He talks about the problems of translation and the difference between literal translation and capturing in another language what the author was trying to convey. He posits an artificial language capable of expressing any idea. The beauty of such a language would be that you could test your translation by translating both your translated sentence and the original into this third language. If they translated the same, you'd know you had a good translation.

Unfortunately, I don't think it's possible. Eco, himself, in his Open Work, talked about writing as an interaction between text and reader and, in my experience, part of what make fiction - and even factual narrative - work is that the reader brings enough of himself to the text. They say a picture is worth a thousand words, and yet in a single sentence we can create an image and lay the foundations for an imagined world. Take, since it is handy, the opening of Borges' "The Shape of the Sword":

A spiteful scar crossed his face: an ash-colored and nearly perfect arc that creased his temple at one tip and his cheek at the other.

You'll note, of course, that this passage is translated from Borges' Spanish. But two things stand out for me: 1) Twenty-five words are more than enough for you form a picture of this guy. 2) Whether in English or Spanish, the picture you're going to get has less to do with these 25 words than with your own ideas about what sort of person has a scar like this. So a perfect translation, in a sense, would not cause someone from Brooklyn to form the same mental image as someone from Buenos Aires. Rather, it would cause the guy from Brooklyn to get exactly the same image from the English that he would have formed were he to have lived an identical life except for speaking Spanish as his native language. (Which is already a paradox.)

In learning a foreign language, there's an urge to translate, as though it's only in getting something back into your own tongue that you know that you really understood it. In the end, however, it doesn't work that way. What you really need to get to is a place where those 25 words give you your own image based on your own experience of living life through the language.

The other day, Josh was writing about global understanding vs. individual words. I sympathize, because I often find, especially with Spanish and Italian, that I get the gist but that I could neither translate nor diagram a given sentence from a passage which I understood globally. I think long term fluency requires that you be able to dissect a passage about as well as a native speaker could. But I'm not sure whether this is something to be consciously attempted (though Lord knows it's what I had to do for French in grad school) or something you just need to let yourself develop through long-term exposure or at least wait to develop until you can use the same tools native speakers use. I do think there is a question you can use, though, to make the decision for yourself:

Do you want to get that individual word understanding because its lack is actively impeding your ability to use the language, or because at some level you are trying to prop up your confidence of understanding through subconsciously translating? If you're really out of your depth with a text in another language, you may well need to use supporting tools to make sure you actually understood. Or you may need to work on simpler texts first. But if you understand what's going on at the end of a text, just not enough to re-create it yourself yet, what you probably need is more exposure to comprehensible input, either with new texts or by running through the same texts until their patterns echo familiarly in your mind, whether you could diagram them or not.