Is translation even possible?

“What a silly question!” you reply. “We translate sentences in high-school language classes all the time!”

Do you? You might not do that at all (perhaps you took a total immersion class) but let that pass. You probably did at least some translating. That might have often been difficult, but certainly not impossible.

I’ve frequently written about problems of translation in this blog, most specifically in a post that began “Translation is impossible.” So this post is something of a follow-up to that one. It was specifically inspired by last week’s surprisingly controversial episode of the wonderful Because Language podcast. Start at 24:39 and listen to the 27-minute interview with Gary Lupyan, which extends to 51:31.

Did you do that? If you didn’t have time to follow my advice, here are a few specific observations and questions from it:

  • When translating, does culture matter? There are arguments on both sides.
  • For example, although “comrade” is literally the correct translation of the Russian word “товарищ,” it isn’t the correct translation in practical terms since the usage is so different.
  • Lupyan takes a computational-linguistics point of view. Is that useful? As one Twitter user asked, “Is computational linguistics Columbusing anthropology?”
  • Should linguists listen to the ill-informed views of anthropologists, economists, philosophers, psychologists, computer scientists, etc.?
  • We can learn something about language from a big-data approach, where we quantify our results. But this methodology annoys those who follow the traditional field-work methods of studying one language at a time in great depth.
  • We still need the expertise of multilingual translators.
  • Snarky comments on Reddit (like “computer scientists discover language” and “what do they know about language?”) don’t help!

How, if at all, does this relate to my post about The Ingenious Language, where I commented on the problems of translating a book about Greek that was originally written in Italian?

Maybe translation is less possible than you or I had thought.


Categories: Linguistics