Monthly Archives: October 2012

David Bellos on the Future of Translation

By | Translations | No Comments

David Bellos, a prominent literary translator recently did an interview with Big Think.  In it he presented a compelling vision of translation in the future.  Check out the video from the article below.

Basically he argues that as machine translation becomes more prevalent people will communicate more with other cultures even more and that eventually this will just become expected.  As it becomes more expected demand for human translation will also rise.  In other words the growth in machine translation will lead to growth in communication in general which we lead to growth in human translation.

He also mentions the oft forgotten fact that machine translation requires human translators to create the language pairs that the machines base their translations off.  As language is an ever evolving phenomenon we will always require translators to provide updated language pairs, even if machine translation becomes good enough to replace human translators.

Vendler Gets to the Challenge of Literary Translation

By | Translations | No Comments

Speaking on a new translation of Dante’s early work the “Vita Nuova,” Hellen Vendler concisely summed up the challenge of translating Dante’s literary works.  But her words are equally applicable to translation in general.

“Although Frisardi’s rendition of the prose of the Vita Nuova is graceful and readable, his resolve to duplicate Dante’s verse in “contemporary American English” sits ill with the archaism of medieval manners and sentiment in Dante’s fiction. In fact, Frisardi cannot really remain within his desired idiom: in the United States we do not say lady, we say woman; we do not say flee, we say run away. One cannot imagine a native American speaker referring to his heart’s demolished core. And yet to offer the poems in a prose translation, as some have done, forfeits the very form of the Vita Nuova—the back-and-forth, prose-to-verse, retrospection to instantaneity—which establishes the rhythm of the whole: how life feels when you are living it (the poem) and how, looking back, you see it really was (the prose).”

This really gets to heart of the challenge of translation, preserving the tone of the source text as much as possible while maximizing the effect of the new translation.  This is something only a trained linguist can do for regular texts and only an expert translator can do for literature.  These experts will always exist, no matter how far machine translation advances.

Read the rest of her review of the new translation here.

One Prediction of the Future of Translation

By | Translations | No Comments

Rob Vandenburg recently wrote a post over at Wired Cloudline on the challenge facing online business we’ve previously discussed.  That is that much of the web is English, while many web users are not.  If business wish to reach and effectively market to these new consumers then they must speak to them in their own language.

Vandenburg’s post, is well worth reading, as it effectively breaks down this issue, which is much larger than many people realize.  The solution he offers is one which the industry has already adopted to a smaller extent.  He says that the problem with machine translation now is that the translation memories(TM’s) used are so small, and cannot cover the broad spectrum of language they have to translate.  Therefore as the size of TM’s available to machine translators grow, the quality will also grow.  But, this will never fully replace the human translator as an editor of the machine translated materials, as humans will have to look out for colloquialisms and idioms.

This is generally how the industry operates now, though on a smaller scale.  Translators use their own TM’s which are usually client specific, and this accelerates the translation process.  TM’s do automatically translate some of the materials, but Vandenburg overestimates how much they can do.  The heavy lifting is still done by the human, and even as the TM grows, the translator experiences diminishing marginal returns as context determines word choice.  The context of the translated material is far more important than any idiomatic words, and takes far more time to translate.  Not to mention the fact that each automatically translated phrase must be approved by a human.

The key thing that Vandenburg misses is that as a TM grows in the broad sense that Vandenburg is advocating, the quality of the automatic translation that it provides actually decreases.  This is due to issues with synonyms and technical language that must be tailored for each industry or even client.  If we were to throw all of these technical TM’s into on great TM we would receive more confused translations as the language would jump around between different disciplines.

While Vandenburg is right that increasing TM size for machine translators would improve the quality of the translations produced, he overestimates the extent to which this can replace human translators.  Improvements in machine translation will at best augment the human translation process, but not to the degree Vandenburg says, where humans are at most minor editors.