Evaluating Translation in Language Learning

By Dr. Katharine B. Nielson –

A couple of months ago, I pointed out that translation is an ineffective methodology for language teaching. That piece was then published on the ELTJam site, where it has generated a lively discussion on the merits of pedagogic translation as well as some more colorful exchanges–it’s been a long time since someone’s told me that he thinks I “should know better.” Rather than respond to those comments there, I thought I would offer up a few points of clarification.

First, the Grammar-Translation method of language teaching has been out of favor for a long time. This summary from BYU’s Humanities Research Center documents the history of this method in a succinct and accessible way. Essentially, the focus on understanding rules and lack of practice with any useful tasks is not an effective, efficient, or pleasant approach to teaching someone a language.

That being said, language learners are going to translate things anyway. We can’t help but use our first languages as we learn our second languages, and using Google Translate to figure out the meaning of a confusing sentence or difficult-to-remember vocabulary word is often an effective strategy for a someone picking up a second or third language. But that is not the same thing as attending a class or using software where the focus is on learning through translation.

Nor is it the same thing as using learners’ first languages in the second language classroom. One commenter claimed that I had ignored a large body of evidence in favor of pedagogic translation, citing this article on “own language” use in English classrooms. It shouldn’t come as a surprise that when teachers and students have the luxury of sharing a common language, it inevitably finds its way into the classroom, and it can even help avoid unnecessary confusion and explain nuance. Obviously, though, this isn’t the same thing as using translation as a pedagogic approach.

But the point of my original piece wasn’t that we should avoid all use of learners’ mother tongues in the second language classroom or that we should prevent language students from consulting dictionaries. It was that translation is not, generally speaking, an effective methodology for teaching languages. That’s because translation is itself an art, not a means to an end. And while there has been recent research calling for more synergy between the fields of translation studies and language teaching, there is absolutely no evidence that asking people to learn through translation is the right approach.

We know that learning a language is learning a skill, and, like all other skills, instruction works best when it’s personalized, relevant, and meaningful. So, if a language learner has a real-world need to translate or wants to become a professional interpreter, then, yes, of course, he or she would benefit from lessons that involve translation. Decades of empirical research have shown us that the most successful approaches to language instruction involve preparing learners to do real things in the languages they are studying. And as most language learners aren’t trying to become translators, we should focus on finding out what they want to be able to do and teaching them how to do it, not wasting their time with irrelevant, inefficient, and outdated exercises.

Katie is Voxy’s Chief Education Officer, which means she leads the teams ensuring that learners are getting the most efficient and effective educational experience possible.  She has a PhD in Second Language Acquisition and years of experience teaching languages, building language courses, and evaluating the effectiveness of language training as a research scientist.  She lectures and writes about all things related to language learning and educational technology.