Machines’ capability to translate natural language is progressing, but there still is room for improvement. Will machine translators always be inferior to human ones?
My answer is no, of course not. That position is backed up by Alec Ross, a former senior adviser for innovation to the U.S. Secretary of State. Writing Jan. 29 in The Wall Street Journal, he suggests that in 10 years, you’ll be able to wear a small earpiece that will provide real-time simultaneous translation—in a voice sounding not like Apple’s Siri but like the foreign-language speaker with whom you are conversing, thanks to advances in bioacoustic engineering.
And conversations won’t be limited to two people. You can host an eight-person dinner party with each guest speaking a different language and all able to understand each other. On a larger scale, he writes, investors—currently hesitant to contend with the 850 languages spoken in Papua New Guinea—can rely on machine translation to help do business in that resource-rich country.
Indeed, machine translators could become so effective there will be no need to study foreign languages in school. Writing March 7 in The New Yorker, Rebecca Mead quotes educator Max Ventilla, founder of the AltSchool in Brooklyn, as questioning the value of studying a foreign language as a means of communication in the era of live-translation apps. As an aside, Mead adds that AltSchool does seem to require fluency in “the jargon of Silicon Valley—English 2.0.” Perhaps there will be an app for that.
On the other hand, David Arbesú, an assistant professor of Spanish at the University of South Florida, sees definite limits to machine translation. Writing March 28 in The Conversation, he contends that “… language contains nuances that are impossible for computers ever to learn how to interpret.”
Arbesú suggests that engineers working on machine translation might try to “log all the rules by which languages work.” But, he contends, languages don’t work that way. However, what he overlooks is that with machine learning, computers don’t work that way either. A machine translator needn’t begin with algorithms—it can begin with data and derive its own algorithms.
Microsoft Research, for example, uses a data-driven approach to machine translation that makes it unnecessary to write explicit rules. With the data-driven approach, machines learn how to translate using human-translated parallel texts. The company offers its Microsoft Translator portal on Bing and recently released a new version of the Microsoft Translator API. The API, the company says, “… is the first end-to-end speech translation solution optimized for real-life conversations (vs. simple human to machine commands)….”
Of course, if it becomes pointless to learn foreign languages, students can always focus on learning machine languages. As Arbesú points out, the Florida State Senate had proposed a bill that high schools offer foreign-language instruction credits for courses in computer coding and that the state college system must recognize those credits as meeting foreign-language credit requirements. Parents and students are warned, however, that colleges outside the state system may not accept the coding credits in lieu of foreign-language study requirements.
It looks as if the Florida bill will not become law, and that’s fine. It’s good to learn computer code, and it’s good to learn a foreign language—there’s no need to confuse the two. Computer prowess at natural languages won’t negate the benefits of people speaking multiple languages. MIT Professor Lawrence Susskind, who is teaching an online course in negotiation at MIT, says that negotiation requires trust, and one way to establish trust is to travel to meet your counterpart face-to-face despite the existence of email and Skype. Learning your partner’s language will always be a further step in building trust.
