Translation technologies has developed rapidly in the last few decades, causing upheavals in the industry as a whole. Where it stands now, human translations are still far superior to machine translations. But will that always be true?

Machine Translations are Catching Up Fast

It’s impossible to project at this point in time what level of quality may be achieved by machine translations in the foreseeable future. At the moment, machine translation can produce texts that are readable, sometimes understandable, and they are getting better. Especially as developers streamline computers’ ability to draw on databases of human translations, there may be a point in the future when machine translations will become flawless.

What Do Humans Have that Machines Never Will?

“Flawless,” however, does not always encapsulate what we are looking for in a translation.

This is especially clear in the world of poetry. The message of a poem is carried not only by the literal meaning of its words, but also by their allusions and connotations, their cadence and rhythm. This means that a literal translation of a poem can never carry its full meaning. Translators of poetry basically rewrite the poem in another language, sometimes straying very far from the literal meaning of the words to convey what they understand as the underlying meaning of the piece.

This kind of nuanced understanding requires sentience. That is the one thing humans have that computers (at least outside the realm of dystopian sci-fi) never will.

Another thing computers cannot be trained to do is recognize cultural differences and potential misunderstandings. Even if a translation has flawless grammar and syntax, it needs to take the sensibilities and nuances of the reader’s culture into account. It is basically impossible to create an algorithm that will account for the cultural nuances of every individual. It’s just too varied and unpredictable.

It’s Not Really Humans v. Computers

Machine translation is not the only type of translation technology that has been developing in recent years. Various computerized translation tools have become available that help translators work faster, more effectively, and more thoroughly.

Perhaps the most effective and efficient model of translation in the future will be a sort of hybrid.

We already see this today to some degree: most translators work with some kind of assistive software. And companies that choose to use machine translation usually hire a human reviewer to go over the text, smooth it out, and make it readable.

It is reasonable to predict that technology will develop to close in the gaps between these two models. The gold standard of translation in the future may be some kind of computer-assisted human translation—or, of you will, human-assisted computer translation.

So, the answer seems to be that no, human translators will never be completely replaced. The challenge for translators will be to stay on top of the changing tides, and be the first to embrace new technology while adapting business models to the new realities of the future.