The Relationship between Humans and Machines

February 5th, 2013

Humans and machines have a long and complicated history.

humans and machines


 

This isn’t uncommon. New technology is often disdained at first, until adoption hits a tipping point an infrastructure springs up to support it. In the translation services world, we’ve seen this happening in slow motion in regards to machine translation.

In the Beginning There was Laughter

People have been trying to make machines perform translation work for about as long as there have been machines. The moment we had working modern computers in the late 1940s and 1950s, scientists and linguists began trying to program them to perform document translation work. This has over the years proven to be much more difficult than initially thought, and even in the modern day the best machine translation tools we have are mediocre, at best, when measured against human work in the field. A pervasive attitude of disdain towards machine translation technology took root in the translation services world, and for many years the common wisdom passed from one translation professional to another was always to never, ever use a computer in your translation work unless you wanted to look foolish.

While the technology has improved, the most important change when it comes to using machines in translation work is the attitude and the thinking surrounding it.

Complement, Not Replace

In the beginnings of work on machine translation solutions the goal was to have a fully automatic translation tool, something you could feed a Russian novel into and get a complete English translation from the other end in moments. A worthy goal, never achieved. Over the years, though, the thinking has changed. Today you’ll find machine translation tools taken much more seriously by translation professionals, but the new watchword is complement.

Instead of trying to build a tool that would replace human translators, the new thinking is that modern technology can give those human translators a lot of assistance. It already does. The fact remains that in order to produce a high quality translation of just about any document of any kind of complexity, you must have a human being who is fluent in both the languages and the cultures involved – something still beyond even the most complex artificial intelligence out there. But those humans can – and do – use computer tools that can both suggest the right vocabulary but also inform on cultural aspects of a translation. Today, machines complement humans in translation work, they don’t replace them!

You might also like: