I use machine translation to communicate with a PhD student in another country about her academic work in our field of translation studies. She is researching a topic I know well and have written several articles on, and we have had a kind of e-mail collaboration going on for several months now.

The main language of our discussions is English, but there are several other languages involved. My command of English is more of a passive one, I am of course much better with my native German. Her English seems to be better than mine, though her research work is done in French and her native language is Chinese.

It usually works like this: She writes me mails in English. My English is good enough that I can read them directly. I want my replies to be well thought-out so that they are useful to her, so I write and edit them in German until I am satisfied. I then put the German text into Google Translate and translate it into English. I still edit the English a bit, most often to correct mistranslations of keywords from our specific field. When it’s all ready, I copy/paste the English into the mail.

I actually often include the original German text too. Sometimes I do that because I’m not 100% sure the English version came out saying what I wanted to say or if it will be clear enough. I figure that if I include the original, then she can compare the English with the German, or try machine-translating the original German into a language she knows better, like French or Chinese. Or she may even have access to a different MT tool that gives her better results.

Strictly speaking, we wouldn’t have to do this through machine translation because I do know some English. But it would take so long for me to produce English from scratch! I would sit for hours looking up individual words. I would definitely end up writing shorter and simpler messages, plus I would write less often. I’ve decided using machine translation is the fastest and best method.

Actually this is not the first time I’ve used this kind of solution. 10 years ago I had some collaboration with an academic in Spain and we used it then. We each visited the other’s university and we spoke English when we were together. But for all the communications needed to plan and organizing these visits, we used Babel Fish to translate between German and Spanish. That worked well. We once had a slight misunderstanding that came from the translation of one word, but it was soon cleared up because the context made it clear that it didn’t mean what he thought at first. It can happen that way sometimes – the context corrects things.

Of course the texts you translate in Google are not ready for any kind of publication. That is clear. But for this purpose, it is a valuable tool that saves me time and makes things easier.

Having said that, I’ll add that on the other hand, I find it interesting that people seem to want to believe that language is a system that can be mechanically produced. So much money – millions – has gone into researching these systems. Much more money than they’ve used for more sensible things. I guess it’s just the way people think; they don’t remember that words are only signs that represent mental concepts. The full meaning of words comes only through their relation to a specific situation and context.