Multilingual communication in a digital world

Posts tagged ‘language services’

Interpreting versus Translation


Interpreter? Translator? What’s the difference? Indeed, some people, even in the language services field, use these terms interchangeably. And there are interpreters who translate as well as translators who also interpret. However, there’s a big difference between interpreting and translation, although they both have to do with the act of translation – that is, rendering meaning from one language to another.


The interpreter’s task is predominantly to listen to utterances, to speech, in one language, and to say the same thing in a different language – or sign – with very little lag time. In fact, a great interpreter makes it appear that there is no lag time. One of my mentors described being an interpreter as thinking of oneself as a large ear and mouth, both ideally working at the same time, actively using both languages in his/her language pair, in a continuous, unbroken cycle. The interpreter neither adds nor subtracts, relaying everything that is said accurately.


Except for simultaneous interpreters, who work from a booth at the back of a conference room (or, like at the United Nations, from a soundproof room with a window through which the delegate floor can be viewed) interpreters are usually found right at the front, next to the speaker, or in the middle of a group. Interpreters work under pressure, with only seconds to think of the right words. They have to work at whatever pace of the proceedings, without being able to consult dictionaries or other reference books. To do so would mean stopping the proceedings, making everyone wait; this is completely unprofessional, because the whole purpose of hiring an interpreter is to remove communications barriers, not add to them by adding time or other difficulties. 


The translator’s task, on the other hand, is predominantly to render written words from one language to another. From the point of view of an interpreter, a translator has the luxury of time (usually mere hours, though, as translators also tend to work with tight deadlines) to think of or to research the perfect match, to go back and revise, edit and correct. Extending my mentor’s analogy, a translator could think of him- or herself as large eyes and fingers (for typing and for rifling through dictionary pages in the quest for the perfect term). Translators also have access to assistance from machines, in the form of computer aided translation (CAT) tools and software programs.


Translators must be good writers in the language that they work into, or the target language. However, they do not necessarily have to have productive abilities in the language they translate from, the source language. In the interpreting realm, only simultaneous, or conference, interpreters have the luxury of working this way, from a source language that they only need to understand (that is, they don’t need to be able to speak it) into their productive language (that is, the language in which they can think and speak, usually their native language). But few interpreters have the luxury of starting out as conference interpreters; this is the top end of the profession and usually achieved after gaining much experience as a court or other interpreter, using all modes and at least two productive languages.





Why Machine Translation Isn’t Enough

English: Basic sketch of brain areas involved ...

English: Basic sketch of brain areas involved in language. Author: Reid Offringa creation date: 1/9/06 (Photo credit: Wikipedia)

Technology is definitely a friend to professional translators. Many of us use on-line glossaries and google academic searches for linguistic and industry research; it’s fast and easy and extremely useful. But many times Machine Translation (MT) tools simply slow me down. I can’t speak for my colleagues in the translation/language services profession, but my brain can process more general terms in context much more quickly than any existing MT tool; I’ve tested it. The human brain is built to store much more data and to be sensitive to context, which is everything for meaning, whereas MT tools are great at recognizing the lexical item and regurgitating an analogous one, but not necessarily the correct one for the context. (Although, Google Translate is pretty good at chunking

Illustration of bad machine translation from E...

Illustration of bad machine translation from English to Swedish, using Shakespeare and a robot (Photo credit: Wikipedia)

but, again, read carefully! If you are not fully bilingual you may not catch the subtle shift in meaning!)

Penny wise but pound foolish

The upshot of all this is that, more often than not, MT is not a suitable replacement for a human translator. If time and money is of the essence, then clients and language service agencies should not be trying to save a penny (or to pay a penny, as one potential client of mine recently did – they went with an agency in Colombia that bid .01/per translated word) by using machine translation because it takes longer to undo (most often it is just more economical, in terms of both money and time, simply to re-do the entire translation) by using MT tools and, perhaps (caveat emptor: this best practice is not often actually applied), a proofreader or translator to review it. Sometimes what seems like a straight line, and therefore the shortest route, is so full of “traffic” that the seemingly less direct route is actually the shortest in time…and cost.

PS See – and be sure to “Like!” – our Facebook page for some entertaining translation gaffes…a few are old chestnuts, but they are still so very valid in today’s “short-cut” world!