A translator’s response to the question, “What do you do?” is often met with a mixture of fascination and confusion, leading to questions such as, “Wow, do you to do it right there, in person?” (no, that’s interpreting) and “But can’t Google translate do it?” (spoiler alert: the answer is no). Essentially, the world of translation holds mystery for many – some of which this piece intends to unravel, to give you a new understanding of what a translator does and the skills they need to do it.
“Monolingual” people tend to be fascinated by the idea of being able to speak another language. From a translator’s viewpoint, this general awe clashes with the commonly held impression that translation is easy – indeed, there is a strong movement in the industry pushing for translation to be recognised as a profession precisely because, as the saying goes, “just having hands doesn’t mean you can play the piano”.
Indeed, a vast number of the translators you’re likely to meet will be signed up to a professional body, such as The Institute of Translation and Interpreting – the only UK-based independent professional membership association for this profession. More on that below.
There are some terms worth knowing to help get your head around the process: the original text is referred to as the “source text” (ST). It’s written in the “source language” (SL). After translation, the text produced will be the “target text” (TT) and that’s written in the “target language” (TL).
Good practice says the TL should only ever be your mother tongue (or a language you speak to that level) and, upon joining the aforementioned professional association, you’re asked to sign a code of conduct that contains precisely that point.
It’s also worth noting that most translators nowadays are trained to use software to help them. Computer Assisted Translation (CAT) tools advance more each year and, naturally, their pros and cons are widely debated and useful to different extents on different text types – but it’s another example of how there’s more to being a translator than first meets the eye.
What lies beneath the surface of a good translation?
Budding translators are encouraged to specialise in an area of expertise. If you have a degree in Economics you’re probably not the right person to write the manual on how to assemble a complicated piece of machinery, just as if you’re a medical professional it’s unlikely a business would come to you to draw up a supply agreement (they’d not only want a lawyer but one specialising in commercial law). It’s no different in translation: so working this field isn’t just about knowing another language but knowing a specific field in both cultures.
Say you want to translate the terms and conditions of using a particular online game from Russian into English – you need to find a way to capture the meaning of the terms in the ST in the TL. The legal terminology will be based around concepts in the SL culture – is there a TL term you can use that captures all the meaning of the SL term? Or do you need to add a gloss? Additionally, T&Cs are intended to be user friendly – the translation thus needs to capture the meaning but also maintain the same balance between technical and comprehensible as it does in the ST.
There is also an argument that the intended use of the TT will govern how you translate it. It’s one of the fundamental principles of translation theory and it’s called “skopos”, a Greek that loosely translates as “purpose”. The most basic example I could give is a text that someone merely needs to understand the gist of in which case full grammatical precision isn’t a major requirement – the aforementioned T&Cs (which will be published) do not fall under that category.
Why won’t Google do?
There are books and books on this, but here’s the basic reasoning: Google Translate has essentially become the household name for machine translation (MT) of which there are two main systems in use, Rule-Based MT and Statistical MT. Each has its own merits; neither is quite AI, but the essence of it is that it uses existing translations (in corpora) and/or rules to turn a ST into a TT. Computers can’t tackle nuance, though, whereas a translator – especially one specialised in a particular field – can nail it.
But basic MT systems deserve some credit for providing entertainment: the story at last year’s Winter Olympics of the Norwegian team ending up with 10 times the number of eggs they needed is gold. Of course, Google, Microsoft and several other enterprises have their own, more sophisticated MT systems and there are plenty of examples of AI under development that promises highly sophisticated solutions – but the bottom line is that, currently, machine translation isn’t good enough.
In summary, there’s a lot more to translation – as a task and as a profession – than many are aware of. Hopefully, this short piece has served to open your eyes to a fascinatingly deep field of practice.