Olympics: the cost of “reduced price” translations

I took my family to the Olympics yesterday. We went to see the rowing at Eton Dorney, and had a great day, despite the return of normal service regarding the British weather. We queued to have our photo taken inside the Olympic rings…

…and my children even managed to apprehend a disturbing cyclops monster that had somehow evaded the security checks on the front gate.

However, a linguist is never off duty, and I thought you’d be interested to see the following sign.

It doesn’t look too bad at first, does it? You can’t argue with “Informations” or “Toilettes”… but wait, what’s happened to the translation of “Concessions”? Tarif réduit (=”reduced tariff”, of course) would be fine IF we were talking about the sort of concessions which are reduced-price tickets – but as you can see from the icons next to the text (a bottle and cutlery, plus a shopping bag) we’re actually referring to concession stands (i.e. food and merchandise outlets) in this case – and if you’ve visited any Olympic venues yet, you’ll be well aware that the prices charged by these stands are anything but reduced. So in short, this is a mistranslation. Oh, dear.

Unlike in my previous ‘Olympic snafu’ post involving meaningless Arabic signs, the translators may need to shoulder a certain amount of blame for this one. Even if they were operating without context (for example, working inside a translation memory system, which typically separates text from its original design layout for translation in the anonymous format of an Excel-like table), it should at least have occurred to a good translator that the English source text had more than one possible meaning; at which point a query should have been passed on and dealt with accordingly.

However, it’s only when you place the translation in its proper context on the sign that the potential ambiguity disappears and the error (in this case) becomes clear. So this gaffe is just as much the fault of the process that failed to provide sufficient context to the translator in the first place (and possibly also ignored a query raised by a diligent translator), and lastly failed to check the final output, which could have been done simply by showing proof copies of the final artwork to native speakers – who would surely have reported a problem before the offending articles actually made it into print.

I’ll have more to say about this in a forthcoming post…