“Quality is the most important thing.”
“A translation must not sound like a translation – it should sound like an original text.”
“We only ever allow translators to work into their native language.”
It’s time to dig around some translation industry clichés. For many buyers of translation, the quotations at the top of the page will be very familiar – and in many ways Translate to English the statements are absolutely right. You would be unhappy to receive a translation that was low quality, that sounded like a translation (and not an original), and had been written by a non-native speaker.
Yes and no.
Yes, you would be unhappy if high quality, polished translations by native speakers was what you required.
But the assumption that high quality, polished native speaker translations is the only requirement is false, and a buyer of translation should be even more unhappy to have been sold an inappropriate service, especially when the service they were sold would have been considerably more expensive than the service that they actually required. Mis-selling is restricted to the world of door-to-door insurance salesmen.
So the question is: when would we use non-native language translators? There are five overlapping factors that will point towards some answers: price, quality, speed, content, and purpose
The “umbrella” reason for using non-natives is purpose. What is the translation actually for?
If a translation is purely “for information” (for internal use only, just to know what something means), then it doesn’t really matter how beautiful the translated language is. The key to such translations is accurate rendition of meaning – and this can (usually) be done absolutely as well by a translator working into a language that is not their mother tongue. In fact (and this now brings in the content reason), the most accurate understanding of any source text is likely to be achieved by a native speaker of the source text. So for a highly technical source document, when the translation is only required “for information” maybe we should after all consider using source language native speakers.
Translations “for information” might be assumed to be chunky texts, maybe a manual, or a tender document that needs to be understood to prepare a response. A classic cause for non-native “for information” translations is in the field of market research. Imagine you have respondents in 20 different languages to a questionnaire. Many of the answers on the questionnaire will be fixed choice responses but perhaps you have some open-ended questions as well: questions to which the respondent could write anything up to several hundred words. You need to know what they are saying, but only so you can incorporate the responses into the overall data for analysis. So you just want to know what has been said, and you don’t even care if the translations have the odd spelling mistake or typo. It simply doesn’t matter (in fact the answers of the respondents will probably have spelling mistakes anyway). You may not even want a record of exactly what they wrote – just for the answers to have been categorised to allow for analysis, with the linguists coding the answers against a code frame, created either by you or by the specialist market research translation company. It absolutely makes sense to consider breaking the “rules” about translators only working into their native language. Here you don’t care if the translations sound like translations. Natives of the source language will almost always be cheaper than English native speakers, just as good for the purpose, and the accuracy and consistency could be higher, especially if the linguist is coding directly.
Price of course can play a big factor. As mentioned above a non-native English speaker will almost always be cheaper than a native English speaker, and dramatically so if the source language is from a region with very low costs e.g. China, India, Eastern Europe.
Supply and demand links into the price, quality, and speed reasons for using non-native translators. This is particularly relevant for translations into English from anything other than the major world languages.
Let’s suppose we need to translate a technical manual from Vietnamese into English, and we want to stick to the rule of only using translators working into their native language. How many native English people learn Vietnamese to professional fluency AND decide to be a professional translator AND happen to be available when we need them AND have relevant sector expertise? Not many, if any, will be the most common answer. And if there is somebody suitable you can be sure they won’t be cheap.
One solution is “translation by committee”. We find a native Vietnamese translator who can provide his best effort at an English translation. An English native (no Vietnamese skills necessary) proofreads and cleans the English text whilst in constant contact with the Vietnamese translator. The result is an accurate, English native quality translation dome at the speed we need it and at a price that reflects the supply of Vietnamese translators who have good English skills rather than a price that reflects the almost non-existent supply of native English Vietnamese translators.