How to insult an entire kingdom with a single translation

Professional translators know that the stakes can be high when providing translation services. Every client requires professional translation that is word perfect, whether it’s for a business card, a sales brochure or a website. 

Huge emphasis is placed on accurate translation by those who work in the human translation industry, so how is it that big corporations still frequently get their translations so spectacularly wrong?

Microsoft’s Bing insults Saudi Arabia

Already this year we’ve seen Nintendo spark protests in Hong Kong over their decision to abandon the Cantonese name for the Pokemon character Pikachu, in favour of the Mandarin name. Tesla has also been in the spotlight for a Chinese translation that described the company’s Model X crossover car as offering ‘self-driving’ or ‘autopilot’ rather than the more accurate ‘automatic driver assistance.’

Now, it’s Microsoft’s turn to make amends for poor translation after the company’s translation service, Bing, translated ‘Daesh’ (the Arabic name for the terrorist organisation known elsewhere as Islamic State or ISIS) as ‘Saudi Arabia.’ Needless to say, Saudi Arabians were none too pleased by the error. 

Microsoft’s apology

As can happen so quickly in our modern age of social media and instant messaging, the translation mistake rocketed around the world, with individuals tweeting videos of the mistranslation happening, along with expressions of outrage. 

Dr Mamdouh Najjar, Microsoft’s vice president for Saudi Arabia, was quick to issue an apology for the error, which he told the Huffington Post was down to the crowdsourced nature of Bing’s translations. The technology relies on contributions from around the world, with translations that receive the same suggestions from around 1,000 contributors being promoted to the top spot. Whether or not intentional manipulation caused the error in this case is being investigated. 

Translation manipulation

The potential for crowdsourcing to manipulate translation is particularly interesting. It is akin to the ‘Jedi census phenomenon’ when many English-speaking countries around the world engaged in a grassroots movement to turn Jedi (from the Star Wars movies) into a religion through official recognition. 

The unofficial campaign resulted in more than half a million people around the world declaring themselves as Jedi in response to the religious question on their census form. The results were most impressive in the UK, where more than 390,000 individuals claimed to be Jedis, and in New Zealand, where 53,000 people (1.5% of the population) declared their Jedi allegiance, making it the second largest religion in the country. The phenomenon shows the power of grassroots campaigns, the potential for which has increased exponentially with the growth of social media. 

The role of professional human translation

Translation manipulation is rare and not necessarily to blame for the Daesh/Saudi Arabia Bing translation gaffe, but Microsoft’s error does highlight the value of professional human translation over and above reliance on machines or crowdsourcing. 

The public as a whole has a sense of humour and it can be dangerous to rely on crowdsourcing. The UK’s Natural Environment Research Council (NERC) found that out when they left it to the public to name their new polar research vessel. Their #nameourship campaign resulted in the overwhelming victory of the name ‘Boaty Mcboatface,’ much to the delight of the British public. The NERC quickly backtracked and named the vessel the RRS Sir David Attenborough, after the famous English naturalist and broadcaster, although they did join in the joke by naming the ship's high-tech remotely operated sub-sea vehicle Boaty. 

The moral of the story? If you need something translated into another language and want it to be professional and accurate, don’t rely on inaccurate machine translation or the whimsical nature of the general public! 

Final thoughts

Why do so many large companies keep getting it wrong when it comes to translation? What more can we do to ensure such mistakes are avoided in future? Share your thoughts via the comments.