During the TAUS Annual Conference 2016 in October, panelists will discuss "How to deliver high-quality translations for long-tail languages". This blog post is written in preparation for this session.
When Nelson Mandela was asked ‘why he spoke the politically charged language Afrikaans with the South African Apartheid government during negotiations’, he answered:
“If you talk to a man in the language he understands, that goes to his head. If you talk to him in his language, that goes to his heart.”
Mandela’s words well explain the growing popularity of long-tail languages in our industry. After all, if we are to touch the hearts of our customers we need to speak to them, not in a language they understand, but in the language that’s theirs. All easier said than done as there are more than 7,000 languages spoken in this world. And we cannot possibly translate them all, can we?
I say we can’t, and fortunately it seems we don’t have to. Of these 7,000 languages, many will not survive the rigors of time. Languages are dying off like flies and it is said that in a hundred years the number of spoken languages will be as low as 2,500. In the digital realm, the outlook is even worse: András Kornai in his paper ‘Digital Language Death’ predicts that no more than 250 languages, give or take, will survive online. The remaining 6,750 will most likely die a digital death.
Should we mourn the loss of so many languages? In a way, yes. Losing linguistic diversity means losing cultural diversity and maybe even human diversity overall. A sad thing, because diversity is beautiful.
In addition, a language without digital presence can still be alive and kicking in the real world. Kornai for example, explains that Mandinka, a language primarily spoken as a first language in Senegal and The Gambia, has more than 1.3 million speakers; it is “neither endangered nor threatened in the traditional sense.” Its digital presence, however, is practically non-existent. This means speakers of Mandinka cannot access online information in their own language and thus suffer from information inequality.
That said, even the few languages that dominate online don’t give speakers access to all information. In her article “The Digital Language Divide”, for example, Holly Young states that “English – the largest and potentially most diverse edition [of Wikipedia] – contains only 51% of the articles in the second-largest edition, German.”
Perhaps that what lies ahead of us is a world of digital diglossia. In traditional diglossia, two languages or language varieties are used by one single community. The first language or dialect is the community's everyday or vernacular language whereas the second one is used in domains such as formal education or politics, but not for daily conversation. A digital diglossia would be somewhat similar to this traditional one; except the domain in which the second language variety is used is the digital realm and not the physical world.
From a business perspective, the loss of linguistic diversity in the digital domain could make our lives easier. Knowing which languages are likely to die out online and which ones are not, may help to decide which long-tail languages to focus our localization efforts on.
About 3.5 billion people currently have access to the internet and the number keeps growing every day. Imagine, not too long from now, we will be able to reach all these people with just a (Big Friendly Giant’s) handful of languages. Localizing 250 languages is still a lot, but if we know that these are the ones giving us access to the world’s hearts and minds, why not work together to get the job done?
3 minute read