It was only a few weeks ago I wrote about the future of the English language, and I’m already about to eat some of my words. At the time, I suggested that the internet would help English grow in popularity. Now that anyone can talk to anyone else in the world, regardless of geography, a common tongue is more useful than ever, and since English is the most widely used second language, I figured it was a sure bet.
But I may have spoken too soon. According to this infographic from Smartling, the fastest growing internet language is not English, but Chinese. Check out the 3-layer pie chart below – it shows the languages spoken by internet users in 2000 (inner circle), 2005 (middle) and 2011 (outer).
You can see that over the last 11 years, English speakers have fallen from 39% of web users to just 27%. Chinese (labelled ZH in the picture) is just behind, at 24%, but is catching up quickly (in 2000, only 9% of web users spoke Chinese). Also seeing rapid expansion are Arabic and Russian speakers.
Contrary to what I suggested, the diversity of languages on the web seems to be growing rather than shrinking. According to Smartling, it’s because developing nations are catching up to their developed brethren. In English-speaking countries, just about everyone who wants the internet has it by now – the market is saturated. But China is still adding users by the thousands, and shows no sign of letting up.
This points to a future where the linguistic diversity of the internet more or less matches the real world – so much for a lingua franca.
The graph above is just a small part of the infographic – check out Smartling’s website here for the full-size, interactive version. Hat tip to Tech in Asia for the find.
Cosmic Revolutions 
Very cool! I love a good graphical display of such complex data. But I wonder if this is a raw document count, or a total based on documents accessed by users. I’m not surprised if these are raw numbers (considering population), but I wonder if usage is very different.
I mention this because accessibility of a language will likely impact its future prevelance, especially across cultural lines. English is a more simple language to learn and is already accessible to many people outside of native speaking cultures. The same is not true of Manderin, which is among one of the most difficult languages humanity has yet created.
There’s a little confusion over what the data represents, actually. The Tech in Asia article suggests that it’s the amount of internet content written in those languages. But if you look at Smartling’s website, they seem to be talking exclusively about the number of internet users who speak a particular language. So I think it’s a snapshot of people, not webpages. But your point about usage is still a good one – it’s entirely possible that the English-speaking users are in the minority, but doing more than their share of internet-using.
In any case, I agree with you that English is still an attractive option as a common internet language. One thing the data doesn’t show us is how many people speak secondary languages – and on that count English is still far ahead of Chinese, in the world if not the internet.
I’m with Brad. It’s totally fun and satisfying to see such a complex issue compiled into a pretty graph.
Also, I think your original argument still holds a lot of ground. Sure, this graph shows that, as you wrote,”the diversity of languages on the web seems to be growing rather than shrinking.” It also shows that the two dominant languages have, together, grown from 45% to 51%. Just because Chinese is pulling ahead doesn’t negate the fact that web users tend to conform toward the languages that more people will understand.
Great observation! That means half the internet’s users speak one of two languages. If that trend continues, perhaps we’d stabilize with a distribution of roughly 1/3 Chinese, 1/3 English (mostly as a second language) and 1/3 everything else. Do you think that sounds plausible?
totally plausible! and really convenient 🙂
Wow. I think cultural predictions of the most difficult to make, because predictions of every other kind-economic, technological, political-play into the picture.
So we are seeing the rest of the world catching up in terms of access to computers, and the knowledge and skills to utilize and program them.
I think widespread real-time translation is still at least 10 years away. It’s not that we won’t have the computer power to achieve real-time translation, but rather it’s about the quality of the translation. In any meaningful business or personal communication, there are subtleties and shades of meaning which are going to take some time to teach to a machine.
Everywhere I’ve gone in the world people want to learn English. They recognize that it grants access in so many ways – to information on the Internet and other places; to job opportunities in their own countries; to a chance to emigrate to the United States or other parts of the West; to the culture of American music and celebrity has reached the furthest corners of our planet via satellite connections. I also agree that English is much easier to learn than Mandarin, not just inherently but due to the exposure that most of the world already has to either language.
I totally agree that we’re heading to a place where “linguistic diversity of the internet more or less matches the real world,” but also that this will be skewed toward those nations with greater economic power. I’m also reminded of a common scenario in India: most people there speak two or three languages (sometimes even more). Because there are dozens of official languages, and hundreds of local dialects, people end up speaking certain languages in the home, and often another, regionally spoken language outside the home. In some cases they may even conduct business in a third language like English or Hindi (India’s official language, though English is its unofficial official language) if they are connected in a national or international way. So I do think English will continue to become a lingua franca of many business and personal interactions which extend beyond regional and national borders.
Smartlings states that despite the rise of Chinese users, “56% of online content is English-only.” I’m sure this is partly because they haven’t ‘caught up’ yet in generating Chinese language pages. But, I’m am 100% certain that far more Chinese speakers are learning and will learn English, than English speakers will learn Chinese, both on a numerical and a percentage basis.
In looking the probably future distribution, we now have to look at the remaining people who are currently not connected – what language do they already know, or will they learn? Overwhelmingy, English is the answer for half the population of India not yet connected, more than 600 million. And despite the fact that the rest of Asia with well over a billion people is in China’s economic and cultural sphere of influence, I think that the United States global cultural presence is a huge factor. Of course, Africa could be a wild card, with one billion people, who speak mostly Arabic, French as first or second languages, but here also, I think the US overculture will win out.
Ranjan,
Thanks for adding your real-world experiences to the discussion! I could spout off abstract theories all day long (and I do), but nothing compares to a good reality-check.
Also, interesting observations of the remaining populations who aren’t online yet. If those one billion Africans come online at more or less the same time, they could create entire spheres of the internet in French or Arabic, and those might be large enough to be self-sustaining. Even a Hindi region of the internet could well be popular enough to persist in the face of other common languages. But if new users trickle in slowly, they’ll be under a lot of pressure to adopt a second, more common language, and in that case English seems the most logical choice.