In space and time, the intensity of intercultural communication and interlinguistic exchanges largely depends on the quantity and quality of information translated from one language to another, but history has shown that circulation and “notoriety” of ideas do not merge quantitatively with the most widely spoken languages. In particular, the number of speakers of a spoken language is not a good predictor of the ability of a message created in that language (or circulating in that language) to be subsequently translated and circulated around the world; According to linguist David Crystal, “Why a language becomes a global language has little to do with the number of people who speak it, but much more with the identity of those speakers”. The network of bilingual speakers and translators is therefore of great importance from this point of view.
Since the 19th century and with globalization and the regulation of “intellectual property rights” and translations, a number of languages and cultures are more or less well “translated”, even dying out more quickly than before or are already dead or forgotten (it being understood that a dead language like Latin can continue to be translated).
Some authors describe the emergence of a new network and a world system of languages, where English plays a role that has become preponderant and central. The cultural-linguistic hegemony of English could, however, be gradually contained by the improvement and generalization of automatic translation software on the Internet and by the unprecedented approach of Wikimedia, which encourages and facilitates “translations and interlinguistic exchanges in Wikipedia and its sister projects (translations into 287 languages possible at the end of 2013, including so-called “dead” languages and Esperanto, with several major bilateral language projects).
Analyzing the relative status of the world’s languages has long remained impossible for lack of relevant data, notes Mark Davis (president and co-founder of the Unicode Consortium, which produces character encoding standards for all computers and mobile interfaces in the planet using writing), even though we sense the importance of the structure of this network; it has long remained impossible to quantitatively study the structure of the global network of language exchanges, but it is becoming easier through the establishment of large open databases of global exchange “places” such as Wikipedia or Twitter and as we know more and more about the proportion of languages spoken on the Internet.
In 2014, an American-French team used network science to create maps to visualize how information and ideas circulate in the world [according to the language of the original message, according to the average GDP of the countries where this language is spoken, according to the language of the first translations and those which will convey the information or according to the medium (books, Wikipedia, Twitter)…]. To draw up this “map”, these researchers studied on the one hand the available data on literary translation (based on 2.2 million translations of books published in more than a thousand languages) and on the other hand the two major global language exchange networks which are:
- bilingual tweets (based on the study of 550 million tweets, 17 million users in 73 languages, selected for the study), which was possible thanks to the open database and because this allows a tweet to be associated with a language and the person tweeting with one or more linguistic communities;
- different language versions of Wikipedia pages (without taking into account the work of robots in Wikipedia), whose database is open (DBPedia).
Here is what analysis of this data reveals:
- there is a significant hierarchy of “interfacing” languages in this network, with nuances depending on the medium studied;
- unsurprisingly, English is the most important and effective language for assuming the function of interface between other languages to disseminate an idea or information in the world (English constitutes, in the mapped network, the most central hub). At the next ranks, especially in Wikipedia, French, German and Russian play a similar role, then comes a constellation of smaller “hubs” with for example Spanish and, far behind, Tamil, Portuguese or Chinese, languages not conducive to the global dissemination of ideas although they are spoken by a very large number of speakers. Unlike English (which is spoken almost everywhere in global ideation networks), Mandarin, Hindi, and Arabic, while immensely popular, are isolated in the network of exchanges. between languages (meaning that communications in these languages reach speakers of other languages less, and less quickly);
- in terms of major nodes in the network of interlanguage information exchanges in the global network, literary translations and Wikipedia’s interlanguage system (283 languages in 2014) still mainly value European languages (and Japanese for translations), but Twitter gives more importance (after English) to non-major languages in the two previous exchange networks (Malay, Portuguese, Spanish, Filipino, Dutch, Arabic). The network of literary translations is more stable and formal. Wikipedia is evolving rapidly, but while being structured, while Twitter offers a totally different model, consisting only of short messages, very reactive to the news;
- speakers of disparate or rare languages benefit from being indirectly connected to other languages via a hub (large or small) if they want their messages to circulate in the world. Twitter can circulate ideas within a group of close languages (e.g. from the Philippines to the Korean area via Malay), while a translation through English will facilitate the circulation of an idea from the Turkish language to Malayalam (spoken in India by 35 million people);
- bilingual or multilingual people or institutions therefore appear as important “nodes” in the network of transmission of information and ideas. The Internet and phenomena like Wikipedia and Twitter have amplified their role as language converters, but their ability to circulate information remains much greater if one of the languages the user masters is English;
- there are a few atypical or emerging phenomena: for example, Dutch is only spoken by a “small” number of people (27 million speakers, which is far less than Arabic which has around 530,000,000 speakers), but the Dutch are both often polyglot and very active online;
- the users studied constitute a kind of elite, literate and active “online”. And although sometimes representing only a small part of all the speakers of one or more languages, this elite has a “disproportionate” power and responsibility because, voluntarily or not, it leaves its imprint (or even certain bias) the messages it translates and relays to other languages, cultures and distant peoples. This is especially the case for Anglophones whose messages seem the most likely to travel far and fast;
- the low rate of translation of texts originally written in many languages into Arabic and the Arab world is an obstacle to the dissemination of “external” knowledge;
- a country that encourages the translation of many documents into English (or into one of the languages that are the best relays) will become better known. The choice of a second language which is very well connected to the other languages on the Internet of social and cultural networks is then an asset;
- a non-English speaker who wishes to circulate ideas or have access to new ideas outside their culture has every interest in choosing English as a second or third language, whereas an English speaker will benefit from choosing Spanish, French or German rather than Chinese or Hindi, at least for the dissemination of ideas in print.
Cultural transmission also involves spoken language, locally and remotely (via telephone or Skype), which could accelerate the dissemination of certain ideas and information.
(Includes texts translated and adapted from Wikipédia)