by Ross Perlin
Technology can help save small languages—or it can help kill them.
Technology alone isn’t silencing our languages. But it won’t single-handedly save them either.
At first it seemed like the internet would reinforce the dominance of “killer languages” like English, Spanish, and Chinese. Then came the hope, in the last few years, that the internet might make room for a multitude of smaller languages. Maybe the virtual world, as it grew ever more oral and bottom-up, would become as deeply, humanly diverse as the physical one.
Now a more muddled, sobering picture is coming into view. As a handful of platforms come to dominate the internet, consolidation and commercialization are putting limits on linguistic variety. Google’s claim to “organize the world’s information” and Facebook’s bluster about “bringing the world closer together” ring hollow when these things only happen in around 5 percent of the world’s languages.
At the edges, or in the interstices, of the official, searchable internet—that’s where the languages are. Click away from the shiny, texty, corporate pages where at most just a few hundred languages flourish, the biggest handful way out in front, and you can chase traces of the world’s actual, vanishing linguistic diversity, some 7,000 languages deep.
How that diversity holds up in the face of new and evolving technologies will depend on their design (how they handle orality, for instance), their architecture (e.g. how open they are), and on the efforts of individuals and communities. Smaller languages and cultures stand a chance if they can organize, muster a critical mass of users, reproduce, and meet and mingle with other cultures on an equal footing.
Not just online, but offline too—where urbanization, analogous to the ingathering of the internet, is challenging the assumption that smaller cultures need isolation to survive. Today’s internet, likewise, is a clamorous site of last-minute linguistic life and death: a “Babel in reverse.” Groups don’t need to wall themselves off—hardly possible now anyway—in order to preserve and develop their heritage. But they do need rights, resources, and respect to manage the transition from traditional territories to diasporic possibilities.
Tearing Down the Tower of Babel, Again
At the Endangered Language Alliance (ELA), physical and digital diasporas are our point of departure. Our home base of New York City, with up to 800 languages spoken in the metropolitan area, is the most linguistically diverse place in the history of the world—precisely and paradoxically at the moment when more and more languages are disappearing.
To take one example, as many as one-fifth of the world’s languages originate in the greater Himalayan region that stretches from northern Pakistan across to Vietnam. Within two generations, fueled by the occupation of Tibet, climate change, and diminishing economic opportunities, a new Himalayan diaspora has taken shape. This diaspora is increasingly digital and urban, rooted in cities like Kathmandu, Dharamsala, Delhi, and New York.
Speakers of Himalayan languages—who also use the English, Nepali, Tibetan, Hindi, or Chinese internets—are forming their own WeChat and Viber groups, hundreds of members strong, making room for local, oral vernaculars like Amdo Tibetan, Sherpa, and Mustangi. They are creating projects like ELA’s Voices of the Himalayas, a series of oral history videos watched across the diaspora, featuring a dozen Himalayan languages rarely seen or heard online before.
Crucial, along with the spread of smartphones and social media, is the new ease of recording and sending audio messages, sometimes video too, within group text streams. Most of the world’s languages were traditionally oral and unwritten. Many still have no script, but the newly widespread literacy in larger languages and the ease of texting are prompting more and more speakers of endangered languages to experiment with writing their languages. Some have started battling with ICANN, Unicode, Microsoft, and the like—the gatekeeping language academies of the digital age—to support non-Latin scripts.
Remixing, reusing, and readapting dominant-culture content is another major strategy for smaller languages, accelerated by the web. A whole universe of diasporic digital media now exists online: Tom and Jerry cartoons dubbed into otherwise invisible Chinese languages; Seinfeld in Yiddish; online television in Inuktitut, a major Inuit language; scenes from Troy, Titanic, and other Hollywood blockbusters, mischievously and hilariously voiced-over in southern Italian “dialects” very different from the Tuscan-based standard. (“Troy in Altamurano” has over 430,000 views; the town of Altamura has just over 70,000 inhabitants.)
YouTube alone, if you know what to look for, contains the flotsam and jetsam of a thousand languages. But it remains a long road from those phone-captured fragments and remixed movies to what sociolinguists call a whole new language “domain”—especially at a time when smaller languages are ceaselessly ceding traditional domains like work, education, and religion to larger languages.
Maybe the Apps Will Save Us
While threatened but still vital languages, concentrated in the Global South, wage a grassroots digital struggle to be heard, languages on the brink or already silent—concentrated in North America, Europe, and Australia—are counting on a tech-led resurrection from above.
“Can an App Save an Ancient Language?” “Can a Podcast Revitalize an Endangered Language?” “Video Game Offers New Life to Ancient Indigenous Language”—these are constant headlines now: positive but counterintuitive, dramatic but non-threatening. Linguists, teachers, and activists, many now employed by tribes and communities, share tools in forums like the Indigenous Languages and Technology (ILAT) listserv run by Phil Cash Cash, a scholar and speaker of Nez Perce from northeastern Oregon. Tribes like the Chickasaw Nation, headquartered in Ada, Oklahoma, are creating sophisticated iPhone and Android apps that aim to preserve the Chickasaw language by teaching the alphabet, along with basic words and phrases, and providing recordings of the remaining native speakers.
In parallel, beginning in the 1990s, many academic linguists—awakening to the disappearance of the very basis of their field—began a substantial push to document smaller languages. One driving force was the availability, for the first time, of relatively cheap, portable, and easy-to use video and audio equipment capable of producing archive-quality recordings. Gathered primarily for academic purposes in digital repositories like the Endangered Languages Archive and Documentation of Endangered Languages (DOBES) collection, these represent a significant but largely frozen legacy, aimed as much at tomorrow as today.
They may be online, but most of the corpora gathered by linguists in the field, like the peoples whose voices they represent, are still too small, poor, and particular to interest the likes of Google and Facebook. Smaller languages may only have whatever material one linguist, often a graduate student, can muster on a short-term grant. How accessible such recordings can or should be on the wider internet remains an open question, where communities differ.
By contrast, big languages have big corpora: annotated text collections gathered by governments, companies, or researchers that allow for all sorts of data analysis. Google Ngram Viewer, for instance, lets users search printed sources over time in a dozen larger languages, like English, Chinese, and Spanish. New technologies and platforms, dependent on capital under the current system, are inevitably developed with big markets, communities, and languages in mind, leaving language activists to push their way up through the digital cracks.
Do Androids Dream of Small Languages?
Dreams of a magic bullet for languages persist, from Douglas Adams’ “Babel fish” to the real-time translation promised by Google’s new Pixel Buds. Online language teaching, universal machine translation, enhanced voice recognition—these are just a handful of possible cyborgian fantasies. Nor is it unthinkable that new techniques of capture and simulation through technologies like virtual reality could make the past effectively eternally present, in some kind of linguistic parallel to the push for indefinite life extension, cloning, and the de-extinction of biological species. The annihilation of space by time may be followed by the annihilation of time altogether—and the past, as we’ve traditionally known it.
In the meantime, there is the risk that futurist fantasies will distract us from developing strategies grounded in the everyday lives of communities of speakers, especially the smallest and most threatened. Any tool not integrated with daily, oral use in families and communities is likely to remain a symbolic gesture—a morale boost, rather than a true linguistic revival. No technology developed so far can rival the intensive, unsung process of language transmission that takes place between adults and young children, almost always at home.
More dangerously, a magic-bullet mindset might make us misconstrue what language is all about. It has always fallen to people—through extensive and unending efforts of understanding and translation—to sustain the age-old ecology of difference that lies behind the real richness of the world’s information and the world’s relationships. If language is prototypically lived through densely contextualized meanings unfolding spontaneously and dialogically in real time, it is hard to see how any device, however fantastical, could be more than “merely one element in a transient flux of compulsory and disposable products”, as Jonathan Crary puts it. Meaning would be flattened, and what would remain is just metabolism: “how the rhythms, speeds, and formats of accelerated and intensified consumption are reshaping experience and perception,” as Crary writes.
Subcultures, speaking chosen dialects, will still flourish online, as they do in cities. But the future of cultures—rooted in history and mostly passed down in families, anchored in distinctive value systems and long-term lifeways—is still up in the air. More deeply different, they are also much more of a threat to any kind of hegemony. This is why we need them, now more than ever. Coding languages are not human languages, which are still an intricate index, after all, of the time that people spend with other people.
Ross Perlin is co-director of the Endangered Language Alliance and is working on a book about the languages of New York.