Half the planet’s 6,000 spoken languages are now in danger of extinction. Why do we care so much about languages most of humanity will never hear?


“Exploring the full range of human languages is to the linguist what examining the abundance of species on the planet is to the biologist. There may well be crucial questions about the structure of human language that can only be answered by working with, say, speakers of Navajo."

— David Pesetsky Ferrari P. Ward Professor of Modern Languages and Linguistics, MIT Department of Linguistics and Philosophy

On Mornington Island

Mornington Island is situated in the southernmost end of the Gulf of Carpentaria, the deep bay that scoops into the northern coastline of Australia. Part of the Queensland territory, the island lies midway between the mainland cities of Darwin and Cairns. A land of soft scrub vegetation and windswept tropical beaches with rocky outcrops, Mornington has for 10,000 years been home to an aboriginal tribe known as the Lardil.           

Until about a century ago, the Lardil lived as a hunter-gatherer society, sharing the island with only the sea turtles, the manatee-like dugong, and other coastal species of Oceania. The indigenous people’s rich ceremonial life included traditional dances and songs (many of them forms of teaching), and spiritual rituals and narratives derived from what aborigines call “the Dreaming”—a parallel “time outside of time” mythology, both ancient and ongoing, that tells about the formation of the landscape and the powerful Rainbow Serpent.  

A language passed on in privacy

The Lardil’s ancient way of life began to change in 1914, when English-speaking missionaries arrived on Mornington Island from the mainland. Intending to save souls, the missionaries did not, alas, have similarly protective ideas toward the aboriginal culture; they banned the Lardil’s traditional tribal rites, including the telling of creation stories, and children from their elders, housing them in dormitories where they were forbidden to speak or hear their own language.

For much of the 20th century, the Lardil language was passed down in privacy, and, as the generations of fluent speakers perished, only a vastly simplified version of the once rich language remained. By the mid 1980s, the late Kenneth Hale, an internationally known MIT linguist and determined savior of the Lardil language, estimated that there were fewer than 50 native speakers left.

Today, there are none.

Indeed, in 2009 the world’s most proficient user of the Lardil language is not an Australian aborigine at all, but a red-haired, Alabama-bred, MIT linguist hand-picked by Hale to continue Lardil restoration efforts.

Thinking like a linguist 

Norvin Richards, PhD’97, now Professor of Linguistics, had never encountered an endangered language before visiting Mornington Island with his celebrated mentor in 1996. A Tuscaloosa native and the son of a philosophy professor, Richards first delved into foreign languages when, as a precocious 12-year-old, he enrolled in Duke University summer courses in Chinese, German, and Russian. There he found he was more interested in decoding the structure and organization underlying the languages than in their potential uses. As he puts it, “I was baffled to learn that my friend wanted to know Russian so he could read Tolstoy in its original; instead, I spent my time trying to understand the complicated rules systems that governed how languages fit together. I wanted to develop a theory to predict the gender of German nouns. Of course it can’t be done, but it was the first time I can remember thinking like a linguist.”

How does a linguist think?

Film actress Ingrid Bergman, who spoke five languages, was often quoted as saying that she preferred English for acting, Italian for romance, French for diplomacy, German for philosophy, and Swedish for secrecy. By contrast, linguists are not mining the world’s vocabularies to diversify their own self-expression; indeed, they are not necessarily polyglots.

Hale was a rare exception, a linguist who was rumored to absorb language systems through his pores. Although Hale himself admitted only to being able to “say some things” in a few tongues, admirers have claimed that he spoke more than 50 languages fluently. One popular anecdote depicts him learning to speak Dutch by reading a single novel en route to the Netherlands for a teaching engagement. David Pesetsky, PhD’83, who succeeded Hale as MIT’s Ferrari P. Ward Professor of Modern Languages and Linguistics, shares a tale that has Hale mastering an obscure aboriginal dialect while his car was repaired during a two-day stopover in the Australian Outback. Says Richards, “Today, wherever linguists gather in the world, they sit and swap Ken Hale stories.”

The late Ken Hale, Ferrari P. Ward Professor of Linguistics at MIT teaching a class on Australian aboriginal languages, early 1980s; photograph courtesy of the Hale family 

The infrastructure of language

For most linguists, though, the interest in diverse tongues is a way of expanding their knowledge of what really fascinates them: the infrastructure of language. Given the basic puzzle of how infants all over the world arrive hard-wired to learn quickly and intuitively the complex rules governing whatever languages they happen to hear in early childhood, linguists want to know why and how. Using what Pesetsky calls “the normal tools of scientific inquiry,” linguists strive to determine what properties the approximately 6,000 languages currently spoken on earth have in common, and what properties differentiate them.

Linguists ask, for example, why the verb precedes the object in English, and, in Japanese, follows it. Why is it that, in many languages, you cannot move the “not” in a negative sentence without changing its meaning? What happens when you turn a statement into a question, and why do some clauses demand a “that,” while others, with similar structures, do not? Why is it that when you say, “John thinks he is smart,” the “he” can possibly mean John, but when you say, “He thinks John is smart,” the “he” cannot possibly mean John? Moreover, why are these tremendously complicated usage systems often identical in historically unrelated languages from opposite sides of the planet? And, perhaps most important, what do the answers to these questions reveal about human cognition and about how people manage to understand each other?

The science of speech

At MIT Linguistics, widely recognized as one of the premier places in the world for linguistics, the faculty is divided about equally among phonologists, who study sound systems; syntacticians, who analyze sentence structure; and semanticists, who explore how speakers endow and listeners extract meaning in sentences.

Linguistics and Philosophy Department head and semanticist Irene Heim says that the department’s researchers also collaborate with faculty from the Department of Brain and Cognitive Sciences who are examining language acquisition, language pathology, and language processing. Professors Ken Wexler, PhD’70, and Ted Gibson, PhD’91, have joint appointments in the two departments. At his Wexler ab/Normal Language Lab, Wexler leads a team of researchers who are looking at the development of language in children, including second-language acquisition; aphasia, or language loss caused by a brain lesion; and unusual language patterns resulting from Down syndrome, autism, and other genetic factors.

Simple words | complex rules

Originally drawn to linguistics as a natural way to link her early interests in math, science, and language, Heim discovered the field, as a teen, when she stumbled on a book at her hometown library in Germany. Today, she specializes in formal semantics, which she defines as the application of logistical analysis to language.  For example, her study on the German equivalents of the  words “a” and “the,” which reveals the  complex rules that govern these apparently simple words.

While a doctoral study on the equivalents of “a” and “the” may sound simplistic at first blush, the rules governing these words are complex indeed. Just imagine, Heim says, that you are trying to explain the subtleties of their usage in English to someone who speaks Japanese, Russian, or another language that does not have articles at all.

To find out what it is about the nature of the human brain that allows people to use language, linguists seek opportunities to mine valuable data from as many languages as possible, explains Heim. Although she has dubbed Richards “our fieldwork guy,” she also notes that because virtually all the School’s linguists supervise and collaborate with graduate students who specialize in studying global indigenous populations, fieldwork is a department-wide effort. Indeed, the distinguishing characteristic that draws scholars and teachers from all over the world to the School’s Linguistics section is its reputation for communication, camaraderie, and an exciting spirit of collaboration. Heim sees each language as a valuable data set, and, in her view, the more endangered the tongue, the more urgent it is for linguists to study, analyze, and preserve it before that unique language passes out of memory and existence.

Endangered or underdescribed?

Of course, not every language for which you can’t pick up a phrase book at the MIT Coop is on the verge of extinction. Linguists categorize the many lesser-known tongues as either endangered or underdescribed. Endangered languages — about half of all languages now spoken — have often passed from widespread use because (like the traditional language of Mornington Island) they were banned by newcomers, or because of practical considerations. Invading cultures often persuaded indigenous peoples that they must adopt the newly dominant language in order to gain a good education, find a job, or participate meaningfully in the modern community. The forces of economics, racism, and elitism have all contributed to making languages endangered. Underdescribed languages, by contrast, may still be thriving - indeed spoken by hundreds of thousands of people, taught in schools, and used in the marketplace - and yet still not have been subjected to rigorous scrutiny by language scholars.

As an undergraduate at Cornell, Richards traveled to the Philippines to study the intricacies of the Tagalog language, a tongue not endangered, but very underdescribed. It was only when he met Hale that Richards began to appreciate the importance of studying the world’s many imperiled languages.

Norvin Richards, MIT Professor of Linguistics, and Irene Heim, MIT Professor of Linguistics, in conversation in the Stata Center; Photograph, Webb Chappell

Cultural treasure

Greatly valued for the knowledge and understanding they provide linguistic researchers, endangered languages are also treasures in and of themselves. Richards recalls that Hale often told him, and anyone else who would listen, that the loss of a language is the loss of a whole intellectual life and a culture. “The death of a language,” he once said, “is a disaster. It’s as if someone had dropped a bomb on the Louvre.” When a language dies, it often takes with it a history and mythology, a distinctive consciousness and way of communicating; a door closes on the past and also on the future, on the countless new things that might have been thought and said in that tongue. Saving a language saves a part of the human story and allows it to continue on, a treasure of the past and a resource for the future.

When Richards, a syntactician, first journeyed with his mentor to Mornington Island in 1996, the Lardil language was nearly moribund. People of the youngest generation who could “speak” the language had long since forgotten all its nuance and its inflections-the prefixes and suffixes that give structure to sentences-and could reproduce only what Richards calls a “me Tarzan, you Jane” kind of dialogue in their ancestral tongue. What’s more, prior to Hale’s arrival in the1960s, those words had never been written down.

Hale began to remedy that, and in 1996 he enlisted Richards to assist in his ongoing work to compile the world’s only Lardil dictionary. After Hale’s death in 2001, the younger researcher continued working closely with the Lardil people on a range of community-based language-preservation and educational ventures. The work is still ongoing.

“One of Ken’s gifts,” Richards says, “was involving native speakers in the work he was doing. For every word we put in the Lardil dictionary, he wanted to have an example sentence, and he assigned me to have the native speakers provide those.

“When I asked them to supply the sentences, the informants seemed confused; it was like pulling teeth. Then Ken came along and explained the same thing, and the suggestions began to flow. One guy in particular became an example-sentence machine; he wouldn’t let us go for hours at a time!”

One of Hale’s greatest legacies, Richards maintains, was a massive dictionary of the previously undocumented Warlpiri aboriginal language. Containing more than 30,000 main entries, the tome, initiated at SHASS, was substantially complete before Hale’s demise and is still being augmented by colleagues in Australia’s Northern Territory.

Closer to home: the Wampanoag language of Massachusetts 

Not all linguistics fieldwork involves travel to distant hemispheres. Indeed, first Hale and later Richards worked closely with Massachusetts native Jessie Little Doe Baird, who received a master’s degree from the department in 2000, on efforts to revitalize Wampanoag. The Algonquian language was once spoken on Cape Cod, where Baird is deeply involved in affairs of the Mashpee tribe, part of the Wampanoag nation.

“Wampanoag has not been in general usage for about a century, but for an extinct language it’s in great shape,” says Richards. Its speakers were uncommonly literate, thanks in part to 17th-century missionary John Eliot, who worked with native informants to translate the King James Bible into Wampanoag, producing the first Bible published in the Americas. Through a reverse-translation project undertaken in the 1990s, MIT linguists were able to discern facts about Wampanoag grammar. Baird’s master’s thesis focused on Wampanoag contracts and legal documents, from which she distilled a useful understanding of Wampanoag nouns.

Together, Baird and Richards are compiling a Wampanoag dictionary and a grammar textbook, and Baird teaches Wampanoag to people at tribal meetings, at community gatherings, and in their homes. In the summer, Richards helps out with a Wampanoag immersion camp Baird runs for adults and children in the town of Mashpee.

“The idea,” he says, “is to speak only Wampanoag for a whole week. This leads to all kinds of charade-playing and awkward silences, but there is now a core group of people who can actually have a conversation in Wampanoag.”

"The loss of a language is the loss of an intellectual life and a culture. 'The death of a language is a disaster, said Ken Hale. 'It’s as if someone had dropped a bomb on the Louvre.'"

Maori: a success story

While Baird’s determination to revive a dead language is ambitious — even gutsy — Richards says similar reclamation efforts are under way in Native American communities on the West Coast, as well as at other sites around the globe. What’s more, there are now several documented successes in revitalizing languages that were within a candle’s flicker of extinction.

“Maori is one success story like that, and Hawaiian was very close to dead when people became alarmed and decided to do something about it. Now it’s possible to go to school in Hawaiian and even get a doctorate in a program where every class is in Hawaiian,” says Richards, noting that MIT hopes to identify a stable source of funding to support other indigenous scholars who, like Baird, wish to pursue master’s-level studies in order to facilitate language revitalization in their own communities.

Other sites, other speakers

Further north, Richards has been working with native New Brunswick speakers of Maliseet, another Algonquian language, while just across the US–Canadian border, in Maine, Benjamin Breuning, PhD’01, is documenting a markedly similar language known as Passamaquoddy. Other recent MIT graduate students engaged in fieldwork include South Carolina-born Seth Cable, PhD’07, an assistant professor at the University of Massachusetts in Amherst, who has done documentary research on the Tlingit language, a distant relative of Navajo that is spoken by fewer than 300 people on the panhandle of southeastern Alaska.

Also in the field is Jessica Coon, a current doctoral candidate from Oregon whose research takes her to Mexico to work with speakers of Chol, a Mayan language used by some 200,000 people. Still other recent MIT graduate students have conducted investigations into the Mohegan tongue, once used in Connecticut; Buli, a Ghanaian language of the Gur family; and Basque, a fully alive but still underdescribed language of France and Spain.

While many institutions have extensive fieldwork programs operating within their own general geographic regions, Pesetsky says it’s unusual for an institution with MIT’s worldwide renown for rigorous, science-based theoretical analysis to be so deeply involved in community linguistics work.

"Saving a language saves a distinctive consciousness and way of communicating. It saves a part of the human story and allows it to continue — as a treasure of the past and resource for the future."

Mebengokre: a language of the Amazon

Argentinian native Andreas Salanova, PhD’07, now a faculty member at the University of Ottawa, has been engaged for more than a decade in studying Mebengokre, a language used by about 10,000 indigenous people in the Amazon jungle. Despite the low number of speakers, Salanova declines to describe the language as endangered. Most languages that become extinct, he explains, do so because outsiders come and create negative attitudes about the indigenous tongues, equating them with ignorance or lack of sophistication.

By contrast, people in the Amazonian area where Salanova works “have a vibrant culture, with pride in their language and heritage,” he says. “Portuguese, which is the dominant language of the region, is rarely spoken in the villages, where young people are learning their native language and using it without embarrassment.”

Beyond informing his own work on how time is represented in Mebengokre, Salanova hopes his efforts in the Amazon will lead to more use of the indigenous language in the local schools and in the development of a written literature that will help preserve the region’s culture. To that end, he has begun to help develop educational materials for the classroom and is participating in the organization of a Mebengokre-Portuguese dictionary.

Salanova’s thesis adviser is Sabine Iatridou, PhD’91. Born in Greece, she first became involved with linguistic fieldwork while employed as a translator on a Dutch Creole project in an immigrant community of Suriname during the 1980s. Now a syntax specialist, she spends nearly all her time in Cambridge but believes firmly in the importance of getting linguists out into the field.

All language speakers have one thing in common 

“One reason we study as many languages as possible,” she notes, “is to test our main hypothesis, which is that languages have much in common not because they share the same historical roots or relationships, but because all the speakers have the human brain in common, and the brain is structured in a particular way. I see this as analogous to walking: We each may walk in a slightly different way, but basically the ability to walk is dictated by the very nature of our brain. The more languages we can compare, the more certain of this we can become.”

Pesetsky makes a similar point when he says, “Exploring the full range of human languages is to the linguist what examining the abundance of species on the planet is to the biologist. There may very well be crucial questions about the structure of human language that can only be answered by working with speakers of Navajo. When and if the last speaker of Navajo dies, that possibility will be gone from us forever.”


Suggested Links 

MIT Linguistics

MIT Indigenous Language Initiative | MITILI Application

Free, online MIT Linguistics courses from MIT Open Courseware

Related Stories

Saving Iñupiaq
Annauk Olin, a graduate student in MIT Linguistics, is working to help her Alaska Native community preserve their language and navigate the severe impact of climate change on their coastal village. September 2020

Jesse Little Doe Baird, SM '00, receives MacArthur grant
for her work to revive the Wopnaak language

Film on Wampanoag language reclamation project, led by MIT alumna
"We Still Live Here: Âs Nutayuneântells the story of the return of the Wampanoag language — the first time a language with no remaining native speakers has been revived in this country.

Wôpanâak Language Reclamation Project

Q and A with linguist Claude Hagege
Author of "On The Death of Languages
The New York Times | December 16, 2009

Jesse Little Doe Baird, SM '00

Prepared by MIT SHASS Communications
Editorial team: Theresa Pease and Emily Hiestand

Published Spring 2009

Soundings, Spring 2009