This is a story¯a creation myth from the Tofa:
In the very beginning there were no people, there was nothing at all.There was only the first duck, she was flying along.
Having settled down for the night, the duck laid an egg.
Then, her egg broke.
The liquid of her egg poured out and formed a lake.
And the egg shell became earth,
And that is how the earth was created.
The Tofa are an obscure, reindeer-herding people in Siberia, and their language is dying. In fact, the history of this story shows the rate of Tofa’s decline: It was reported to scholars thirty years ago; today, only thirty or so Tofa speakers remain, and few of them even remember that there had been a duck creation story. None could recount it in its fullness.
This Tofa myth comes from When Languages Die 
Harrison is director of the Living Tongues Institute for Endangered Languages and assistant professor of linguistics at Swarthmore College (full disclosure: he was also my undergraduate thesis adviser). He and Gregory Anderson, working with the National Geographic Society , have identified "language hot spots"¯places on the globe where there are multiple languages that will soon fall silent when their last remaining speakers die.
Stories about their work appeared everywhere, from the New York Times and the Washington Post to The Guardian . Harrison even appeared on Stephen Colbert’s popular television show , sharing with Colbert popungkoontam , the Sora word for "I will stab you in the belly with a knife."
It was funny, but should we really care? I personally feel as though I have led a happy life without popungkoontam or intimate knowledge of the Tofa creation duck. When a language dies, what is lost and does it matter?
The answer has to do with the significance of language itself. Language is the matrix in which the human mind creates and passes down its work. Knowledge, culture, human interaction¯humanness itself¯are possible only because of language. Each individual language, in turn, is a product of the particular minds that have learned, used, and changed it over centuries, and each language is unique in its ability to capture human expression.
In many instances, languages can be translated, but every speaker of more than one language knows that no such thing as a perfect translation exists. So, if languages are not interchangeable, and if they are crucial to understanding humanity, then Harrison is right¯and we really should have people like him out in the field recording dying tongues, especially the smallest and most eccentric. We need to know about things like popungkoontam and duck creation stories, because without them we lack the information bound up in those languages.
To which the answer, of course, might be: So what? Who needs a term for a male domesticated reindeer, during the first mating season, which may be castrated, but even if not, will probably not be allowed to mate? ( Döngür , as the Todzhu say.)
And yet endangered languages have often developed in small regions, with a small group of speakers. They contain intricate systems of classification specific to the life and surroundings of those speakers. Their folk taxonomies are built on observations accumulated over generations, and they can describe and classify, say, the eating habits of Amazonian birds or Hawaiian schools of fish in ways unknown and helpful to modern science. They can also be useful for pharmaceuticals: In 1990, drug companies made $85 billion in profits on medicines derived from plants first known to indigenous peoples for their healing properties.
More to the point, the taxonomic information in indigenous languages cannot easily be translated. As Harrison writes, "You cannot merely substitute labels or names from another language and hold on to all of the implicit, hidden knowledge that resides in a taxonomy or naming system." The Danish naturalist Johan Christian Fabricius was more blunt: Nomina si pereunt, perit et cognitio rerum ¯If the names are lost, their understandings are lost as well.
For that matter, languages also contain counting systems, calendars, cartography, and duck creation stories¯together with epic tales, songs, and poetry: an abundance of human culture. Take the epic tales of the Tuvans, from a part of Russia near Mongolia. My favorite begins with man out grazing his flock by a river. Standing on the bank, he sees a lung floating down the stream. He reaches for it with a stick, but the moment he touches the lung¯poof!¯it turns into a black monster, who ravages the man’s family and herds. In the end, though, the herder’s infant son kills the black monster with a toy axe, "and they live prosperously on the high grassy plains with their sheep and camel herds."
It would be better, of course, to keep a language spoken, but at least people like Harrison are there to record as much as they can before a language dies. For beyond the local knowledge and epic stories they communicate, languages provide us with important linguistic data. Linguistics is not just about the study of syntactic structures and semantic meanings. It also serves as a means of looking into what makes us human.
If linguists are going to study languages, they need many samples, from the fullest possible spectrum. Indeed, endangered languages show the mind at work in unexpected and impressive ways.
In Sora, for instance¯a language with 288,000 speakers in eastern India¯many words can combine into a single word. At first the process might seem like compounding, the process that allows German to call Batman fledermausmann (flying mouse man). Speakers of Sora, however, do not compound their words. Instead, as Harrison puts it, "in Sora, verbs literally swallow other words like pythons, sucking them in by a process linguists call incorporation."
Take popungkoontam , the word Harrison used on The Colbert Report . It’s not just a combination of words but one large verb based on the root verb poo-t . Thus poo¯pung¯koon ¯t¯am , Stab + belly + knife + will + you. Even though Sora has no written form, we know that compounding takes place, not only because two nouns appear inside a verb, but also because poo, koon , and am are all shortened from their original forms.
Onomatopoeia provides another example. In Tuvan, speakers can make up their own onomatopoeia¯and be understood. Pairs of consonants in Tuvan represent different kinds of sounds. So, for example, k and ng represent metallic ringing or impact sounds. High vowels, like the vowels in he and hay , represent rapid or high-pitched sounds, while low vowels, like the vowels in hah or hawk , represent slow or low-pitched sounds. So, Harrison writes, " Kongur is the sound of a big bell ringing or a large metal pipe striking an object. Kingir or küngür would be jangling stirrups or clanging keys, while kangyr might be a giant empty metal barrel rolling along. With eight vowels, Tuvan provides many possible combinations, and speakers can use and understand most of these combinations, even if they have never heard them used before."
Each of these languages tells us something we didn’t know from English, Italian, Czech, or any other common language. Part of it is the local knowledge and the stories they contain. But the larger part is this: They reveal the different ways people can perceive and organize their world. Surely such insights into the human mind are worth some field time with a tape recorder.
Nathaniel Peters is a junior fellow at First Things .
