Eyak may not be long alone. Children are still learning only two of the twenty native Alaskan languages÷Central Alaskan Yupik Eskimo and Siberian Yupik Eskimo of St. Lawrence Island. Of North American Indian languages, Iowa and Osage have five fluent speakers each, Mandan six, and Abenaki-Penobscot twenty. According to 1977 counts, Menominee had fewer than 50 speakers, Tuscarora fewer than 30, Coeur dâAlene fewer than 20, and Yokuts fewer than 10. This sad litany is not unique to North America. Sirenikski Eskimo has two speakers; Ainu, formerly spoken on Hakkaido, Sakhalin, and the Kurile Islands, may be extinct. Until linguists discovered, in 1989, an elderly woman in Turkey who spoke Ubykh, the northwest Caucasian language with more consonants (more than 80) than any other language, Mr. Tevfik Esen of Turkey was thought to be the last speaker of Ubykh.

Looked at by continent, the threat to languages takes on catastrophic proportions. In North America, children are no longer learning 149, or 80 percent, of 187 native languages. About 50 of 300, or 17 percent, of the indigenous languages of Central America, and 110 of 400, or 27 percent, of South American languages may be moribund, meaning that unless children begin speaking them, they will vanish when the current speakers die. Thus, in the Western Hemisphere, unless drastic measures are taken to save them, 300 of 900 languages will soon disappear forever.

In the Eastern Hemisphere, the picture is equally bleak. Australia faces the greatest language loss, with 90 percent of 250 aboriginal languages still spoken now moribund, most very near extinction. In lands where it is the dominant language, then, English assumes exclusive dominance, wiping out an average of nine out of ten native languages. Russian hegemony, by contrast, has reached this level only among the northern Russian minorities; in Russia, 45 of 65 indigenous languages, or 70 percent, are moribund, while in the former Soviet Union as a whole, the total is closer to 50 percent.

How many of the worldâs roughly 6,000 languages, then, are likely to go extinct in the next century? According to Ethnologue: Languages of the World, edited by Barbara Grimes (Summer Institute of Linguistics, 1988), the Americas contain 900 (15 percent), and Europe and the Middle East 275 (5 percent) of the worldâs languages. The other 90 percent lie in Africa (1,900) and in Asia and the Pacific (3,000). The two countries with the largest number of languages are Papua New Guinea with 850 and Indonesia with 670, followed by Nigeria (410), India (380), Cameroon (270), Australia (250), Mexico (240), Zaire (210), and Brazil (210). Based on political conditions in these countries, and on the factors that have led to the present linguistic mortality÷cultural genocide, habitat destruction, ethnic displacement, language suppression, and electronic media bombardment÷I estimate that, at the present rates, 3,000 languages will become extinct by the end of the next century.

In addition to these, numerous other languages, if present conditions persist, will cease to be learned by children sometime in the next century. The number of these endangered tongues is even more difficult to calculate; it is easier to count those languages that are neither moribund nor endangered. Two obvious factors indicate a language that is free from danger: official state support, which is usually nonexistent for minority languages, and large numbers of speakers. A language with fewer than a million speakers may be safe in a country where that language is dominant. Icelandic, for example, with 250,000 speakers, is in no danger of disappearing. But even a million speakers does not make a minority language invulnerable. Consider Breton, which within living memory had perhaps a million speakers. Today, it is struggling for survival after decades of French government suppression and neglect. As late as the 1960s, the Navajo, with some 200,000 members the largest Native American tribe in the United States, still taught 90 percent of their children Navajo as a first language. But after a decades-long policy of the U.S. Bureau of Indian Affairs to impose English as the first language, today, according to some reports, the first language of most Navajo children is English.

The median number of speakers for languages of the world, however, is only 5,000 to 6,000. With this and the factors leading to linguistic morbidity in mind, I estimate that the coming century will likely see either the death or the terminal stage of 90 percent of humankindâs languages.

Language extinction is nothing new. It is impossible to calculate how many languages have gone extinct since humans first started speaking, because unwritten languages donât leave fossil records. But the figure is in the thousands, even if the number of languages historically known to have become extinct may only be in the hundreds. Like biological extinctions, language extinctions are occurring faster today than at any other time in human history. Languages, like some biological species, however, may not vanish entirely, but evolve. Latin, for instance, evolved into French, Italian, Romanian, and other Romance languages. One language, on the other hand, though in one sense extinct, has been revived as a natively spoken language. In the late 19th century, the Jewish scholar Eliezer Ben Yehudah led the effort to revive Hebrew, allowing only Hebrew speakers into his home and raising his son as the first native Hebrew speaker in at least 1,500 years.

One reason why Hebrewâs revival was successful was that it gave a sense of identity to the Jewish people who otherwise had no common language. Language is identity. Though there must be many descendants of, say, Etruscans in Italy and Hittites in Turkey, they do not think of themselves as Etruscans or Hittites because those languages died out and were replaced by what are now Italian and Turkish. Even though there is no Basque king or Basque state, strong ties bind speakers of Basque. As long as a people speak a language apart, they will maintain their identity apart. If they do not, long-term survival of that identity is÷at best÷in question. If their language disappears, in the long run, what, except for their facial features, will set the Passamaquoddy Indians of Maine apart from other Americans?

Saving Passamaquoddy is not only good for Passamaquoddies, but for speakers of any tongue. For Passamaquoddy, as any language, is the supreme achievement of a uniquely human collective genius, as divine and unfathomable a mystery as a living organism. Anyone who speaks more than one language knows that each language contains not only words and phrases that are untranslatable, but entire concepts that are unique to that tongue. Might not Eyak and Ubykh contain concepts as valuable as the Chinese notion of aligned meridians in the body (which underlies acupuncture) or as spiritually unique as the Australian Aboriginal notion of the dreamtime? Lose a language and you lose knowledge that that language was invented to express. If English, with its huge vocabulary and myriad ways of expressing myriad ideas, seems sophisticated, it is in part because English is so well traveled and has borrowed so extensively from other languages (kangaroo from Australians, kayak from the Inuit, quinine from Quechua, champagne from the French).

Englishâs cosmopolitan nature has allowed it to become arguably the language of international commerce; if you donât speak it, youâre in danger of being left out of the trade loop. Yet its dominance should force speakers of minority languages neither÷at one extreme÷to abandon their native tongue for English nor÷at the other÷to speak only their native language. Rather, native groups might, for instance, do as the Danes: learn and use English as a second language, with no thought of relinquishing Danish.

Proper documentation is essential to language preservation. To document a language, linguists work with native speakers to make an analysis of the sounds of the spoken language and from that devise a writing system. From this work, which is enhanced by the use of audio- and videotape, linguists (who ideally are native speakers themselves) can then create a grammar and dictionary as well as take down traditional stories, legends, and accounts of everyday life. This corpus of texts can then be preserved in archives, such as that of the Alaska Native Language Center. With careful documentation, even a language no longer spoken as a first language by anyone can play a limited but crucial role in native society÷in schools or in ceremonial life, for example. Furthermore, even after the last native speaker has died, any adequately documented language, as in the case of Hebrew, theoretically could be used again.

There is more to saving a language than documenting it, however, just as there is more the saving an endangered species than taxonomically describing it. There are hundreds of environmental organizations, from international giants to special-interest groups like Friends of the Sea Otter. By contrast, there is so far no broad language survival movement, nor a World Language Fund, much less a widespread Friends of Passamaquoddy. Linguists would do well to take a leaf from the book of biologists and conservationists, who have honed the techniques of environmental monitoring, publicity, and activism on local, regional, national, and international scales. Beyond educating the public, linguists must also try through education, cultural, and political means to increase endangered languagesâ chances for survival. This means working with community members to develop teaching materials, to promote literature and language development in appropriate domains (including television), and to work with agencies, and, where possible, governments, to launch supportive language planning.

The best antidote to language morbidity is for native speakers to nurse their own languages back to health. In Brittany, the Diwan School movement has begun to teach children Breton at an early age. Kohanga reo (ălanguage nestsä) in New Zealand have been so successful in reviving Maori that they have spread to Hawaiâi for restoring Hawaiian. The language revival movement for Irish Gaelic is relatively well known. Even Cornish, after going extinct in the late 18th century, was reconstituted in this century from the texts of 14th-century miracle plays. Today, reportedly, there are several young native speakers in Cornwall.