Skip to main content

Preserving Language Diversity: Computers can be a tool for making the survival of languages possible

Preserving Language Diversity: Computers can be a tool for making the. survival of languages possible.

The people of the world speak between 3,000 and 6,000 languages. Of these, 80 to 90 percent are spoken by indigenous peoples, representing almost all linguistic diversity today. A few native-language communities, like the Aymara and the Tswana, are large and robust, but most are small and fragile. Only 276 Languages are spoken by a million or more people.

Languages seem to be disappearing faster than ever before. I estimate that there are about 15 percent fewer languages now than in 1500 A.D. This is alarming in itself, but, just as important, the consequent reduction of cultural diversity may threaten humanity's survival. Our adaptive success as a species - with over 5 billion people in such diverse environments as jungles, deserts, and the Arctic - is due to "culture," implying the communication of ideas through language. Linguistic diversity relates to adptational ideas about property, health care, food, children, power, and disputes. The loss of language diversity diminishes our ability to adapt because it decreases the pool of knowledge from which to draw.

The existence or disappearance of languages has particular political and economic implications for native peoples themselves. Consider language-related politics in India, Belgium, Canada, Lithuania, and Estonia. Cultural uniqueness - ethnicity - reinforces claims to a share of political power, land, jobs, and other resources in heterogeneous states. Language is a powerful force in legitimating those claims.

Conversely, not speaking its own language can hurt a group's claim to special ethnic status. A group of Mexican Indians recently sued a power company that had received funding to build a hydroelectric generator and dam. The lake resulting from the dam would flood thousands of acres of ancestral Indian land. The Indians argued that not only was the company offer to pay for the land inadequate, but also that the land was important to their identity. Company lawyers noted that only a few elders spoke the Indian language, and none of the younger generation were learning it. How could Indians expect special status if they didn't speak their own language?

Computers can help preserve both vanishing native languages and language diversity. First, native peoples can use computers to write in previously non-written languages. Authors seeking readers can help teach their people to read. Second, computers can reduce the cost of publishing in native languages. Books, pamphlets, articles, letters, and so forth can help spread literacy. Third, computers can be a tool for creating dictionaries that may be more extensive than those produced by linguists. A linguist can compile a dictionary of perhaps 5,000 words over a decade. After a two-week seminar with computers, five Kom speakers in Cameroon produced a 2,000word dictionary from a 25,000-word body of literature they wrote in those two weeks.


A few indigenous groups have literary traditions that flourished in the past or flourish now - Zulu, Xhosa, and Luo in Africa; Cherokee and Javajo in Northern America; Quecha and Aymara in South America. However, most native languages lack "popular literacy" - many people reading and writing regularly and authors writing books in their native language. Without popular literacy, all but a few languages will soon disappear.

It's not for lack of writing systems. Virtually every indigenous languages has an alphabet or some other writing system. Most of these have been created by missionaries and linguists over the past 500 years. A great many peoples even have two or three books in their native language; a grammar, a dictionary, a translation of the Bible.

Nor is it for lack of bilingual education programs that teach children to read and write in their first language and the major language of their country. Children typically learn to read primers and occasionally the Bible in their native language. However, they rarely grow up to write books. In most cases, these programs don't produce popular literacy in the native language, and the tribal languages of the world remain largely unwritten.

Moreover, many books go unpublished because the costs of printing are high. One copy of a 200-page book in a Mexican native language may cost 45,000 pesos (about$9) - two days' wages for a rural laborer.

Nevertheless, there is no more potent force for literacy than authors who want others to read their work. Computers can help preserve language diversity by enabling native authors to produce literature in their own languages.

Acting on this reasoning 20 years ago, I taught Jésus Salinas Pedraza, a Ñähñu Indian from Hidalgo State, Mexico, to read and write his own language. We used a modified version of an alphabet developed by missionary linguists some years earlier. Several years later, he authored a description of his culture, which I translated into English, annotated, and typed into a computer. This collaboration resulted in two Ñähñu English books that Salinas and I coauthored, Otomi Folk Tales, Parables, and Jokes and The N!chn!u, an ethnography of the Ñähñu people.

Problems arose with adjusting the Ñähñu alphabet to make it possible to write text on a computer. The alphabet contained several symbols not on a standard computer keyboard. Salinas and other Ñähñu educators didn't like the compromised writing system that I'd concocted, and they balked at the limited keyboard letters available. They insisted technology should serve their needs. They were right, although the wisdom then current in linguistics was to construct writing systems from characters available on standard keyboards.

To answer Salinas's criticism, we switched to an Apple II word processor called Gutenberg. This made it possible to fashion new characters and print them exactly as they appeared on the screen. For example, the letter "e" in Ñähñu stands for the sound of "a" in the English word "cat." In the 1970s we used a "w" for that sound. The "w" is close to the "e" on a standard keyboard, and "w" is not needed elsewhere for writing Ñähñu. Gutenberg and the Apple II let us design an "e" for both the screen and the printer.

With Gutenberg, Salinas could write the characters of Nahnu by 1981. He later wrote a 250,000-word work, Native Ethnography: a Mexican Indian Describes His Culture, which I annotated and translated. It was published in 1989.


As early as 1984, Salinas was so comfortable writing Nahnu on the Apple II that it was possible to consider extending the technology to Mixtec, Tsotsil, and other languages. That led to the establishment in 1987 of the Native Literacy Center to enable Indian people to learn to read and write and publish books in their own languages using computers. Four organizations sponsored the center: the National Directorate for Indian Education (the part of Mexico's Ministry of Education in charge of Indian-Spanish education programs), the Center for Advanced Studies in Social Anthropology (CIESAS), the Interamerican Indian Institute, and the University of Florida Anthropology Department. In 1988, the center opened at the Oaxaca headquarters of CIESAS. Seven language families are represented in the state of Oaxaca, with 16 major Indian languages.

In the first phase of the project, ending in 1991, Indians spent up to three months at the center. Mostly bilingual school teachers, the trainees were literate in Spanish but needed to learn to write in their native languages. They found this easy, once they learned to use a word processor that produced the necessary characters. Mixtecs used a Mixtec-Spanish word processor, Chinantecs used a Chinantec-Spanish word processor, and so on.

The Oaxaca native literacy project has been using Gutenberg and Apple IIs, but it is changing to IBM-compatible equipment, capable of handling all the characters needed. The technology has become relatively inexpensive, and Word Perfect now allows people to define most characters they need for text. As Windows programs become more popular, many more ways to fashion new characters should become available. (Many people have pointed out, of course, that it would be far easier to create and print special characters with a Macintosh than with IBM-compatible machines. The decision to use the IBM-compatible format was entirely a matter of practical convenience: when we began the project, there was no Machintosh repair facility in Oaxaca.)

The Native Literacy Center is now becoming CELIAC, the Centro Editorial en Lenuas Indigenas, A.C., or Indian Language Publishing Center. The A.C. at the end of the name indicates that the center is a non-profit corporation. It can enter into contracts, receive grants, train authors, and publish native-language works.

So far, 52 people speaking a total of 12 Mexican languages have trained at Oaxaca. They have produced works in Nahnu, Mixtec, Mixe, Zapotec, Chinantec, and Mazatec. Books in other languages are in the works. The participants have produced original prose and poetry, ethnographies, and biographies. In each case, these authors have written directly in their own languages.

In the second phase of the project, now underway, the center is editing and publishing these literary works and distributing them to the various language communities. There are plans to expand a few works, translate them into Spanish, and publish bilingual editions for wider distribution. The author/teachers will use their books to help adults and children in their home regions learn to read.

While the computer has not made it possible for most Indians to afford books, it has made it clear that many native people want to write them. The same is proving true in South America. After Salinas and I presented our work to the Society for Applied Anthropology in 1989, Norman Whitten, an anthropologist at the University of Illinois, arranged for Alfonso Chango, a trilingual (Shwara-Quichua-Spanish) teacher from Ecuador, to visit Oaxaca. During his three-month stay at the center, Chango wrote 50 pages on Shwara culture in Quichua, a language spoken by about 5 million Ecuadorans. In 1990, he acquired a computer and began teaching others to read and write Quichua and Shwara.


In July 1988,l Salinas and I demonstrated our project at the Twelfth International Congress of Anthropological and Ethnological Sciences held in Yugoslavia. Paul Nchoji Nkwi, an anthropologist at the University of Yaounde in Cameroon and a member of the Kom people, suggested that the technology might help his people write in their own language, too. About 127,000 people speak Kom and almost all are bilingual in English and Kom. According to Nkwi, many would read Kom books - if any were available.

In 1989, with a grant from the WennerGren Foundation for Anthropological Research, I spent two weeks training five Kom speakers to use a computer and write in their native language. The five participants included two bilingual teachers, a lawyer, an anthropology graduate student, and a Catholic priest. All knew the Kom alphabet developed by German missionaries in the 1880s, but none had written a substantial text in it.

The task was to transfer their skill in writing English to writing Kom, but an immediate problem arose with tones. Kom is a tonal language, with three tones. The Kom speakers had learned to mark tones so others could read their work. However, if people have to mark all the tone, they often don't write at all.

To enable the Kom to write tones on the computer, I programmed a character set that included all the vowels with the tones. However, the system was cumbersome and required learning the placement of more than a dozen extra keys. By the third day, four participants had abandoned the tones, although one teacher persisted in marking tones to the end. He felt it was more correct to do so, according to what he'd been taught by linguists.

In two weeks, the five trainees produced 25,000 words of casual, adult, literate Kom. It was causal in that each participant would turn on the computer in the morning and start typing. Anything they could say, they could write. It was adult, not the simple sentences in school primers. It was literate: flowing Kom text, organized around a theme, went on for pages. For example, the lawyer had recently married and wrote an essay about Kom marriage contracts.

The evidence indicates that Kom people don't need to mark tones in writing. At the end of the training, the group held a ceremony and invited two dozen other Kom. The priest wrote a speech in Kom about there training, and he didn't mark any tones in the text. One of the teachers presented the speech orally, without difficulty, after just one preliminary reading.

Leaving out tone or accent marks is happening in other languages as well. Many contemporary Spanish writers leave out accent marks out of convenience. This may not replace the tradition of accenting, but native Spanish speakers don't need to see accents to read their language fluently. Modern Hebrew is written entirely without marking vowels.

In other words, people can tolerate a great deal of ambiguity in writing. Word processing made it possible to test this in Kom, as it had in Mexico earlier.

Word processing has also helped the Kom people hasten the development of a Kom orthography. Like most non-written languages, Kom has an alphabet but little else in the way of rules for writing, such as for spelling, syllabigication, and punctuation. The size of an orthography depends on the amount of literature in a language. English orthographic rules developed in part out of the large body of English literature.

Just as the Kom participants changed the rules about marking tones, word processing affected other aspects of Kom orthography. For example, the writers developed rules of punctuation because they wanted the computer to produce justified text, which looked attractive to them. Since Kom's long words left big, ugly spaces in justified text, the cure was a clever use of hyphens. The participants worked together to hyphenate text, discussing how to hypenate certain words and whether some words should be hyphenated at all. In some cases, they decided that long words were really composed of two words. The participants thereby squeezed decades of convention building into two weeks.


Teaching people to read primers and Bibles doesn't produce authors; it produces readers. Printing presses and publishing houses produce authors, and teaching a few highly motivated people to write and print their own books can help many people become literate. This is no different today than it was in medieval Europe.

The computer-based ability to publish such books raises a question, however. While few native-language communities can afford computers and desktop publishing systems, the technology is not too expensive for government agencies, development agencies, foundations, missionary groups, and community self-help groups, or even wealthy native individuals. But should such agencies invest in publishing systems for native-language communities? The technology could irrevocably transform a people and culture, and scholars like Etienne Verne warn that insisting on the importance of written literature may lead native people to shun their oral tradition. Walter J. Ong and others respond that people must give up the beauty and power of the oral world to gain the power of literacy.

Both arguments are wrong. Oral traditions remain strong in the most technologically advanced societies, through plays, movies, television, music. And if literacy does fundamentally alter a culture, it is arrogant for Westerners to decide that a particular transformation should or should not happen to preliterate peoples.

Whatever the consequences of literacy, I'd still choose to preserve language diversity over just preserving oral tradition. In Australia, the lack of native-language literacy contributed to reducing the number of Aboriginal languages from 260 to only 40 or 50 in daily use in the 1970s.

This is not to argue that native authors should feel compelled to write in their native languages. Instead, potential authors need the opportunity to write. Computers help create this opportunity, providing many people with the tools to write and print books. While this won't guarantee language survival, it will make it possible. The richness of human knowledge is at stake.


The Native Literacy Center in Oaxaca, Mexico, has become a nonprofit corporation called CELIAC, the Centro Editorial en Lenguas Indígenas, A.C. CELIAC publishes books in Latin America's indigenous languages and trains indigenous people from around Latin America on how to write a publish books in their own languages. Some CELIAC books will be bilingual - in Spanish and the indigenous language of the author.

You can support CELIAC by purchasing the books they produce or by asking your library to do so. Tax-deductible donations can be made to the University of Florida Foundation, Inc. Equipment donations are also welcome.

For more information on how to help or become involved in the project, contact H. Russell Bernard, Dept. of Anthropology, 1350 Turlington Hall, University of Florida, Gainesville, FL 32611; fax: (904)376-8617.

Article copyright Cultural Survival, Inc.

Our website houses close to five decades of content and publishing. Any content older than 10 years is archival and Cultural Survival does not necessarily agree with the content and word choice today.