In some ways, it is surprising that languages change. After all, they are passed down through the generations reliably enough for parents and children to communicate with each other. Yet linguists find that all languages change over time—albeit at different rates. For example, while Japanese has changed relatively little over 1,000 years, English evolved rapidly in just a few centuries. Many present-day speakers find Shakespeare’s sixteenth century texts difficult and Chaucer’s fourteenth century Canterbury Tales nearly impossible to read.

Why They Change

Languages change for a variety of reasons. Large-scale shifts often occur in response to social, economic and political pressures. History records many examples of language change fueled by invasions, colonization and migration. Even without these kinds of influences, a language can change dramatically if enough users alter the way they speak it.

Frequently, the needs of speakers drive language change. New technologies, industries, products and experiences simply require new words. Plastic, cell phones and the Internet didn’t exist in Shakespeare’s time, for example. By using new and emerging terms, we all drive language change. But the unique way that individuals speak also fuels language change. That’s because no two individuals use a language in exactly the same way. The vocabulary and phrases people use depend on where they live, their age, education level, social status and other factors. Through our interactions, we pick up new words and sayings and integrate them into our speech. Teens and young adults for example, often use different words and phrases from their parents. Some of them spread through the population and slowly change the language.

Types of Change

Three main aspects of language change over time: vocabulary, sentence structure and pronunciations. Vocabulary can change quickly as new words are borrowed from other languages, or as words get combined or shortened. Some words are even created by mistake. As noted in the Linguistic Society of America's publication Is English Changing?, pea is one such example. Up until about 400 years ago, pease referred to either a single pea or many peas. At some point, people mistakenly assumed that the word pease was the plural form of pea, and a new word was born. While vocabulary can change quickly, sentence structure—the order of words in a sentence—changes more slowly. Yet it’s clear that today’s English speakers construct sentences very differently from Chaucer and Shakespeare’s contemporaries (see illustration above). Changes in sound are somewhat harder to document, but at least as interesting. For example, during the so-called “Great Vowel Shift” 500 years ago, English speakers modified their vowel pronunciation dramatically. This shift represents the biggest difference between the pronunciations of so called Middle and Modern English (see audio clips in "")

Agents of Change

Before a language can change, speakers must adopt new words, sentence structures and sounds, spread them through the community and transmit them to the next generation. According to many linguists—including David Lightfoot, NSF assistant director for social, behavioral and economic sciences—children serve as agents for language change when, in the process of learning the language of previous generations, they internalize it differently and propagate a different variation of that language.

Linguists study language change by addressing questions such as these:

• Can we trace the evolutionary path of a language?
• How do language changes spread through communities?
• How do historical circumstances influence language change?
• What is the relationship between language learning and change?

                    ENDANGERED LANGUAGES

As "globalization" increases, so does the loss of human languages. People find it easier to conduct business and communicate with those outside their own culture if they speak more widely used languages like Chinese, Hindi, English, Spanish or Russian. Children are not being educated in languages spoken by a limited number of people. As fewer people use local languages, they gradually die out.

Why It Matters

At least 3,000 of the world’s 6,000-7,000 languages (about 50 percent) are about to be lost. Why should we care? Here are several reasons.

• The enormous variety of these languages represents a vast, largely unmapped terrain on which linguists, cognitive scientists and philosophers can chart the full capabilities—and limits—of the human mind.
• Each endangered language embodies unique local knowledge of the cultures and natural systems in the region in which it is spoken.
• These languages are among our few sources of evidence for understanding human history.

Other Implications

Those who primarily speak one of the world’s major languages may find it hard to understand what losing one’s language can mean--and may even feel that the world would be better off if everyone spoke the same language. In fact, the requirement to speak one language is often associated with violence. Repressive governments forbid certain languages and cultural customs as a form of control. And conquered people resist assimilation by speaking their own languages and practicing their own customs.

On the positive side, one language can enrich another—for example, by providing words and concepts not available in the other language. Most languages (including English) have borrowed words of all kinds. Learning another language often brings an appreciation of other cultures and people.

The study of endangered languages also has implications for cognitive science because languages help illuminate how the brain functions and how we learn. “We want to know what the diversity of languages tells us about the ways the brain stores and communicates experience,” says Peg Barratt, NSF division director for behavioral and cognitive sciences. “Our focus is not just on recording examples of languages that are soon to disappear, but on understanding the grammars, vocabularies and structures of these languages.”

Preserving While Documenting

Documentation is the key to preserving endangered languages. Linguists are trying to document as many as they can by describing grammars and structural features, by recording spoken language and by using computers to store this information for study by scholars. Many endangered languages are only spoken; no written texts exist. So it is important to act quickly in order to capture them before they go extinct.

To help preserve endangered languages, (Electronic Metastructure for Endangered Language Data) aims to boost documentation by:

• duplicating and digitizing high-quality recordings in an archival form;
• emphasizing self-documenting and software-independent data;
• giving linguists a toolkit to analyze and compare languages;
• developing a General Ontology for Linguistic Description (GOLD) to allow interoperability of archives, and comparability of data and analysis.

In another kind of archiving, Joel Sherzer, Anthony Woodbury and Mark McFarland (University of Texas at Austin) are ensuring that Latin America's endangered languages are documented through (AILLA). This Web-accessible database of audio and textual data features naturally–occurring discourse such as narratives, ceremonies, speeches, songs, poems and conversation. Using their Web browsers, scholars, students and indigenous people can access the database, search and browse the contents and download files using free software.

Documentation is the right thing to do for both cultural and scientific reasons. According to NSF program director Joan Maling, we must explore as many different languages as we can to fully understand this uniquely human capacity—"Language" with a capital L. “Just as biologists can learn only from looking at many different organisms, so linguists and language scientists can learn only from studying many different human languages,” she says. “Preserving linguistic diversity through documentation is critical to the scientific study of language.”