Why Do Languages Die?

The mechanism is language shift. Speakers of a smaller language increasingly use a larger one because it offers access to work, education, government and media, and eventually stop passing their original language to children.

The classic pattern takes three generations. Grandparents speak the heritage language, parents are bilingual, and children are raised in the dominant language, understanding but not speaking their grandparents' tongue. Once transmission to children stops, the outcome is largely determined.

Several forces drive shift, and they usually combine:

• •Economic pressure, where jobs require the dominant language
• •Education systems that exclude or forbid minority languages
• •Media and entertainment overwhelmingly in a few large languages
• •Migration away from communities where the language is spoken
• •Deliberate suppression, including policies that punished children for speaking their language
• •Stigma, where speakers are treated as backward or unsophisticated

There are around seven thousand languages spoken today, but they are distributed extraordinarily unevenly. A small number have hundreds of millions of speakers while roughly half have fewer than ten thousand.

Estimates suggest a language stops being spoken every few weeks, and many linguists project that a large share of current languages will have no speakers by the end of the century. Documentation efforts are racing against that timeline.

The obvious loss is cultural: oral literature, songs, humour, ways of describing relationships and the environment that may have no equivalent elsewhere.

There is also a scientific loss. Each language is evidence about what human language can do, and grammatical features found in only a handful of languages carry disproportionate weight in linguistic theory. Detailed ecological and medicinal knowledge encoded in local languages has also been documented as disappearing along with them.

Language death is not inevitable, and there are real successes. Hebrew was revived from a liturgical and literary language into an everyday spoken one with millions of native speakers, which is the most complete example.

Welsh has substantially increased speaker numbers through education, broadcasting and official status. Maori, Hawaiian and several Indigenous languages have benefited from immersion schools where young children are taught entirely in the language. What these efforts share is a focus on children and on giving the language genuine everyday use rather than ceremonial status alone.

When revival is not possible, documentation preserves what can be preserved. Modern language documentation aims for far more than a word list.

A thorough record typically includes audio and video recordings of natural conversation, a grammatical description, a dictionary, and transcribed texts including stories, songs and everyday speech. Best practice now involves community members as researchers rather than as subjects, with the resulting archive accessible to the community itself. That shift matters, since earlier documentation often ended up in distant universities and was effectively unavailable to the people it came from.

Calling a language endangered is not a guess based only on speaker numbers. Linguists use scales that ask who is still learning the language and where it is used. UNESCO's framework, for example, runs from vulnerable, where most children speak the language but perhaps only at home, through definitely and severely endangered, to critically endangered, where the youngest speakers are elderly and use it rarely. The central test is always whether children are acquiring it as a first language. A small but stable community can be safer than a larger one where transmission has stopped.

Many communities also object to the word dead for a language that has no fluent speakers but is well recorded. They prefer terms such as sleeping or dormant, because the language can wake again. The Wampanoag language of Massachusetts is a well known case. It had no native speakers for over a century, but seventeenth-century documents written in it, including a translation of the Bible, allowed community members working with linguists to reconstruct its grammar and vocabulary. Children are now learning it again, which shows why careful records matter even when the last speakers are gone.

People sometimes call Latin a dead language, but its story is quite different from the one described above. Latin was never abandoned in favour of another language. Instead, as spoken Latin changed gradually across different regions over many centuries, it developed into Italian, Spanish, French, Portuguese, Romanian and other Romance languages. Each generation spoke much like its parents, yet the accumulated changes eventually produced languages that are not mutually understandable. Linguists see this as continuous change rather than death.

True language death breaks the chain of transmission. Cornish, the Celtic language of Cornwall, shows the contrast. Over several centuries its speakers shifted to English, and by the end of the eighteenth century it had essentially stopped being used as a community language, though fragments and phrases lingered. In the twentieth century revivalists used older texts and records to bring it back into use, and it now has a small community of learners and speakers, including some children raised with it. Its path is shift, sleep and revival, not gradual transformation.

In its final stages a language often changes in recognisable ways. The linguist Nancy Dorian studied the Scottish Gaelic of fishing communities in East Sutherland and described what she called semi-speakers. These were usually younger people who understood the language well and took part in conversations, but spoke it with simplified grammar. They regularised irregular forms, dropped some distinctions and borrowed freely from English. Interestingly, many were skilled at the social side of talk, knowing when to speak and how to respond, which kept them inside the community of speakers. A language at this stage can be heard every day while losing complexity with each generation.

The very end is the last speaker, a situation lonelier than it sounds. Ned Maddrell, who died in 1974, is usually described as the last native speaker of Manx, the Gaelic language of the Isle of Man. For years before his death, native speakers had been few and scattered, so they rarely had anyone to talk to in their own language. That is why the date of a last speaker's death can mislead. A language usually stops working as a community's everyday tool decades earlier, when it is no longer the natural choice between neighbours, friends and family at home.

Discussions of language death often blame English, but in most of the world the language replacing a small one is a national or regional giant. Indigenous languages of Latin America mostly give way to Spanish or Portuguese, small languages of Siberia to Russian, many minority languages in China to Mandarin, and local languages across Indonesia's islands to Indonesian. Even a language with millions of speakers can begin to slip if city parents raise children in the national language for the sake of schooling and work. The pressure comes from whichever language opens doors locally, not simply from the most widely spoken one on Earth.

Paraguay shows that shift is not inevitable even under strong pressure. Guarani, an Indigenous language, is spoken by a large majority of Paraguayans, most of whom do not identify as Indigenous, and the 1992 constitution made it an official language alongside Spanish. For centuries the two have coexisted, with Spanish dominant in government and formal writing and Guarani strong at home, in rural areas and in casual talk. Many speakers mix them freely in a blend called Jopara. Linguists study stable bilingualism like this to understand why some communities keep two languages for generations while others drop one within three.

Languages die when speakers shift to a dominant language under economic and social pressure, typically within three generations once transmission to children stops, and successful revivals have all worked by restoring the language to children in daily use.