Are dying languages worth saving?
About 6,000 different languages are spoken around the world. The Foundation for Endangered Languages estimates that between 500 and 1,000 of those are spoken by only a handful of people. Every year the world loses around 25 mother tongues, which works out at 250 languages over a decade.
Put like that, the arithmetic is brutal and the question follows immediately. Are dying languages worth saving, and by whom?
At a conference in Carmarthen, Wales, organised by the foundation and held at Trinity College, about 100 academics gathered to make the case. They did it, pointedly, in nine different languages, and they spent the week discussing indigenous languages in Ireland, China, Australia and Spain.
The argument from knowledge
"Different languages will have their quirks which tell us something about being human," says Nicholas Ostler, the foundation's chairman. "And when languages are lost most of the knowledge that went with them gets lost."
That second sentence is doing more work than it appears to. Ostler is not being sentimental about vocabulary. Small languages carry information that exists nowhere else: which plants treat which ailments, how a reef behaves across a season, where the water is in a bad year, how a kinship system actually functions. This knowledge is rarely written down, because the communities that hold it did not need to write it down. It lives in the language, and it leaves with the last fluent speaker.
There is a scientific argument too. Linguists reconstruct the history of human migration partly from language, and every grammar that vanishes narrows the sample. Theories about what human language can and cannot do are built on the range of languages we know about. A field studying 6,000 systems is a different field from one studying 3,000.
The argument from identity
Ostler's other point is blunter. "People do care about identity as they want to be different," he says. "Nowadays we want access to everything but we don't want to be thought of as no more than people on the other side of the world."
That is the honest version of the case, and it is the one that survives contact with the people actually involved. Language shift is almost never chosen freely. It happens when speaking your own language costs you a job, a school place, a hearing in court, or a beating in a playground. Communities do not casually abandon their language. They are taught, over a generation or two, that it is worth less than the one being offered instead.
What killing a language looks like
Language death is a process, not an event, and it follows a recognisable sequence. Grandparents speak the language fluently. Parents understand it but answer in the dominant language. Children can follow a few phrases. Grandchildren have a handful of words for food and family. Then a newspaper prints an obituary for the last speaker, and the language has been over for thirty years by the time anyone notices.
The Welsh setting of the conference was not accidental. Wales is one of the few places that halted the slide and partially reversed it, through compulsory Welsh in schools, bilingual signage, a broadcaster, and legal status for the language in public administration. It took decades, it cost money, and it worked well enough that Welsh is now the standard reference point for anyone arguing that language revitalization is possible rather than merely admirable.
The counter-argument, taken seriously
The case against is rarely stated openly but it is widely held. Maintaining a language spoken by two hundred people costs real money. That money could buy vaccines, or schools, or roads. A child who learns only a minority language is, in cold terms, locked out of the economy that surrounds them. Is the linguist's interest in a rare grammar worth a family's economic prospects?
The answer that has emerged over thirty years of practice is that the choice is false. Nobody credible proposes raising children monolingual in an endangered language. The goal is bilingualism: keep the heritage language at home and in school alongside the national one. Children handle this easily. It is adults, and ministries, who find it difficult.
The economic framing also cuts both ways. Learning a second language well is genuinely hard work, as anyone who has argued about the hardest language to learn knows. But a child who grows up bilingual has done that work by the age of six without noticing, and nobody has yet shown that keeping the second language costs them anything.
Who does the work
Bodies such as UNESCO, which maintains an atlas of the world's languages in danger, and the Foundation for Endangered Languages itself provide the documentation, the grants and the international pressure. But recordings in an archive are not a living language. Revival happens when a community decides it wants the language back and organises schools, playgroups, radio and social media around that decision.
Linguists can hand a community a grammar. They cannot hand it a reason to use it. That is the uncomfortable finding underneath the whole field, and it is why the delegates at Carmarthen spent the week talking to each other in nine languages rather than one.
Twenty-five languages will go this year. The question the conference was really asking is how many of the remaining 6,000 anyone intends to fight for.