For best experience please turn on javascript and use a modern browser!
You are using a browser that is no longer supported by Microsoft. Please upgrade your browser. The site may not present itself correctly if you continue browsing.

Just more than half of now living languages – about 3660 – are currently threatened, endangered, moribund or nearly extinct. This is the finding of a new study by an international team of linguists, which includes Frank Seifart from the University of Amsterdam (UvA), on the progress made over the last 25 years in documenting and describing endangered languages. The linguists call for research on endangered language to be intensified. The study is published in the December edition of the journal Language.

Credits: FlickrCC

The study presents the most reliable figures on worldwide languages endangerment to date. More than half of the close to 7,000 now living languages are currently endangered. Around 600 of these are already nearly extinct and are now only spoken occasionally by members of the grandparent generation. About 950 endangered languages are still spoken by children, but the proportion of children acquiring these languages is getting smaller and smaller. Seifart and his fellow researchers warn that if this trend is not reversed, these languages will also die out.


The study is a follow-up to the seminal multi-authored article ‘Endangered Languages: On Endangered Languages and the Safeguarding of Diversity’, coordinated by Ken Hale, which appeared in 1992 in Language. To estimate the current dangers of language loss, the researchers unified data from three major sources on language endangerment: the Catalogue of Endangered Languages (ELCat), UNESCO’s Atlas of the World’s Languages in Danger and Ethnologue: Languages of the World. The article will appear just before the start of UNESCO’S International Year of Indigenous Languages 2019. 

Progress and shortfalls

As far our knowledge on endangered languages is concerned, the researchers see both significant progress and critical shortfalls. With a growing network of researchers carrying out language documentation around the world, and helped by technological progress for data collection, processing and archiving, our scientific knowledge of the world’s languages has significantly increased over the past 25 years. So has the engagement of indigenous researchers in their own languages. Over this period, many hundreds of languages have been documented in sustainably archived audio and video collections, as well as more traditional products like grammars and dictionaries. ‘But our study also shows that well over a third of the world’s languages, including over 1,400 endangered languages, are still severely under-described’, says Seifart. ‘Even basic information on their grammar and lexicon are lacking, let alone proper documentation of culture-specific language use.’ 

Major effort needed

According to Seifart, it is high time to sound the alarm bell. ‘The potential loss if linguists do not up their game is enormous on all accounts. The documentation of linguistic diversity keeps turning up new phenomena and there are no signs that new discoveries are tailing off. These discoveries keep driving linguistics to broaden its canon of possible grammatical categories. Whole new meaning domains have been discovered, and entirely new speech sounds are also still being brought to light. Beyond such core categories of linguistic structure, work with little-studied languages is expanding our knowledge of how language is learned, processed, socially organised, aesthetically extended, and how it evolves, within as little as one generation. There are thus many reasons to intensify research on small and often endangered languages.’

Publication details

Frank Seifart, Nicholas Evans, Harald Hammarström & Stephen C. Levinson: ‘Language documentation twenty-five years on’, in: Language (december 2018). 

Pre-print version: