One language dies every two weeks. But where does that statistic come from? Everyone is saying it without explicit citation.

Who or which academic paper first noted the death of a language every two weeks?

1 Answer 1


Short answer, David Crystal (2000), see catdir.loc.gov/catdir/samples/cam032/99053220.pdf‎

Long citation:

A middle position would assert 50% loss in the next 100 years. This is the view independently arrived at by three linguists reported by Krauss in 1992. 34 50% is 3,000 languages. 100 years is 1,200 months. To meet that time frame, at least one language must die, on average, every two weeks or so. This cannot be very far from the truth.

It's a conjecture of how language would die according to guess-gestion of other linguist's statistics not reporting language death.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.