Learning an Endangered Language (Part 3)

Part 1 | Part 2

In this installment I want to discuss more about what it means for a language to be called “endangered.” In doing so I will draw on David Crystal’s book Language Death. The picture below is from National Geographic’s Enduring Voices Project:

Enduring Voices Project, Endangered Languages, Map, Facts, Photos, Videos -- National Geographic

There are some reasons for anthropologists to be skeptical about the various discourses surrounding language endangerment (for a good discussion of those see Joseph Errington’s excellent article, “Getting Language Rights: The Rhetorics of Language Endangerment and Loss” as well as my AN editorial, “The metaphor of ‘endangered languages'” [PDF]), but I’ll be putting that aside for this post, focusing instead on some of the issues involved in trying to determine “threat levels” for endangered languages.

When is a language a dead language? When the language can no longer be used for daily conversation. By this definition, having one living speaker is not enough, because she will have no one to speak to. There are some complicating factors, such as languages which have literate or religious traditions, but its good enough for our purposes.

Where things get difficult is determining the other end of the equation. How many speakers does a language need to be “safe”? Is 20 speakers enough? 200? 2,000? 200,000? Unfortunately, as Crystal points out, there are too many contextual (linguistic, social, political, economic, etc.) factors to be able to answer the question with a simple number. In a stable, rural setting with little migration in or out of the community, it might be enough to have 500 speakers (barring wars or natural catastrophes). On the other hand, in a highly urbanized environment where members of the linguistic community are scattered throughout the city even 50,000 speakers might not be enough to keep a language alive if it is not taught in the schools or used on the radio and television.

Crystal provides the following threat levels for endangered languages:

  • Potentially endangered: Fewer children are learning the language.
  • Endangered: The youngest good speakers are young adults, few or no children are learning the language.
  • Seriously Endangered: Youngest good speakers are 50 years old or older.
  • Moribund: Only a handful of very old speakers.
  • Extinct: No speakers

This chart is useful for the discussion I started in parts 1 and 2 of this series, as it can help us understand when, exactly, we are dealing with an endangered language. Some anthropologists work in rural, isolated, communities which lack a written language, but these languages are not (yet) properly called endangered languages since the linguistic community remains fairly stable. Others might be working with large populations in urban areas and yet encounter a language that is “seriously endangered” or even “moribund.” In the former situation it might be feasible for the anthropologist to learn the language herself after an extended period in the field (although a translator might be required for comparative work in the next village over). In the latter case, however, the need or ability of the anthropologist to learn the language might be entirely dependent on the age of the population she is working with.