July 25, 2026 by languagehat

Frequent commenter Y wrote me as follows:

Two interesting bits. I don’t know if they are blog material but I hope you’ll find them interesting in any case:

— First one comes from Pystynen’s Xitter (I don’t have an account there. I use nitter to browse it.) […]:

Has there ever been work in something like documentary linguistics that was not downstream of the European missionary–colonial tradition? That is, someone producing a description/grammar at some detail of a previously unwritten language foreign to the writer?

We find occasional snippets about foreigners’ languages for long time (Greeks writing down a handful of words of Scythian, etc.), but is going beyond that basically single-invention?

Al-Kashgari comes sorta close (at least a dialect documentation pioneer), but he is both a native speaker and this is early enough that presumably most Common Turkic varieties would have been still intelligible with the centuries-extant Old Turkic literary language.

I was curious about that, and after poking around a bit, found this. More specifically, the first Manchu grammar was written by Shen Jiliang, of whom you can read here. He seems to have been the sort of person Pystynen was looking for, a language enthusiast at a time and place when that was unusual.

— Second one is an article in Science, still firewalled. But somehow I was able to get it through this link, from the first author’s LinkedIn. The gist of the article is that there was a peak in global language diversity around 2000 years ago. There are details to quibble about, but it’s an interesting thought experiment.

You can learn more about it from the accessible supplementary materials; also from an older article it’s based on, and from secondary coverage:
The lost golden age of language diversity
Study uncovers lost ‘golden age’ of languages

Thanks, Y!