When this came up yesterday, I thought rspeer's comment was the most insightful, at https://news.ycombinator.com/item?id=16241656 .

> "he ran this somewhat anachronistic model and it told him that the text was a good statistical match for being in... Esperanto. He threw this prediction out on the basis that it makes absolutely no sense. You'd think this would be the first sign that the model did not fit the data.

> The #2 prediction was Modern Hebrew ...

The actual ACL article: https://transacl.org/ojs/index.php/tacl/article/download/821...