Do f58r/f58v show Currier languages evolving?

[ReneZ]

These two marks tend to align most with the pharmaceutical text (yellow crosses) [...]

Hi Rene,
unless I messed up something in the top plot You are not allowed to view links. Register or Login to view., the bigram that makes f58 different from pharma is 'eo'. That bigram is typical of pharmacese, where it is three times more frequent than in f58 (~6% vs ~2%).

That is right, of course, and this bigram was not part of the first plots.
In the PCA diagram, the two pages end up near the edge of the pharma cloud.

It seems as if one of the reasons why the two pages are different is the frequency of "qo", but that is perhaps also not the full story.