Character Classes

[MarcoP]

Sean Palmer produced a study along similar lines a few years ago You are not allowed to view links. Register or Login to view., showing glyph affinity in general.

Thank you, David! The colored matrix is very interesting.
As the author wrote:

"Comes before" and "comes after" are measured in a binary way, meaning that distance and number of occurrences are ignored. So for example "ab" ranks as "b-after-a" just as much as "acccbbb".

So this is all about the position of characters inside words: prefix characters produce blue horizontal lines; suffix characters produce red horizontal lines.

This analysis seems to me to be mostly relevant as a morphological investigation, while I think that Emma's character classes are most likely due to phonetics. But of course there could also be a significant overlap between morphology and phonetics: I guess that morphology is constrained by what is phonetically possible.