Statistical analysis, two findings, one potentially new

[mjtemkin]

I honestly don't have a hypothesis yet on the word doubling at all (well, a hypothesis other than what's been tested, but everything has turned out to be false). And this relates to the nulls as well- I don't have a good feel for how the nulls arise while still showing a positive result in the grammatical ordering. This is what I'm planning to dig into next (though as I thought more about texts to compare with, I realized some of the Hebrew mystical texts like Sefer Yetzirah could be interesting as another test of the doubling pattern, as could be contemporaneous mathematical/technical texts). I don't have a theory yet about the conflict between the nulls and the grammatical structure but I'm going to keep poking. 

Let me know if you have any ideas!