Well, my comment was that one of the things occurring to me when observing the (apparently) "Jurchen" elements that you demonstrated in Kitan is that the Mohe/Jurchen speakers to the east were always more populous while also being close. That being the case you would kind of expect an influence --when most of the people in a border region speak one language and a minority speak another, you'd expect the majority language to be more influential unless there is a strong countervaling factor (maybe political). But it seemed to me the influences might really be pre-Jurchen, and that was what interested me about Beckwith's book. Turkic political vocabulary came east for pretty obvious reasons, but strangely Turkic horse vocabulary didn't. I'm sure somebody knowing something about linguistics can probably find evidence for "Tungusic" words going west.  I'm not sure the west/Turkic to east/Tungusic scheme will prove dominant in all genres.