Answering that would take a large team of people scanning old sources like local newspapers. It's the sort of thing dialect-dictionary compilers look at, and the results can be diagrammed as "isoglosses" - lines on the map that mark the boundaries between areas where one word rather than another is used. (Most of the time, they don't correlate very well with other words - the idea of "dialect" is much less definite than most people think). One of the more surprising maps like that I've seen is in Laurence Picken's book on the folk instruments of Turkey, mapping the preferences between the words "darbuka" and "dumbek" to describe the same kind of goblet-shaped drum. It looked like a chessboard. There was no large area where one word predominated, each province was different from its neighbours. I'd guess that the words in that list from Google will be distributed in an even more chaotic arrangement, down to the level of words preferred by particular families.
|