Funny - since I almost cited the article that (I think) you have in mind. The article I remember wasn't an actual proof though: it was a reductio ad absurdum, u

Funny - since I almost cited the article that (I think) you have in mind. The article I remember wasn't an actual proof though: it was a reductio ad absurdum, used in a review of a book about Homer that used statistical methods. The reductio was, in a nutshell, to show that by the same argument as the one in the book on Homer, the reviewer could prove that Milton didn't write Paradise Lost. In any case, here's a JSTOR reference to the article I'm thinking of: http://www.jstor.org/stable/641985 http://www.jstor.org/stable/641985. (I can provide a pdf copy if anyone wants it. Check my profile for email.) I wonder if this is the one you're remembering.