Maybe we statisticians should study a little linguistics?

Mark Liberman has an interesting post about digital linguistics over at Language Log.  I recall that Joe Roy wrote an interesting masters’ thesis at UTSA doing some basic statistical linguistics.  Could be a growth area!