Unless I'm missing something, I don't think non- is a Latin root in the first place? > I suspect purification attempts will produce better results for language

Unless I'm missing something, I don't think non- is a Latin root in the first place? > I suspect purification attempts will produce better results for languages that have a less diverse, more linear (and perhaps shorter, in some sense) history and are spoken by a relatively small number of people concentrated in a particular area. Ah, but they're only (or at least more) interesting on languages that are already a mess of influences to contrast with.