I'm still reading John McWhorter's The Power of Babel. I haven't spent as much time reading nonfiction this week, but I am enjoying what I read. He does a good job of showing why many assumptions people have about language are simply incorrect, especially relating to any kind of idea of a pure language. One that had me laugh was this quote from a John Cheke in 1561:
Our own tung shold be written cleane and pure, vnmixt and vnmangeled with borrowing of other tunges...
which as McWhorter points out the words "pure" and "mangled"...both French borrowings. This kind of language change (and actually much more extreme borrowings and interminglings) is the rule, not some exception, and is constant. Any snapshot of a language is like a a single glance at a lava lamp, though things like widespread literacy do slow down the rate at which the language changes.

Anyway, one interesting change he noted relates to a grouping of grammatical features common to many European languages. What's interesting is that as far as linguists can tell, proto-Indo European did not have these features, and neither did proto-Uralic (ancestor of Finnish, Hungarian, Estonian). Yet they're prevalent in many of the mainland languages descended from both of these, but not in the known Celtic languages, which were seemingly more isolated for a key stretch of time. Written records of early Germanic languages even show some of these changes spreading over time.

So I'm curious what's going on there? Where did these changes come from? It's clearly, given all the points McWhorter adds to the picture, a matter of grammatical changes spreading from language to language, a continent-wide Sprachbund (meaning a group of languages that influence each other). But McWhorter doesn't offer any explanation of why this happened. Did some of these language groups encounter people who spoke another language, something neither Indo-European nor Uralic, that's now lost? Or perhaps not entirely lost? Basque is an intriguing language isolate...but I've seen enough out-there, conspiracy-tinged theories connected to Basque that I'm hesitant to focus just on that. I'd think it'd have to have been a fairly prominent language at least somewhere, though I suppose it could have spread virus-like even from a less prominent language. Or maybe it was just a random mutation (of sorts) in one language and spread from there.

The epilogue of the book addresses the question of whether we could ever recover the original proto-language, probably spoken some 150,000 years ago (hint: chances aren't good, despite what some researchers claim). I've already read that part, but I'll probably take another look at it and maybe blog about that next week.