That "recipe" is completely unhelpful in explaining the method. It boils down to the author saying "it's a bunch of mumbo-jumbo I don't understand, therefore it

That "recipe" is completely unhelpful in explaining the method. It boils down to the author saying "it's a bunch of mumbo-jumbo I don't understand, therefore it's wrong". But I largely agree that it's a stretch. As far as I understand it, they discovered a correlation between phylogenetic markers in historical populations and the presence of these folktales in their cultures. Which is completely unsurprising, populations that mix genetics will share stories. From this, they link "The Smith and the Devil" to the genetic marker ATU 330 (going back to the Bronze Age), which is the only story they could trace back. > The Bayesian ancestral state reconstructions failed to support the presence of three out of the four tales that were tentatively inferred in Proto-Indo-European (table 2). However, the analyses reconstructed one tale, ATU 330 ‘The Smith and the Devil’, in this corpus, with a posterior probability of 87%. A fossil test returned positive support for the presence of ATU 330 (Bayes Factor 3.59). It's not nothing, it's plausible, but I wouldn't call it "evidence". > [we] find evidence that one tale (‘The Smith and the Devil’) can be traced back to the Bronze Age Even if a story is correlated to a genetic marker, it doesn't mean they originated at the same time. Although it might be possible to infer that the story originated in a certain population before it got mixed with others, maybe. Is that what the paper is saying? I'd need more time with it.