Dalton Luther reflects on the Denisovan admixture paper [1] that I wrote about earlier this week ("How widespread is Denisovan ancestry today?"), by referring to John Moore's work on ethnogenesis [2].
Getting back to the original quote about Denisovan legacies, just because the Denisovans aren’t “around” anymore, doesn’t mean they’re not “around.” An ancient population is present even though in a very different form. Using the braided river metaphor, the name Denisovan refers to the contents of a particular stream that mixed back into another stream, which grew larger, amplifying its original contents.
What seems to be the challenging concept to some geneticists is that some people today have that legacy and others don't. But it's not at all unusual for that to be true of families, kindreds, cultural traits, or even languages. So why should it be unusual for populations?