One problem with arguing with ignorant people is that they don’t usually have indicator lights that tell you exactly how ignorant they are. This matters when you’re trying to explain something: it’s not always clear how much goes without saying.
About the imaginary high levels of gene flow between distant populations in the past: if that had happened, Fst between those populations would be close to zero. It’s not. And when an inevitable consequence of a certain proposition is false, the proposition is false.
But here’s another example: falciparum malaria probably originated in Africa ( it’s closely related to a kind of malaria that infects gorillas). At some point it spread to humans, and it’s a bear – in terms of evolutionary pressures, probably the worst disease in the world.
Many different expensive genetic defenses have arisen in populations exposed to malaria. But those seen in southeast Asia/PNG are, at the molecular level, entirely different from the ones seen in Africa. As far as I know, there is no overlap at all. You see mutations of the same gene showing up in those far-separated populations ( convergence) , but the SNPs are never the same.
The sickle-cell allele would have been advantageous in, say, PNG lowlands, and if even a few copies had ever arrived there, it would have become common rather rapidly. A smidgen of gene flow from sub-Saharan africa , even 20 individuals total, would have left a genetic signature in PNG. [ just as a few copies of Denisovan altitude-tolerance alleles were enough to transform Tibetans] That smidgen would have grown geometrically with time: it would function as a very sensitive detector of African gene flow. But it never happened. Contrariwise, some of the malaria defenses in PNG ( like Melanesian ovalcytosis) would have spread widely in Africa with even a little gene flow. But that never happened either.