I have got to read Clive Finlayson's book. If this bit about people drifting to Australia is typical, it should be funny as hell.

1. There are areas that are unclear, but Neanderthals were clearly in Europe, the Middle East, and southern Siberia. 2. Nothing like a campaign or war, those are late development. Little bands must have sometimes fought. Our side won more often than not, due to something or other: smarts, numbers, who can say for sure ? 3. Last was about 28k years ago in southern Spain, but dating revisions might change that a little either way.

1. So far, Melanesians only. More exactly, I would say Australoids only: Australian Aborigines must have it too. 2. I suspect so, the authors probably disagree. 3. There should be, but there isn't. I think it's clear that Neanderthals contributed some ancestry to everyone outside of Africa, but some disagree. I haven't heard any argument supporting that disagreement that makes sense to me. 4. How's that Stamford bridge thing working out for you?

There has for a long time been a suspicion that Australoids had erectus admixture. I've also seen funny genetic anomalies that are probably due to this. There were further hints this year. Long, looking at microsatellites, found evidence for one admixture that showed up in all Eurasians and another that showed up only in Melanesians. Moreover, Linda Vigilant (from Max Planck) found Long's work interesting and said that it fit certain patterns they had seen in Melanesians. Later, in the fall, I noticed the clues in Table S48. I thought that the Denisova sample might be from the same population (from Occam's razor), but was somewhat discouraged from this when Paabo said the Denisova pinkie was Neanderthal, as recently as two weeks ago. As for ancient population substructure in Africa - the idea that it explained the evidence of Neanderthal admixture was silly. The idea that it might explain Denisovan admixture in New Guinea is the turducken of silly.

Franz Weidenreich certainly talked about a special connection between Australians and archaic hominid fossils in Indonesia in the 1940s, maybe earlier (I haven't read that much of his work). Herman Klaatsch talked about it as far back as 1908. I think it's pretty obvious if you compare the skulls. In recent decades, it became generally unobvious, for some reason.

Most of the 70 new mutations per individual don't affect functional regions of the genome: by definition all of the ~200 deleterious mutations in your model do. Most must be only mildly deleterious and fairly old. If you could spell-check the genome, errors would creep back in, but it would take many generations before the mutation burden approached the original level. So, not obviously futile.

You expect higher failure due to mutational load in extremely complex systems, ones that depend on the action of many genes: thus, you see a higher frequency of genetic retardation than you do, say, genetic albinism. Such a syndrome would be caused by mutations in many genes: we know of more than 100 that can cause congenital deafness. Of the many mutations causing such a grab-bag mutational load syndrome, some will be syndromic - i.e. will have other distinctive phenotypic effects, such as the white streak of hair in Waardenburg syndrome deafness, or macroorchidism in fragile X. In the case of congenital deafness, about 15% of causal mutations are syndromic. Is there any reason to suspect that sexual orientation is extraordinarily complex? No. Birds do it. Bees do it. Do we see syndromic forms of homosexuality in homosexual men? No. In principle they may exist, but they must then be very rare.

About 1 in 1500 kids are born deaf: about 1 in 3000 are born blind. Why so low? I mean, it would hardly be surprising if those numbers were 1 in 30, right? Just as no one would be surprised if a few percent of birds flew north - or east - for the winter: odd that this isn't the case.

Yes, that paper, but the candidate source group was homo erectus. I suggested that there might be a Neanderthal-derived host cell line infection in humans: more plausible now, since we clearly had intimate contact with Neanderthals. On that same note, I would suggest that Chris Stringer's notion that Neanderthals died out just before anatomically modern humans showed up seems less likely in view of the recent results on admixture, if we discount necrophilia. I am reminded of the paleontologists who have suggested that the dinosaurs died out just before the asteroid hit, and those have argued that the American megafauna died out just before the paleoIndians showed up. There is some common thread to those arguments - and I don't believe that it's Occam's razor.

My guess is that he never said anything about a relict Neanderthal population in Europe - but what would I know?

Of course some populations might have been descended from the first erectus expansion, about 2 million years ago, and thus genetically distant from both AMH and Neanderthals. As for the technical difficulties with degraded ancient DNA, we're getting somewhat better at it.

The ~2% Neanderthal mixture may be the same over Eurasia, but that number applies to the genome as a whole - mostly neutral. The action is in the coding and regulatory area, a small fraction of the overall - there we don't know the impact of admixture. As for none of the skin color genes introgressing from Neanderthals, how do you know?

Sheep happen. I suppose mentioning _any_ fact cheapens my argument. Pretty much anything one might say invites responses from people who have no clue.

Another possible reason, at least for some people: a lot had been invested in mtDNA studies that assumed neutrality. Maybe the idea of selective advantages or disadvantages in mtDNA was unwelcome? On the other hand, in most of the strong-bottleneck models, losing mtDNA variants with an initial frequency of 2% was the likely outcome, even with neutrality.

In relevant examples, say hybrids of chimps and bonobos, 'very decreased fitness' doesn't happen. At least not in any obvious way. If someone had said that we see noticeably decreased fitness in even one cross between mammalian sister species with comparable time depth of separation, I would call that a non-silly argument. I haven't ever seen that argument, and I don't think that there is any example of such a cross. And to all the people out on the net who are speculating about all gene flow being from Neanderthal males to human females - we don't see any Neanderthal paternal _or_ maternal markers, so stop already.

My guess is that the experts (people like Stringer on the one hand, Paabo on the other) never checked out the literature on mammalian hybridization, never checked out what is known about adaptive introgression (either known cases or pop-gen theory). So their priors were based on nothing.

"a test that is based more on sensory data than language and reasoning." One that would lose all that nasty predictive ability. One at which myopes would do poorly and coyotes excel. Chimps know more about which African wild plants are safe to eat than I do. I guess they're smarter than me. And they survive in the wild. I guess _they_ must have an IQ above 50. And turtles survive in the wild, so... I am reminded of a discussion in which most of the class argued that pigs are just as smart as people, in their own way. Now that I think of it, those undergraduates were pretty good evidence for their position. Sarcasm off, scores on western-designed IQ tests are what matter, because they do a fair job of predicting achievement in western-influenced technological societies - which are the only societies that matter. When hunter-gatherers figure out how to beat those technological societies - when they can trump fusion bombs, jets, Iphones and penicillin - then their other ways of knowing will be worth revisiting. I'm not holding my breath.

If ants somehow transmit complex learned stuff to other ants, they have culture. Somehow I don't think they do. Humans have never been static, but the selective pressures they experienced have to have changed a lot more rapidly in recent millennia than they did 150k years ago.

I have been known to give subroutines odd names. Once, when writing a simulation of a multiprocessor system solving a sparse set of linear equations, I named all the routines used in the first step of the solve after German generals (manstein, guderian, leeb, etc), all the routines in the second step after Soviet generals (zhukhov, koniev, etc) and the routine used in both sections was, of course, named vlasov.

Just another example of the square-cube law. 
 
As for cancer, calculate the effect of adding a couple of extra tumor suppressor genes. Works in Down's.