I think I came in at the middle of this thread, but I can't find what came before that first tweet and it's not worthwhile to dig further. So let me begin by asking, just by the way: Has anyone said that eugenics "wouldn't work"? In the absence of an example to discuss, I'll examine what it would mean for eugenics to work.
On the most basic, neutral technical level, I can't see that eugenics would work. In The Pure Society (Verso, 2000), the historian of science Andre Pichot quotes a 1930 book by the French biologist Jean Rostand:
All kinds of objections, both practical and theoretical, were raised against the application of eugenic ideas. In particular against the sterilization of severely defective individuals, in which connection the discouraging slowness of the possible effects was adduced. Fischer calculated that it would take twenty-two generations to reduce the proportion of mentally ill from 1 in 100 to 1 in 1,000, and ninety generations to reduce it to 1 in 10,000. The more serious objections, however, were of a moral order; to many consciences, any restriction on the right to procreate was seen as an injury to individual liberty and the dignity of the human person [Pichot, 135]."The Fischer referred to here is probably Eugen Fischer, a German anthropologist and geneticist, and the author, along with Baur and Lenz, of what at the time was the bible of human genetics...; Fisher was a eugenicist, and compromised by Nazism" (ibid.).
I don't know if Fischer's calculations would stand up today, but it's telling that even a proponent of eugenics acknowledged that it would take about five hundred years to reduce a (supposed) trait by a factor of ten, and a couple of millennia to reduce it by another factor of ten. This doesn't mean the enterprise is impossible in principle (which I take to be what Dawkins means), but it's definitely impossible as a practical matter. And that's only one detail, one flaw of many in the genome, that would have to be managed to improve the race. Perhaps Fischer believed that a Thousand Year Rule would endure long enough to carry out the project, but I doubt Dawkins does, or that he's even thought of the problem.
Then too, mental illness is not a genetically determined trait, though scientists are still trying to claim that it is. (On some level, many are still eugenicists, keeping the faith in their well-funded catacombs against the day when Political Correctness is destroyed and they can carry the faith of their fathers back out into the light.) Sterilizing the mentally ill (touted as a humane alternative to simply killing them) would not eliminate mental illness. Nor could a eugenic program affect most of the qualities eugenicists hope to eliminate or augment, since they are not genetically determined.
Presumably in order to ward off moral objections, Dawkins offers the neutral example of breeding "humans to run faster or jump higher." Even those are dubious, since we have no idea how to breed for such skills, which are a combination of physical endowment and decades of training for each individual. I've been reading Dick Francis's mystery novels, which are set in the world of high-end horse racing, and a point that crops up repeatedly is that breeding horses to run faster and jump higher is still a crap shoot, with no guarantee that the offspring of elite sires and dams will be a champion. Training is a factor, as is the ability of the jockey, but also dumb luck in the roll of the genetic dice. Yet the reproduction of racehorses is under human control in a way that human reproduction isn't. The most that could be hoped for (if that's the word) is to produce a strain of human beings who might be able to run faster and jump higher in competition with each other, while most of us remained unaffected. That's a far cry from improving human beings as a species, and as with the fantasy of sending a few colonists to the stars while most people are stuck on a dying earth, it's fair to doubt that eugenicists have ever had any interest in human beings as a species.
When you bring in the breeding of other animals or plants by human beings for human ends, more practical problems appear. I've already suggested one; human generations are a lot longer than those of the other species. A racehorse, I gather from Dick Francis, can start competing by the age of two. Homo sapiens can't breed for at least fifteen years, and preferably longer: the convention for human generations is around a quarter of a century. That's why Fischer's eugenic program would take so long to have any detectable effect. Eugenic fantasies of the twentieth century didn't consider this. The Noble Engineer Robert Heinlein, in his stories about the long-lived Lazarus Long, imagined a privately-funded foundation that encouraged individuals from long-lived lineages to marry individuals from long-lived lineages. This foundation took the "long view," Heinlein postulated, but it produced significant results in only a couple of generations: people who lived long enough to be noticed and persecuted by their short-lived neighbors. Heinlein, it should be recalled, is regarded as a practitioner of "hard science-fiction," the coldly realistic, science-based SF that doesn't indulge in airy-fairy wishful thinking and fantasy.
An additional disadvantage of those who would breed humans is that they can't just cull the failures, as selective animal or plant breeders can do. Even sterilizing them is in bad odor nowadays, thanks to Hitler. And that brings up another point where Dawkins fumbles: it's not possible to object to eugenics in purely practical terms on one hand or moral/political terms on the other. There's not a clear boundary between the moral and the practical in any domain that affects human beings. Advocates of eugenics, then and now, argue in moral terms: don't you want to decrease human suffering, eliminate preventable diseases, and move us ahead to the next evolutionary stage? If it's pointed out that eugenics will do no such thing, they appeal to the wonders of Science and accuse their critics of anti-intellectual bias, and superstitious hatred of evidence and reason. As Dawkins did here. I don't accuse him of being a crypto-eugenicist, because I have no idea what lurks in his heart. I don't think he does, but he argues in such bad faith, he's so intellectually and morally dishonest, that I see no reason to take his protestations seriously.
Finally, just as a purely scientific issue, no one knows what a superior human being would be. Jumping higher and running faster do not a Homo superior make. Getting rid of debilitating genetic illnesses would be an improvement, but it wouldn't be a new species. (Remember too that advocates of genetic manipulation claim that they can do the same job without the traditional eugenic methods, though their claims must be regarded critically too. Scientists love to promise wonderful outcomes they are nowhere near able to deliver.) Dawkins is confusing or equivocating between selective breeding, a local intervention, and eugenics, a global one. Eugenicists appealed to the analogy of breeding animals and plants, but they had a much more ambitious project in mind: that of replacing Homo sapiens with Homo sapiens 2.0. The racism in their propaganda wavered between the implicit and the explicit, as it does today.
One difference between selective breeding and eugenics is that the former involves the control and manipulation of other species by human beings for human ends. Breeding doesn't really have the good of its subjects in mind. Are cows made happier, or even objectively better in evolutionary terms, by being bred to produce more milk for human consumption? I think it's safe to doubt that. (Self-styled Darwinians love to accuse people who, e.g., wear spectacles of interfering with Evolution; I can't recall them ever complaining that selective animal breeding does, though it certainly does so.) When scientists use this analogy, it's reasonable to ask who's breeding whom, to what end, and for whose benefit? I think they tend to imagine themselves as specimens of Homo superior, selflessly guiding human cattle into the glorious future. This was also at least a subtext of mid-20th century eugenicist science fiction; aside from Heinlein's Methuselah's Children, see his 1949 novelette "Gulf" in the collection Assignment in Eternity, where the scientific racism is explicit in the text, and its 1982 sequel Friday. Again, I'm not accusing Richard Dawkins of such an attitude, though his use of the selective-breeding analogy relies on it; I just think that he hasn't thought about it, which doesn't speak well for him either.
This careless equivocation is typical of Dawkins. It goes back to the beginning of his career as a popular-science writer. In The Selfish Gene (1976), he flipped between insisting that he called genes "selfish" and "ruthless" as terms of art, 'not even mean[ing] the words in a metaphorical sense," without ever suggesting that organisms are selfish; and casually saying that "This gene selfishness will usually give rise to selfishness in individual behaviour" (TSG, 2) and "we are born selfish" (ibid.). When critics accuse Dawkins of saying that selfish genes produce selfish organisms, he and his defenders point to his disclaimers while ignoring the passages which undermine the disclaimers. It's like reading Christian theology: even in a rather short book, you can find contradictory doctrines which can be used as prooftexts for whatever position you like. I said above that I'm used to British writers expressing themselves clearly, and maybe Dawkins isn't an exception: he dumps his mistakes and contradictions openly on the page, so blatantly that only the mind of faith can miss them. Dawkins should be dismissed as thoroughly as he dismisses theologians, and for the same reasons.
