Originally a comment by Claire on It would totally work.

The only way to do it is to emulate how we breed animals and plants for specific traits: you have to very carefully select for the traits you want and literally force them to breed. People wouldn’t have any choice on who to breed with or indeed whether to breed at all. And any inferior offspring would be killed. It’s even more dystopian than Brave New World.

Anyway, electing for high intelligence would probably be in contradiction to docility. In authoritarian regimes, killing off the intellectuals is fundamental for a reason. You can’t have a docile herd without it. And we are not naturally herd animals, whereas most domesticated animals are. They’re easier to handle.

I’m not defending Dawkins here but I can see where he’s coming from in a way that would probably be lost on most people outside of biology. It’s easy for me to see this dispassionately, as Dawkins does, with humans as just another animal. Unlike Dawkins, I appreciate the concepts of sociology, anthropology, and psychology – humans (and some of the non-human primates) are not in fact quite like other animals. Although, for clarity, I don’t consider him a part of my field. Dawkins is an evolutionary biologist, not a geneticist. And frankly, not a particularly good one, just a loud one.

But most importantly, he’s wrong.

Francis Galton, who is considered one of the fathers of genetics and in more specific terms one of the pioneers of my specific field. He was a big fan of eugenics and was Darwin’s cousin. He spent years studying trait variation in populations and this led naturally (naturally at the time, this was the 19th century) to considering how to improve the stock of the human race. I won’t go into more detail because it’s a whole lecture series on its own but he actually invented a number of methods and approaches we still use today.

This is the context that Dawkins has stripped from this conversation. Twitter is not the right medium for this kind of discussion.

It’s an uncomfortable heritage (the irony of which is not lost on me). Those of us in population genetics, genetic epidemiology, and statistical genetics owe a lot to Galton, but some of his ideas were hideous. We have to accept that.

So, why are Dawkins and Galton wrong?

When we raise cattle for milk or pigs for meat or wheat for flour, we observe traits and deliberately select for the ones we want. But to do that, we have to account for the traits that come with that. Cattle raised for milk or meat would not survive in the outside world. Their biology has been distorted to the point that they are reliant on humans to support characteristics that evolution would otherwise select against.

The idea of breeding (or genetically engineering) a superior human is biological nonsense. You select for one trait, you’re selecting for and against a bunch of others and you never know what they are in advance. Most genes are pleiotropic (which means they have more than one function) and they interact, so it’s just impossible to engineer either by breeding or by engineering a “superior” human, whatever your metric of superior is. They wouldn’t be able to survive.

tl;dr Any species that can’t afford to be reliant on another’s care for survival cannot survive on its own. All species survive by being adaptable. A human herd bred for “superiority” would die off pretty quickly because evolution is no longer working with the environment of the outside world. The environment always wins.