Jared Diamond, in discussing animal domestication, claims that the local availability of species with the right qualities for domestication was key, rather than anything special about the biology or culture of the humans living there. In some cases that may be true: there aren’t many large mammals left in Australia, and they’re all marsupials anyway. Stupid marsupials. He claims that since Africans and Amerindians were happy to adopt Eurasian domesticated animals when they became available, it must be that that suitable local animals just didn’t exist. But that’s a non sequitur: making use of an already-domesticated species is not at all the same thing as the original act of domestication. That’s like equating using a cell phone with inventing one. He also says that people have had only mixed success in recent domestication attempts – but the big problem there is that a newly domesticated species doesn’t just have to be good, it has to be better than already-existing domestic animals.
Indian elephants, although not truly domesticated, are routinely tamed and used for work in Southern Asia. The locals in Sub-Saharan Africa seem never to have done this with African elephants – but it is possible. The Belgians, in the Congo, hired Indian mahouts to tame African elephants, with success. It’s still done in the Congo, on a very limited scale, and elephants have recently been tamed in other parts of sub-Saharan Africa, such as the Okavango delta. Elephants have long generations, which makes true domestication difficult, but people have made domestication attempts with eland, African buffalo, and oryx. They’re all tameable, and eland have actually been domesticated to some extent. If a species is tameable, economically useful upon taming, and has a reasonable reproductive schedule, domestication is possible: selection for even a few generations can change their behavior enough to make dealing with them a lot easier.
As for the Americas – have you ever had a deer eating out of your hand? Bison seem too wild and scary to have ever been domesticated, but then I’m sure you would have said the same thing about the aurochs, the wild ancestor of cattle.
In fact, in my mind the real question is not why various peoples didn’t domesticate animals that we know were domesticable, but rather how anyone ever managed to domesticate the aurochs. At least twice. Imagine a longhorn on roids: they were big and aggressive, favorites in the Roman arena.
Let me throw out an idea originated by an old friend, Ivy Smith. Consider mice, cats, and toxoplasma. Toxoplasma is a protozoan with a two stage life cycle: one in an intermediate host (mice and rats, among others) and a definitive host (some feline). Toxoplasma only reproduces sexually in the definitive host, and it ‘wants’ to end up there. It manipulates the behavior of the intermediate host in ways that increase the probability of transmission to the definitive host. For one thing, it makes mice like the smell of cat urine, which elicits fear in uninfected mice. In fact, it seems that toxoplasma-infected mice are sexually excited by cat urine. How weird – a parasite rechanneling sexual interest…
The idea is that at least some individual aurochs were not as hostile and fearful of humans as they ought to have been, because they were being manipulated by some parasite. The parasite might have caused a general reduction of fear or aggression without infecting or aiming at humans – or, maybe, humans really were the definitive host, and the parasite knew exactly what it was doing. The beef tape worm – which we originally acquired from lions or hyenas back in Africa a couple of million years ago – might have gained from making infected bovines quiet, passive, maybe even overly friendly in the presence of humans. This would have made domestication a hell of a lot easier.
Parenthetically, such host manipulation may play a really important ecological role. For all we know, if canids and felids had to rely purely on their own abilities, they’d starve.
The beef tape worm may not have made it through Beringia. More generally, there were probably no parasites in the Americas that had some large mammal as intermediate host and Amerindians as the traditional definite host. Amerindians simply hadn’t been there very long. Domesticating bison may have too hard for unaided humans, back in the day.