> You aren't even addressing Diamond's argument.
I think Jared Diamond could be correct, but I would still say that his argument is lame because he failed to establish a high likelihood of prompt invention once prerequisite conditions are met before concluding what those supposedly missing prerequisites were. I'm not addressing his argument because he failed to convince me that his argument addresses a real question in the first place. Supposing hypothetically that ancient Mexico had pack animals, it still might have been thousands of years before pack animals and wheels were put together.
> Going from [rollers] to the wheel doesn't seem like a huge leap of insight.
I really think it was. There's a big conceptual gap between rolling something on logs, and making the rollers captive through the use of an axle. The roller goes from being a tool used to move a thing to being an integral part of the thing. Wheels may have been invented more than twice, but there's not much evidence for that. And in any case, it certainly did take a very long time for some seemingly simple applications of wheels to be considered after wheels were popularized for more sophisticated and demanding purposes.