10k years is plenty enough: http://www.salon.com/2013/03/10/paleofantasy_stone_age_delus...
In fact, 10k years is enough for dogs to have evolved to digest carbs a lot better. There's even specific genes that are known to help that.
In fact, 10k years is enough for dogs to have evolved to digest carbs a lot better. There's even specific genes that are known to help that.
Obviously there has been some adaptation to the change of diet, and it is also obvious that this diet was not letal to human (or nobody would have adopt it), but this does not implies that it is optimal for human consumption. Most people in my region are not lactose intolerant... but do you know what? I threw up every time that my mother had me drink a glass of milk because it was "healthy" and good for my bones. If adaptation depends that much between humans (the % of lactose intolerant or lactase persistent is not the same in the US than in China or Africa, for example), trying to correlate anything with animals seem completely off to me.
The paleo hypothesis is that we haven't adapted to carbs to the point that they don't cause disease. Carbs aren't dangerous enough to kill us before we reproduce, so why would we expect the genes to be selected out?
Population members aren't useless to preserving genes once they're done having babies.
Elders use and share collected knowledge (particularly skills that take a long time to master), care for babies while the parents are hunting/etc., help with gathering/cooking/cleaning/etc.. If they are all sick & dying instead of doing this work, your group isn't going to do as well as the competing group whose elders are in good shape.
Grain eaters are still overall pretty healthy, so they can still accomplish the beneficial things you describe, allowing their less-than-optimal genes to carry on.