First of all, women are fertile into their forties, and then they still have to raise their offspring to a point where they can fend for themselves. That would result in a lifespan of about 60 years.

Secondly, older people represent the stored knowledge/values/experiences of a tribe, especially in times before written language, so they are valuable to have around (healing, plant knowledge, cyclical events like floods etc).

And they're useful for babysitting when everyone else is off hunting/gathering.

Longevity is a "biologic novelty" and not something evolution seems to have selected for. But it does appear our brain is capable of altering that set of circumstances. We just have not empirically figured out the key points. Some of us will keep that search up

Yes, tribal elders are a valuable resource. I think the Quilt is saying that our learning abilities may overcome our innate short lives. Dawkins somewhere says something like: evolution in the human case seems to have said, 'you've got a brain, so just do whatever you can to keep going'.

Anyway, Galapagos tortoises live pretty long, and so do giant redwoods.

Here's a pretty relevent quote from the book Genome: "Our Stone-age ancestors began breeding at about twenty, continued until about thirty-five and looked after their children for about 20 years, so by about fifty-five they could die without damaging their reproductive success. Little wonder that at some time between fifty-five and seventy-five most of us gradually start to go grey, stiff, weak, creaky, and deaf".

The author explains that after we pass average child rearing age, we are at a point where genes that damage us at this stage of life cannot be evolutionarily selected to die out.

I doubt a "correct" human diet, or anything else really, would have the effect of turning on some kind of gene for altruistically dying, if that was what you were asking.

I'm assuming this is based on the idea that following reproduction and child-rearing, it would be better for the group if the individual dies, thus freeing up resources for the more reproductive members? Good thinking; this is an idea which has been proposed several times by evolutionary biologists. However it has been refuted by modelling and by more accessible logic. Evolution doesn't work at the level of 'what is good for the species', or even the group (except in certain very particular circumstances). Natural selection operates on individuals, not groups. For the individual to die in the way you are suggesting it would have to be programmed i.e. selected for. There is no strong selective force at the level of the individual to drive the evolution of such an outcome; it is not in the interests of the individual. Unless perhaps the groups were very small and highly related.

But that is background. Probably it would be too difficult a design task, even for evolution, to have a body which functioned optimally on a diet up to a point, and which was killed in fairly short order by the same diet.