[gonehome]

I think the idea that aging may be an evolved anti-cancer mechanism is interesting. Death of individuals is itself probably a selective advantage for the species. Though I suppose shorter lifespans are also more commonly seen in species that reproduce sexually so that preference for continued biological variance is probably also a factor.

[narrator]

Well why don't we just stay hyper fertile till we drop dead? That would be very positive evolutionarily! Truth be told evolution is a chaotic imperfect process that does all kinds of glitchy seemingly suboptimal things so aging is not perfect and divine because it's evolution. Sorry Candide, we do not live in the best of all possible worlds.

[nostromo]

Most animals don't survive past their reproductive years. Humans and some whales do, but that's it.

Some evolutionary biologists suggest that animals that share wisdom or share childbearing duties with their offspring evolved to survive past fertility because doing so increases the competitiveness of their shared genes.

[tjradcliffe]

Grandparents transmit culture to their grandchildren. This is a very good bet as to why humans live such absurdly long lives.

It might even be testable.

I often wonder if the Upper Paleolithic Revolution (https://en.wikipedia.org/wiki/Upper_Paleolithic_Revolution) could be correlated with genetic changes related to longevity. If the genes responsible for our double-lifetimes (we live about twice as long as you'd expect a mammal of our heart rate to survive) were small in number and had mutations that could be traced back to ~50 ky BP it would be strong evidence for this idea.

[darkmighty]

Also, we didn't have enough time to evolve greater lifespans and longer fertility spans, we've done so artificially much quicker than evolution can handle.

[TTPrograms]

Well arguably we haven't evolved to survive past reproductive years - life expectancy past 40 is only about 150 years old. That's much too short of a time for evolutionary effects to have an impact.

[nostromo]

You're misreading the statistics a bit. The biggest gains in life expectancy have been at birth. There have been old people for a lot longer than the last 150 years.

Even Aristotle knew about menopause. And Plato lived to be 80.

[darkmighty]

You have to remember a significant time in human evolutionary scale is probably something like 10k-100k years. Looking at wiki page for life expectacy at the Paleolithic [1] (even at 15 years) shows it lines right up with woman menopause.

[1] http://en.wikipedia.org/wiki/Life_expectancy

[FeeTinesAMady]

Life expectancy is a mean value, and it's skewed by people dying young. So the life expectancy at the Paleolithic doesn't mean that people didn't survive past menopause: it means that a lot of people died young.

[pierrec]

That's actually a very interesting question. Evolution means that out of all states that stem from the current one, the optimal one tends to endure, so the "best of all possible worlds" is indeed a pretty good way of phrasing it, if a little vague.

Staying fertile till we drop dead would be sub-optimal because germline stem cells (the "factories" that produce sperm and eggs) accumulate mutations as we age. Those mutations are useful as one of many mechanisms of evolution, but they must only occur in reasonable amounts, as they also increase the probability of nonviable offspring.

In short, protecting genetic information while still allowing evolution is a hard problem. This problem is not solved in exactly the same way by different organisms, and occurs on different time-scales.

[eurleif]

>Evolution means that out of all states that stem from the current one, the optimal one tends to endure, so the "best of all possible worlds" is indeed a pretty good way of phrasing it, if a little vague.

"Best of all possible worlds" suggests a global maximum to me. Evolution can get stuck in local maxima.

[Florin_Andrei]

> In short, protecting genetic information while still allowing evolution is a hard problem.

In some ways, it seems like this is all that life on Earth is about. A planet-size computer running for billions of years to try and come up with a good solution to this problem.

[lqdc13]

It could also be sub-optimal because there would be less time spent rearing the offspring that you do have. So they would have a lower reproductive success.

[waps]

You have to keep in mind that while technically you are correct, evolution has been running for at least 3 billion years.

Fertility is the main thing it optimizes. To think you know the way this should work better than evolution is like betting you know adwords better than Google.

In general it would be extremely safe to assume that human fertility is the best optimized part of the human genome and that in 1000 years we'll still be discovering factors that evolution took into account "designing" human fertility. There is (or will turn out to be) a very good reason for every single tiny detail about how our bodies procreate. If you can't see the reason, the fault is likely with you, and assuming otherwise is effectively betting against an algorithm that has had 3 billion years to ponder this question. It thinks slower than you, of course, but you're still quite unlikely to have the upper hand in that bet.

> does all kinds of glitchy seemingly suboptimal things so aging is not perfect and divine because it's evolution

This is how evolution works. It's called "mutation". And you're right. Species are a big fan of it, as it massively improves them. For the large, large majority of individuals, mutation is a small or large disaster.

Every human is an experiment meant to improve the human species as a whole. This is great if you're a successful experiment, but the extremely large majority of individuals in any species will be a failed experiment that don't get to propagate their genes (meaning the genes that are different in that particular individual). The vast majority of individuals will turn out to have a (usually small) weakness that will get filtered out. Usually that will be something like a toe that's a few millimeters shorter than most, sometimes it's ALS.

The key here is that evolution very likely won't decide any particular human is a success or failure until the human species gets into trouble again. When that happens, something like 98% or more of all lineages will go extinct, and until that happens, even evolution itself won't know or care what fitness even means.

Unless you know what is going to cause the human species to lose >90% of it's population at some point, you have no point of comparison.

[thaumasiotes]

> Fertility is the main thing [evolution] optimizes.

This is more hedged than it really needs to be. In an infinite-integral-across-time sense, fertility is the only thing evolution optimizes.

[DavidSJ]

Keep in mind that evolution works primarily at a subspecies level, so any mutations which benefited that lineage at the expense of the species as a whole would tend to be favored.

[tjradcliffe]

Actually, evolution is almost entirely dominated by selection at the individual level, with a modicum of kin selection thrown in. The notion that selection acts at other levels is highly speculative (I wrote a novel based on the idea, but it's the purest kind of speculative fiction: http://www.amazon.com/Darwins-Theorem-TJ-Radcliffe-ebook/dp/...)

[DavidSJ]

Sounds like you're agreeing with me? By "lineage" I'm talking primarily about the lineage of a gene, or in many cases the so-called inclusive fitness of an individual.

[waps]

Lifespan is a function of organism size (ironically measured in weight, mostly). The bigger an organism, the longer it's lifespan. Keeping in mind normal human lifespan would be about 30-35 years, you find that many animals of similar size have a natural lifespan within a factor 2 or so of a formula that also applies to humans. Normal whales, for instance, have a natural lifespan of about 70 years.

That this formula works would seem to indicate that you are right. Death is natural, but it's "planned". The easiest way for such a formula to work would be that your genes somehow contain a death clock.

But there are multiple death clocks. One limits number of cell divisions. There is another one known that limits the amount of energy that can pass through a cell, after which it will kill itself. There are various others, one that kills the cell if it isn't deactivated on a regular basis (presumably meant as a check on DNA integrity), one that is triggered from the outside of the cell, ... the list goes on.

[ehvatum]

Size alone has limited predictive power for lifespan. Consider the example of the naked mole-rat (30-35g, 31 yr) vs Norwegian rat (450g, 3 yr). Naked mole-rats dwell in environments shielded from predation, whereas no R. norvegicus in the wild could possibly hope to avoid attrition for 10 years, even if its ageing process allowed 20 years of healthspan, let alone 30.