I agree with this criticism in the case of sleep apnoea, since the respiratory system is complicated. By intuition, it seems like the evolution of a complicated trait should require significant genetic changes, and thus be more likely to get stuck in a local maximum. I'd expect the respiratory system to be quite resistant to change: what if you've got an SNP that reduces sleep aponea but also prevents peristalsis, speech, weakens your neck, etc.? You might need a number of SNPs to all change at once in order to fix the problem.

I don't think the argument holds so strongly for questions like "why don't we produce GLP-1 hormone at a 20% higher level?" Increasing GLP-1 production levels seem like a smoother transition with fewer side effects. If higher production is better, and you've got an SNP that increases production by 5% on its own by increasing the number of L cells in your intestines (or something), that sounds like it should make you more reproductively viable in today's world and shouldn't have as high an impact on everything else as changing the shape of a throat.

Most of this is just my intuition, and I'm absolutely not an expert.

Edit: final note that I think in today's world is the key gotcha here, per the original text: "getting fat in times of plenty was a feature and not a bug", meaning that it was evolutionary advantageous for most people to put on weight until very recently, evolutionarily speaking.

> it was evolutionary advantageous for most people to put on weight until very recently

And it will perpetually remain possible for it to very abruptly and unexpectedly prove advantageous once again.

Primates in the wild really hit the nail on the head here. We clearly lost the equivalent at some point and haven't rapidly regained it. There has to be a reason for that (though it might well prove to be entirely incidental).

> The body is a marvelous machine, but it is far from a perfect or intelligent design.

And I think that we can only expect the process of evolution through natural selection to operate on traits that cause differential survival of individuals before reproduction. All sorts of new traits acquired through random mutation and recombination that only affect individuals after typical reproductive years will exist even while natural selection is still operating. The answer to “why doesn’t the human body operate like [some idealized state]?” Is either 1) insufficient time has elapsed to allow natural selection to change the gene pool, or 2) It doesn’t affect survival before or during reproductive years. (Or, I suppose a third explanation is that the required mutation that drives the change just hasn’t happened yet. Our species hasn’t been around that long after all!)

Yet women have a longer lifespan despite shorter reproductive years.

If in a hunter-gatherer society most people stop reproducing after 40 (not entirely realistic, but bear with me), me being alive and fit at 45 can still improve the survival of my genepool. I can hunt meat and gather plants to make sure my children and grand-children are well-fed and able to produce plenty of offspring, carrying my genes on. I can build shelter, care for the young (improving their survival chances and freeing up the parents to do more physically demanding work).

And my siblings share half of my genes, and my cousins a quarter (assuming no inbreeding at all, otherwise the numbers go up). So helping their children survive and proliferate is a win for the proliferation of my genes. Evolution gets less effective the more indirect the effect and the lower the gene contribution, but it doesn't completely stop working. If there is one clan where everyone dies at 30 and one where people stop reproducing at 30 but stay useful members of society until 60 then then the second clan will do better and outcompete the first

But you also take up resources, continue to decline physically, etc.

We know that there is strong selection pressure for things that occur during reproductive years. No one debates this.

How much the sort of scenarios you're discussing impact selection pressure is significantly less clear.

But it's also a moot point for this discussion - we haven't been in a world where people can get this fat so easily and in such large quantities for long enough to make significant impact to evolution even if we assumed that selection pressure works the same in said world - and again, we know it doesn't.

Grandparents are a selective advantage well after reproduction. Older members of the tribe can watch the young while fitter members engage in hunting and foraging. They can also teach the young. Most tribes around the world that still exist you see this pattern.

But for an ancient tribe grandparents means 35-50 years old people. Just as health starts to decline in modern humans.

Correlation is not causation. Plenty of people theorize there is large selective advantage here, sure. They may even be correct! Others disagree with them on it being a large advantage.

But even if there is it doesn't mean that there is any reason whatsoever that increased GLP-1 production would be enough of an advantage to have had significant impact on evolution. We're going from A is true, B is likely true, C might maybe be true, all the way to Z is definitely true.

> Yet women have a longer lifespan despite shorter reproductive years.

After industrialized nations have reduced both maternal mortality and violent death (warfare) to small fractions of their pre-modern levels. The life tables of 1500, 500, or 5000 BC would look very different than their modern versions.

To increase chances of survival of the population you need just an extra 10y after the end of reproduction to grow last child, no more.

Reducing the reproductive fitness of each individual after a certain life time might even increase the reproductive fitness of the species (or tribe) as a whole, because it might reduce competition between generations and foster cooperative behavior: if parents and their children compete for the same resources – from sexual partners to food needed to raise a new generation – you'd never develop the cultural norms of life long mutual support across multiple generations.

Evolution should make intelligent choices for us but there's thousands years where a few extra pounds meant surviving a famine.

Also I bet it will kill a few people with eating disorders.

As long as you're in a 1st world country and have a BMI over ... maybe 25 (and aren't on steroids so you're both anorexic and overweight at the same time) I'd guess you're dodging a lot of the most obvious risks.

How do you define 'intelligent' and 'evolution?' I think we might differ on what those words mean, which would make this conversation difficult, so let's clarify where each of us is coming from.

My understanding (which may be slightly misguided because this is definitely not my field) is that Intelligence is the ability to use information for reasoning, decision making and achieving understanding. This ability, as a directed action, only applies to creatures with consciousness.

Regarding evolution, my understanding is closer to a process that is more akin to a passive filter, with no guidance to speak of unless we want to stretch the limits of the concept of guidance to encompass survival, which is the genetic materials of those who survive the environment getting passed on, which would also implies that sometimes garbage/noise stays in the gene pool. While information (genetic material) is being passed on, it is not being selected through reasoning or consideration, but more of a mechanical process (although there's an argument to be had that we, humans, might be changing that game).

How would you define those words? Happy to consider understanding different than my own, since I do find it interesting.

The OP said something about how the human body doesn't make "intelligent" decisions.

I was being cheeky, not using the best word for the job.

Evolution is not really passive, it's a combination of mutation (which is not intelligent, but not really passive) PLUS a passive filter, which gives the illusion of intelligence since it ends up as a kind of optimiser.

Now, due to various factors, evolution isn't perfect. For gene expression reasons (I think) the layout of mammal skeletons are not really effected much mutations, but bone sizes are; so you have giraffes with huge neck bones that are the same size as human neck bones (this isn't the best way to do it). You have bats with the same finger bones as humans (or at least very similar) acting as wings. Elephant feet are almost the same as human feet in an x-ray. Some genetic "knobs" are easier to turn than others, and it can be that shrinking something is easier than getting rid of it. Just google "vestigial structure". Actually, LLM are kind of similar (it's easier to optimise a value to zero than to remove it, even if removing a value will make the model better and faster).

It's also more adaptative than optimal. Some argue it's the "selfish gene" that "wants" to be passed on. If a gene makes you kill all the competition, that's a "fit" gene. If working in a team increases the survival rate of your cousins (who also kind of carry your genes) it's "fit". If having a certain percent of a village be gay would somehow increase survival rates (especially of their kin) then having a chance of being gay is "fit" even if the individuals themselves could be less likely to reproduce.

There's also some meta stuff about how the rate of mutations could partly be genetic.

Evolution is seen as intelligence because it solves hard problems, and IMO that's kind of the only thing intelligence needs? People who talk about "consciousness" talk a lot about how they think they think, but is how we think actually an illusion? The feeling of understanding is just a chemical pleasure hit when the brain hits the "problem solved" button. Consciousness might be an illusion, we might make a decision and wonder how we made it, then an AI in our brain quickly justifies it in some BS way that isn't actually our real thinking process.