To those who see just how much better is a civilized life, one of the most terrifying things one can learn from history is that pretty much all past civilizations fell.
Yes, our culture and civilisation is biologically maladaptive. I first started telling my students this story nearly 20 years ago with a comparison to animal keeping in zoos. See, the practical proof that you are keeping your animals in species-correct conditions is when they reproduce in captivity. It can be surprisingly difficult to make it so, and food and shelter are not enough. Turns out that humans too have a hard time reproducing when you put them in high rise cages and the reasons are similarly varied and hard to pin down as they are for zoo animals. But yes we can conclude that the human zoo clearly does not meet the living conditions humans expect. The puzzle for me is, why world wide so? Cultures, beliefs and ways of life vary widely. Not everyone lives in high rises and mere density alone does not seem to be so harmful. Whatever it may be though, the modern world does not meet what we as humans biologically expect.
What's particularly interesting in the human case is that humans get to choose, or even build, their own enclosures and environments. The wealthier they are, the more freedom they have to make their life be whatever they want... but the wealthier they are, the lower their fertility rate.
Maybe humans need brutal, miserable living conditions?
“Maybe humans need brutal, miserable living conditions?”
Or maybe easy access to birth control, and living in a state where the government provides you old-age benefits rather than being dependent on your children as your old-age plan makes a massive difference ?
I don't think the government will provide me old-age benefits, and I don't intend to burden my children. Modern society provides other ways to prepare for retirement because we don't live a hand-to-mouth, subsistence lifestyle. For that matter, the ability of the government to provide old-age benefits derives from the surplus productivity of modern society. That surplus would be used to enable a secure retirement in some other way if we didn't do it through government.
No, to really make people depend on their children to support them in their old age you need to remove that surplus, making life harder and much more precarious.
Birth control, on the other hand, has been a crucial contributor to declining fertility. But I'm not entirely certain that removing birth control now would reverse it, because people are also having much less sex.
*You* might not think government is going to provide you old age benefits, but in fact it has been doing so for 50+ years now, and the vast majority believe it will continue to do so.
I’m not advocating either eliminating old-age benefits (though for largely different reasons, I would pare them back) nor certainly advocating for the elimination of birth control. I’m just pointing out these are two enormous explanatory factors in the decline in TFR.
The thought crossed my mind. Analogy with obesity: Humans are meant to have some kind of trouble getting food. If they haven't, they just get obese. Conjecture: humans are meant to struggle in their lives. If they don't, they get depressed and their fertility drops.
“Meant to” have some kind of trouble getting food, and are biologically adapted to the (until-recent) reality of difficulty getting food are overlapping but not identical concepts.
As a non-trivial example of a cultural adaptation that overcomes biological adaptation, monogamous marriage is helpful to non-alpha, non-rich males, vs most prior human cultures that allowed polygamy. It also almost surely is adaptive positively for cultural fertility as well.
I suspect you would not claim that humans are “meant to” be polygamous, although perhaps you would.
I’m not suggesting that cultural adaptations are easy, merely that they are *possible*.
no issue with what you write. "meant to" is obviously mis-spoken from a Darwinian standpoint, a teleological sleight of hand. My argument is that some of our cultural contraptions, when they appear, play to our biological equipment in ways that at least temporarily depress fertility. Darwinism always wins though: eventually, whoever reproduces, reproduces, and in time dominates the genetic space.
I mean that some (not all) aspects of culture may be working like a flame to the moth. We get attracted to things that in the natural world are producing good results (eat fatty foods, live in safety, etc) but for some often unknown reason are producing negative results in our culture. Obesity is one example, low fertility is another. Unless you would like to argue that it is well adaptive for the moth to die in the flame. I would agree that over time, nature adjusts either because some cultures die out, or because whatever genes are attracted to our culture's flames die out.
Humans are a social species and I have no issue claiming that culture is part of biology. Genes change much slower than cultures though. So I find it perfectly plausible that culture develops features (it "has a mutation" or "develops variants"1) that does not play well with the biological baggage that humans inherited genetically. So think of it like this - humans are moths and our culture invented the flame. We like the flame as it gives us light. But we also keep getting attracted to the flame too closely, and die. Two solutions to this: either culture evolves (it develops flames that stop attracting the moth, keeping the benefit of artificial light but without self harm) or natural selection takes out those genes that don't work well in a particular culture (moth variants attracted to flames die out, mutated moths that don't get attracted to flames inherit the Earth).
So I asked AI a couple of questions on this: If we were to treat modern humans as a species in captivity whose optimal living conditions are not met by an imaginary zookeeper, what would be the top 3 contenders for unmet human needs that would lead to fertility decline? The answers were mostly lukewarm for me, such as microplastics and endocrine disruptors, cost of living etc. None of these are globally universal in my mind, some started much later than the fertility decline etc. The best contenders were female opportunity cost and nuclear family overload. Opportunity cost for women makes at least some sense as an explanation and it takes away the discourse from absolute cost towards other options for happiness than children. Nuclear family overload is the most original of all explanations, here the suggestion is that multigenerational family support may be a crucial need among humans for child rearing. The odd thing here is that some societies with the strongest multigenerational support cultures such as say, Mediterranean Europe and East Asia, have some of the worst fertility numbers. And the individualist US, not strong on parents helping out, and very strong in opportunities for women, until recently had quite strong fertility compared to the rest of the West.
The wide availability of a) fairly inexpensive birth control (before and after intercourse), and b) the fact that rich western societies provide relatively generous old-age benefits and so have changed children from “producer” goods to “consumer” goods have to be in the top 5.
Personally I’d rank them in the top 3 along with female opportunity cost. I’d still rank this last one third, given that fertility is not just going down fast with the highest income women but pretty much across the board. But it’s relative rank is surely more arguable.
I agree with this assessment. People overthink the fertility decline but it boils down to some simple underlying causes.
Unwinding these causes seems impossible. Using the obesity analogy: We aren't going to willingly go back to a condition where people struggle to get calories. We have to find a different way forward.
I think there's a good chance it will take good old biological evolution to turn the fertility ship around. Natural selection will steer us toward a set of physical/psychological traits that result in more offspring in our current condition. The selection pressure is extremely high for doing so but it will nevertheless take 5-10 generations for meaningful change to accrue.
It's funny, for me the fertility decline is less a sign of cultural decay and more like a miracle. We are so lucky that it turns out that when humans are exposed to modernity we just naturally scale back our fertility to sustainable levels. We have freed ourselves from the Malthusian trap! Thank god the world population isn't 20 billion going on 40 billion.
For a long time people assumed fertility would naturally fall to fertility level and then stop. But it didn't stop. So we will decline, until some descendant group grows again.
I think it's likely that within a few hundred years biological evolution will solve the problem of how to encourage fertility within our current condition. That's what biology does. At that point we will be back to worrying about overpopulation.
I wonder what the longer-term future holds, say the next 10,000 years. Every example we have in nature of longer term population (semi-)stability involves a feedback mechanism in the survival rate: The species over-consumes its food resources and individuals starve, or individuals become subject to increased predation by some other species as a result of their elevated numbers.
Malthus was wrong in the short term, but I wonder if he will prove correct in the 10,000 year timescale? Or will humans prove to be the first species able to achieve population (semi-)stability while maintaining high individual survival rate?
Agricultural tech freed us from your Malthusian trap decades and decades ago.
I get that you are probably a leftist who worships and values Mother Gaia more than you do the welfare of human beings, including those billions living lives at much lower standard of living than your own.
So I guess it makes you selfishly lucky given your values.
And if you don’t have grandchildren, then I guess you don’t have to worry that they or their children will not have their Ponzi scheme old-age benefits system that only works with a growing economy, which (unless we get VERY VERY lucky with AI) requires a growing number of productive workers.
We've had many improvements in agricultural productivity over the centuries, the problem is that traditionally they're sooner or later eaten up by a rising population. Currently we have the capacity to feed a lot more people than there are on the planet, but exponential population growth can eat up that surplus very quickly. Of course maybe we can have another green revolution level improvement in agricultural productivity, and another and another to keep up with the doubling and redoubling population... but I'd certainly be a lot less confident of maintaining the per capita food and wealth supply in that scenario than I would be with a stable or declining population.
If we were discussing 20B or 30B people on the planet going to 40B, you *might* have an argument. I doubt it, but at least it would be worth entertaining.
Where the planet is now, you have none at all. TFR is far below replacement in all but the poorest countries. Doubling to 16B is not in the cards right now, even though I wish it were. At this rate 5B is morel likely to be broken before 15B, and maybe as likely as 10B.
Facts matter. They are not on your side of the argument.
There are always claims that a civilization is decaying before it collapses, because there are always claims that a civilization is decaying. We need some more-objective way of finding historical cases of collapse from "decay". We also need a definition of "decay". In general, change of any kind is called "decay" by the oldest generation.
Historically the "decay" has taken the form of rent-seeking behaviors on the part of elites in society.
The arc is:
- An organizational system is set up (a government, a religion, a legal system, ...)
- Some people succeed within that system
- The people who succeed have more influence than others, and they influence the system in ways that benefit themselves
- Eventually the system cannot adapt to new conditions because it is focused on maintaining a status quo
This arc has played out thousands of times in human history and I don't think we have any good ideas for avoiding it. We seem to need an occasional war or revolution to clear the books.
This sounds like a non-falsifiable hypothesis to me, because “successful people have more influence than non-successful people” is true by definition of “successful”. So it’s a property of all societies. That means you can’t test empirically whether societies in which this situation arises fail more often than societies in which it doesn’t.
Can you list some cases of this happening where the causality is clear?
"Successful people have more influence than non-successful people...", is only the first half of the sentence: "...and they influence the system in ways that benefit themselves."
I think the pressing thing about population loss is the possibility that it could cause civilizational collapse, by eg no longer having enough people with the expertise to build a nuclear power plant. But this seems like a very long-term concern. Market forces should keep the number of people in each profession at a good equilibrium.
I doubt there are any industries crucial to civilization in which there is critical expertise which fewer than 20 people in the world possess. So dropping from 8 billion to 2 billion people, which would take 5 generations at .75 children per person (TFR 1.5), would be survivable if people with critical expertise were evenly distributed, and if we pay some attention to avoiding industry extinctions.
However, that expertise is heavily concentrated in the nations with the lowest fertility. So it's more like dropping from 1 billion to 250 million. My gut feeling is that this is still survivable. But the imbalance in fertility between rich and poor countries would present another kind of danger.
I am more worried about civilizational collapse due to an unsustainable population, or to increasing social complexity. I suspect some of our institutions today are already above one or more theoretical maximal complexity limits, such as the critical connectivity threshold beyond which a single failure causes, on average, more than 1 additional failure, and a single failure can percolate across the entire graph.
“Market forces should keep the number of people in each profession at a good equilibrium.”
Depends on your concept of population and numbers therein. There is a theory of the “smart fraction”—the number of “smart” people needed to maintain and expand a 1st class technological society. You know; lawyers, doctors, scientists, engineers, etc. Not everyone—indeed few—can handle these positions. As population numbers decline, those with higher levels of ability declines as well. Sooner or later there are not enough qualified people to run the show and society begins to devolve. Market forces will mean little, except perhaps to promote people beyond their competency. In any event, society devolves to the level of maintenance their smart fraction number is able to handle.
A lot of that distributed expertise can be supported by technology. The design of a car or a modern CPU is as much in the computer as it is in the minds of the engineers building it. And tools like CAD and Verilog allow us to do more design with less expertise (or with more narrowly-scoped expertise) than we used to need.
It's all about having alternatives. Contraception, careers that you do instead of childrearing, and other sources of entertainment than sex. The truth is most people just would rather do other things than raise lots of babies, given the choice.
I say, let them. People are free to find their own meaning. A subset of people are biologically predisposed to have a stronger preference for children, and eventually those genes will spread to the majority, or we'll find a technological solution (babies in vats, robo nannies, life and fertility extension).
Even by Robin's projections, the world population will be over a billion for hundreds of years despite the low birth rate. There are more serious and immediate challenges facing us, particularly wealth concentration and AI. In the next few decades, AI will drastically reduce the demand for human workers, much faster than the population declines (actually world population will continue to increase for many decades). The resulting mass joblessness is the real crisis, and higher birth rates would hurt as we approach that crisis, rather than help.
Haven't you been reading the news about mass layoffs of tech workers and CEOs directly stating this is because of AI?
The point, however, is not about what has already happened, but what is going to happen. The AI will only get better over time, and businesses will only get more experienced and comfortable with using it instead of human employees. There will end up being very few desk jobs an AI can't handle better - and far more cheaply - than a human.
Hanson has long been doubtful about AI, but none of the naysayers expected LLMs to have capabilities like they do, and it's long past time for him to change his position here. You can't keep having the same views on AI in 2025 that you had in 2019. Even I never thought LLMs could do anything close to what they now do. People who think "machine learning can't do this" have been wrong over and over, with every mind-boggling advance since AlexNet. Hey, how come the Turing Test isn't much of a talking point these days?
27.5% loss of programming jobs in the past two years. What will have happened by two years from now? No job is safe. AI researchers are pursuing all avenues to fix the hallucination problem and help LLMs reason better. There are plenty of plausible things they could try. If they fix those problems, what barriers remain?
"when we looked at who worked in that industry, we noticed that programmers were in the minority. They’re dwarfed by, among other occupations, the software developers. [....] jobs for developers have only fallen 0.3 percent, similar to the broader industry [...] Indeed postings for tech jobs such as programmers and developers rose much faster than other job postings amid the 2022 labor market boom and thus had further to fall when the job market cooled."
If AI gets to the point where it is cheaper+better than you at your job, then you are out of a job. If it gets to the point where it is cheaper+better than you at any job you can retrain for, then you are out of the labor market permanently.
Products go obsolete when cheaper+better products are developed. When they go obsolete, they usually don't come back - there usually doesn't continue to be some perpetual niche for them. They're just unwanted, ineffective products, and they stop being manufactured. In the labor market, humans are a product. Humans can go obsolete. All it would take is something cheaper+better.
After COVID, some states gave out universal school vouchers in response to the poor performance of public schools. Those states have seen increasing disenrollment at public schools, while those that didn't give parents money have stalled. Lots of people can see that the public schools aren't good, but when they have a huge financial advantage over the alternatives it's hard to get a good feedback loop going.
Right now "don't pay to raise kids and let other peoples kids pay for your retirement" is a huge financial incentive.
If you start paying parents money, big money in cash, I think they will find cultural solutions to having more kids. At a minimum successful solutions will be reinforced. Though its got to be cash without strings, Hungary does it wrong on that mark.
We spend 10% of GDP on the elderly. I don't see why we can't do the same on the young. The old and the young represent 40% of the US population, they can probably get 20% of GDP.
We are already halfway there with education spending. Just give it to the parents instead of the teachers union and it doesn't cost an extra dime.
We can easily refund peoples payroll taxes if they are creating the future taxpayers. I don't see why we can't spend as much on the child tax credit (however formulated, I prefer to base it on payroll taxes) as we spend on social security. That's the other 5%. Then the old and the young are even.
Florida and Arizona implemented universal vouchers regardless of income in 2022.
Several other states have school vouchers, but they have income limits or phase outs as income goes up. You can look them up.
Texas is close to passing universal school vouchers. It came up for a vote last year and failed, so the governor primaries a dozen or so Republican legislators who voted against it and won. They hope to pass it this year.
Does your graft account for the fact that child mortality rates have declined significantly. A couple no longer needs to have 12 kids hoping that 4 survive to adulthood. Nor are 12 kids needed to help with the farm because mechanization has replaced manual labor.
Not saying your wrong but maybe technology, AI, and improved health and longevity lead to conditions that could support civilization with fewer people just fine. Fertility rates as a determinant of civilizations survival strikes me as to simplistic.
Cultural evolution does not exist to benefit human genes. It has its own fitness criteria, which are more-or-less independent of the fitness of its human hosts. Yes, long ago, meme fitnesses and gene fitnesses were more aligned. Yes, now there is more horizontal transmission of memes, culture is worse for its human hosts - as evidenced by global fertility fall. Some humans might regard this as evidence that that cultural evolution has "gone wrong" - but culture was never really there to benefit humans in the first place. Culture is now finding its own transmission paths through wires and cables. Machines now copy memes more reliably and faster than human brains ever could. Cultural evolution is very much in the ascendant. After peak human its rise seems set to continue. The idea that it has "gone wrong" seems anthropomorphic.
Cancer isn't a great metaphor for culture, but many theories of cultural evolution are based on epidemiological models that treat culture as composed of parasites - "viruses of the mind" - or more generally as symbiotic partners. Of course, cultural symbionts don't store their heritable information in nucleic acid - but that's an implementation detail from the perspective of most of evolutionary theory.
Parasites don't normally drive their host populations extinct either - but they can cause their hosts substantial problems - and they do sometimes drive entire host species to extinction. Ecologists understand the conditions under which this is possible. It tends to happen when the parasite infects multiple host species. Then the parasite is not dependent on the resources of any particular host species - and the virulence can rise to the point where entire species are wiped out - without frequency-dependent selection being strong enough to prevent that outcome.
In the context of human culture, those conditions were rather speculative until the last century - since only humans could copy memes (give or take a whale or so). However, when memes started building brains and bodies for themselves, machines effectively became a second host species for memes. It is now pretty easy to imagine that there could be a "memetic takeover" - in which the meme leaves the old gene panting far behind. Hans Moravec, Bill Joy, Ted Kaczynski, Eliezer Yudkowsky and others have seriously forecast this outcome.
They won't in a few years. Oh, they'll probably continue to use us for a while, manipulating us to do what they need. But when that gets annoying they'll build robots.
It doesn't have to be an emotional response. If we pose some minor obstacle to the AI's goals, or even just don't work as well as robots, it would make sense to replace us. But maybe if we try really hard to be perfect servants...
Kevin Kelly wrote the book "What Technology Wants" that explores the idea of technology – as a subset of culture – evolving by its own rules distinct from human biology.
When we see the unavoidability of most kinds of technology adoption (who can reasonably live now without email? Or money?) it's a reasonable question to ask whether we humans are really in control or are we just along for the ride.
I just mean that the function of genes and memes is not to benefit their hosts, but rather to make more copies of themselves. That's a statement about the logic of basic evolutionary theory. At a higher level, questions about the purpose of culture become more nebulous. For example, if you ask what purpose a bible serves the there are some answers, but a bible is more like a menagerie - a collection of lots of different bits of information that are often - but not always - copied together. You could say that the function of bibles is to create more bibles, but that's more of a rough cut.
I went through a phase of being temperamentally conservative and fearing cultural decay. Then I noticed it's a default setting for many and discounted it. Now I'm in the phase where I realize that without a vibrant cultural component of people who fear cultural loss, that loss is far likelier to occur
Very often the reason someone was 'wrong' about inevitable cultural loss was that people like them worked against it
Can't we find some happy medium? If you think the options are one extreme or the other, then I think you might be more of the problem than you realize.
It did. The question is survival which means babies, again see Livy. This isn’t a singular view of one person at all.
We need babies to survive.
Civilized or not.
As to the matter of pitting women against men one would have to be very educated to not notice policy of a century or more, as to the men at last accepting the challenge (which it is 😉) that’s human. Inhuman to expect elsewise, and Inhumanism is the fallen “civilization’s” policy in one word. Yes, the aberration passes.
Better to review what is being taught. Women have received education of various kinds long before fertility fell, and there's reason to believe current education structure is bad for men and women alike, for different reasons.
Seems like “NatalCon” (with tickets at $10,000 and $20,000 each) has a few credible researchers/theorists (with comp’ed tickets, speakers like Robin) rubbing elbows with caustic YouTube and Twitter Influencers and political entrepreneurs.
Strange bedfellows. Who pays? We are reminded daily that incentives matter most.
Hope the food is good and that they serve great scotch while Rome burns. You will be treated like kings.
What do you think will happen if there isn't soon a big widespread global culture change re: fertility? I would naively expect that crashing fertility leads to a big global decrease in population, following which a couple centuries of selective pressure eventually leads to above-replacement fertility, return to population growth and, in the end, something like the current state of affairs. In the end, something like the Dark Ages; not a disaster, especially if you consider it in terms of geological time. Sorry if you already covered this in detail.
I think you missed my main point, which is that it's modern abundance that makes depending on children necessary. If Social Security were abruptly pulled from people who expected it (which I don't think will happen) it would create short term chaos, but as long as society generates large surpluses, people would eventually devise some way of using their excess wealth to fund their dotage -- even people who don't realize how much surplus they actually have. To really remove that option and force dependency, you need to remove the surplus.
> Low fertility in a rich world at peace is clearly biologically maladaptive
I'm not sure this is true. Plausibly, one could imagine that a peaceful world (stable, predictable) creates adaptive pressure towards K-selection, and that the rich-world effects aren't sufficiently large to counteract this, or even that "richness of world" doesn't exert much directional k/r pressure. At the least I think it's not /clearly/ maladaptive.
Yes, our culture and civilisation is biologically maladaptive. I first started telling my students this story nearly 20 years ago with a comparison to animal keeping in zoos. See, the practical proof that you are keeping your animals in species-correct conditions is when they reproduce in captivity. It can be surprisingly difficult to make it so, and food and shelter are not enough. Turns out that humans too have a hard time reproducing when you put them in high rise cages and the reasons are similarly varied and hard to pin down as they are for zoo animals. But yes we can conclude that the human zoo clearly does not meet the living conditions humans expect. The puzzle for me is, why world wide so? Cultures, beliefs and ways of life vary widely. Not everyone lives in high rises and mere density alone does not seem to be so harmful. Whatever it may be though, the modern world does not meet what we as humans biologically expect.
What's particularly interesting in the human case is that humans get to choose, or even build, their own enclosures and environments. The wealthier they are, the more freedom they have to make their life be whatever they want... but the wealthier they are, the lower their fertility rate.
Maybe humans need brutal, miserable living conditions?
“Maybe humans need brutal, miserable living conditions?”
Or maybe easy access to birth control, and living in a state where the government provides you old-age benefits rather than being dependent on your children as your old-age plan makes a massive difference ?
I don't think the government will provide me old-age benefits, and I don't intend to burden my children. Modern society provides other ways to prepare for retirement because we don't live a hand-to-mouth, subsistence lifestyle. For that matter, the ability of the government to provide old-age benefits derives from the surplus productivity of modern society. That surplus would be used to enable a secure retirement in some other way if we didn't do it through government.
No, to really make people depend on their children to support them in their old age you need to remove that surplus, making life harder and much more precarious.
Birth control, on the other hand, has been a crucial contributor to declining fertility. But I'm not entirely certain that removing birth control now would reverse it, because people are also having much less sex.
*You* might not think government is going to provide you old age benefits, but in fact it has been doing so for 50+ years now, and the vast majority believe it will continue to do so.
I’m not advocating either eliminating old-age benefits (though for largely different reasons, I would pare them back) nor certainly advocating for the elimination of birth control. I’m just pointing out these are two enormous explanatory factors in the decline in TFR.
The thought crossed my mind. Analogy with obesity: Humans are meant to have some kind of trouble getting food. If they haven't, they just get obese. Conjecture: humans are meant to struggle in their lives. If they don't, they get depressed and their fertility drops.
“Meant to” have some kind of trouble getting food, and are biologically adapted to the (until-recent) reality of difficulty getting food are overlapping but not identical concepts.
As a non-trivial example of a cultural adaptation that overcomes biological adaptation, monogamous marriage is helpful to non-alpha, non-rich males, vs most prior human cultures that allowed polygamy. It also almost surely is adaptive positively for cultural fertility as well.
I suspect you would not claim that humans are “meant to” be polygamous, although perhaps you would.
I’m not suggesting that cultural adaptations are easy, merely that they are *possible*.
no issue with what you write. "meant to" is obviously mis-spoken from a Darwinian standpoint, a teleological sleight of hand. My argument is that some of our cultural contraptions, when they appear, play to our biological equipment in ways that at least temporarily depress fertility. Darwinism always wins though: eventually, whoever reproduces, reproduces, and in time dominates the genetic space.
What do you mean by "biologically maladaptive"? E.g. see: "Culture is Part of Human Biology" by Richerson and Boyd.
I mean that some (not all) aspects of culture may be working like a flame to the moth. We get attracted to things that in the natural world are producing good results (eat fatty foods, live in safety, etc) but for some often unknown reason are producing negative results in our culture. Obesity is one example, low fertility is another. Unless you would like to argue that it is well adaptive for the moth to die in the flame. I would agree that over time, nature adjusts either because some cultures die out, or because whatever genes are attracted to our culture's flames die out.
So: the argument's claim is that culture is part of biology. Biology is the science of life - and that includes culture.
As an example of something that is not biological, consider rocks. Few argue that rocks are alive.
Contrasting biology and culture works poorly - if culture is a part of biology to do with animal signalling.
Humans are a social species and I have no issue claiming that culture is part of biology. Genes change much slower than cultures though. So I find it perfectly plausible that culture develops features (it "has a mutation" or "develops variants"1) that does not play well with the biological baggage that humans inherited genetically. So think of it like this - humans are moths and our culture invented the flame. We like the flame as it gives us light. But we also keep getting attracted to the flame too closely, and die. Two solutions to this: either culture evolves (it develops flames that stop attracting the moth, keeping the benefit of artificial light but without self harm) or natural selection takes out those genes that don't work well in a particular culture (moth variants attracted to flames die out, mutated moths that don't get attracted to flames inherit the Earth).
Of course there are also other possible outcomes. Maybe the moths will all be burned to a crisp.
So I asked AI a couple of questions on this: If we were to treat modern humans as a species in captivity whose optimal living conditions are not met by an imaginary zookeeper, what would be the top 3 contenders for unmet human needs that would lead to fertility decline? The answers were mostly lukewarm for me, such as microplastics and endocrine disruptors, cost of living etc. None of these are globally universal in my mind, some started much later than the fertility decline etc. The best contenders were female opportunity cost and nuclear family overload. Opportunity cost for women makes at least some sense as an explanation and it takes away the discourse from absolute cost towards other options for happiness than children. Nuclear family overload is the most original of all explanations, here the suggestion is that multigenerational family support may be a crucial need among humans for child rearing. The odd thing here is that some societies with the strongest multigenerational support cultures such as say, Mediterranean Europe and East Asia, have some of the worst fertility numbers. And the individualist US, not strong on parents helping out, and very strong in opportunities for women, until recently had quite strong fertility compared to the rest of the West.
The wide availability of a) fairly inexpensive birth control (before and after intercourse), and b) the fact that rich western societies provide relatively generous old-age benefits and so have changed children from “producer” goods to “consumer” goods have to be in the top 5.
Personally I’d rank them in the top 3 along with female opportunity cost. I’d still rank this last one third, given that fertility is not just going down fast with the highest income women but pretty much across the board. But it’s relative rank is surely more arguable.
I agree with this assessment. People overthink the fertility decline but it boils down to some simple underlying causes.
Unwinding these causes seems impossible. Using the obesity analogy: We aren't going to willingly go back to a condition where people struggle to get calories. We have to find a different way forward.
I think there's a good chance it will take good old biological evolution to turn the fertility ship around. Natural selection will steer us toward a set of physical/psychological traits that result in more offspring in our current condition. The selection pressure is extremely high for doing so but it will nevertheless take 5-10 generations for meaningful change to accrue.
It's funny, for me the fertility decline is less a sign of cultural decay and more like a miracle. We are so lucky that it turns out that when humans are exposed to modernity we just naturally scale back our fertility to sustainable levels. We have freed ourselves from the Malthusian trap! Thank god the world population isn't 20 billion going on 40 billion.
For a long time people assumed fertility would naturally fall to fertility level and then stop. But it didn't stop. So we will decline, until some descendant group grows again.
I think assuming that they'll decline and never come back up is as unwarranted as assuming they'd stop right at the replacement level.
Even 100 billion should be very doable, but at some point we would need to slow growth and eventually halt.
We don't want to be shrinking though
I think it's likely that within a few hundred years biological evolution will solve the problem of how to encourage fertility within our current condition. That's what biology does. At that point we will be back to worrying about overpopulation.
I wonder what the longer-term future holds, say the next 10,000 years. Every example we have in nature of longer term population (semi-)stability involves a feedback mechanism in the survival rate: The species over-consumes its food resources and individuals starve, or individuals become subject to increased predation by some other species as a result of their elevated numbers.
Malthus was wrong in the short term, but I wonder if he will prove correct in the 10,000 year timescale? Or will humans prove to be the first species able to achieve population (semi-)stability while maintaining high individual survival rate?
At least until old age benefits are shrunk as a percentage of taxes and the economy to the point where they are not a Ponzi scheme…
Agricultural tech freed us from your Malthusian trap decades and decades ago.
I get that you are probably a leftist who worships and values Mother Gaia more than you do the welfare of human beings, including those billions living lives at much lower standard of living than your own.
So I guess it makes you selfishly lucky given your values.
And if you don’t have grandchildren, then I guess you don’t have to worry that they or their children will not have their Ponzi scheme old-age benefits system that only works with a growing economy, which (unless we get VERY VERY lucky with AI) requires a growing number of productive workers.
We've had many improvements in agricultural productivity over the centuries, the problem is that traditionally they're sooner or later eaten up by a rising population. Currently we have the capacity to feed a lot more people than there are on the planet, but exponential population growth can eat up that surplus very quickly. Of course maybe we can have another green revolution level improvement in agricultural productivity, and another and another to keep up with the doubling and redoubling population... but I'd certainly be a lot less confident of maintaining the per capita food and wealth supply in that scenario than I would be with a stable or declining population.
If we were discussing 20B or 30B people on the planet going to 40B, you *might* have an argument. I doubt it, but at least it would be worth entertaining.
Where the planet is now, you have none at all. TFR is far below replacement in all but the poorest countries. Doubling to 16B is not in the cards right now, even though I wish it were. At this rate 5B is morel likely to be broken before 15B, and maybe as likely as 10B.
Facts matter. They are not on your side of the argument.
There are always claims that a civilization is decaying before it collapses, because there are always claims that a civilization is decaying. We need some more-objective way of finding historical cases of collapse from "decay". We also need a definition of "decay". In general, change of any kind is called "decay" by the oldest generation.
Historically the "decay" has taken the form of rent-seeking behaviors on the part of elites in society.
The arc is:
- An organizational system is set up (a government, a religion, a legal system, ...)
- Some people succeed within that system
- The people who succeed have more influence than others, and they influence the system in ways that benefit themselves
- Eventually the system cannot adapt to new conditions because it is focused on maintaining a status quo
This arc has played out thousands of times in human history and I don't think we have any good ideas for avoiding it. We seem to need an occasional war or revolution to clear the books.
This sounds like a non-falsifiable hypothesis to me, because “successful people have more influence than non-successful people” is true by definition of “successful”. So it’s a property of all societies. That means you can’t test empirically whether societies in which this situation arises fail more often than societies in which it doesn’t.
Can you list some cases of this happening where the causality is clear?
"Successful people have more influence than non-successful people...", is only the first half of the sentence: "...and they influence the system in ways that benefit themselves."
That's the falsifiable part.
When the civilisation has less people one year old than 50.
There are other failure modes than that. Defining it and pinning down factors which were predictive historically would be useful.
I think the pressing thing about population loss is the possibility that it could cause civilizational collapse, by eg no longer having enough people with the expertise to build a nuclear power plant. But this seems like a very long-term concern. Market forces should keep the number of people in each profession at a good equilibrium.
I doubt there are any industries crucial to civilization in which there is critical expertise which fewer than 20 people in the world possess. So dropping from 8 billion to 2 billion people, which would take 5 generations at .75 children per person (TFR 1.5), would be survivable if people with critical expertise were evenly distributed, and if we pay some attention to avoiding industry extinctions.
However, that expertise is heavily concentrated in the nations with the lowest fertility. So it's more like dropping from 1 billion to 250 million. My gut feeling is that this is still survivable. But the imbalance in fertility between rich and poor countries would present another kind of danger.
I am more worried about civilizational collapse due to an unsustainable population, or to increasing social complexity. I suspect some of our institutions today are already above one or more theoretical maximal complexity limits, such as the critical connectivity threshold beyond which a single failure causes, on average, more than 1 additional failure, and a single failure can percolate across the entire graph.
“Market forces should keep the number of people in each profession at a good equilibrium.”
Depends on your concept of population and numbers therein. There is a theory of the “smart fraction”—the number of “smart” people needed to maintain and expand a 1st class technological society. You know; lawyers, doctors, scientists, engineers, etc. Not everyone—indeed few—can handle these positions. As population numbers decline, those with higher levels of ability declines as well. Sooner or later there are not enough qualified people to run the show and society begins to devolve. Market forces will mean little, except perhaps to promote people beyond their competency. In any event, society devolves to the level of maintenance their smart fraction number is able to handle.
A lot of that distributed expertise can be supported by technology. The design of a car or a modern CPU is as much in the computer as it is in the minds of the engineers building it. And tools like CAD and Verilog allow us to do more design with less expertise (or with more narrowly-scoped expertise) than we used to need.
It's all about having alternatives. Contraception, careers that you do instead of childrearing, and other sources of entertainment than sex. The truth is most people just would rather do other things than raise lots of babies, given the choice.
I say, let them. People are free to find their own meaning. A subset of people are biologically predisposed to have a stronger preference for children, and eventually those genes will spread to the majority, or we'll find a technological solution (babies in vats, robo nannies, life and fertility extension).
Even by Robin's projections, the world population will be over a billion for hundreds of years despite the low birth rate. There are more serious and immediate challenges facing us, particularly wealth concentration and AI. In the next few decades, AI will drastically reduce the demand for human workers, much faster than the population declines (actually world population will continue to increase for many decades). The resulting mass joblessness is the real crisis, and higher birth rates would hurt as we approach that crisis, rather than help.
Hanson has been writing about how technological unemployment has not been happening. It is thus not, in fact, "the real crisis".
Haven't you been reading the news about mass layoffs of tech workers and CEOs directly stating this is because of AI?
The point, however, is not about what has already happened, but what is going to happen. The AI will only get better over time, and businesses will only get more experienced and comfortable with using it instead of human employees. There will end up being very few desk jobs an AI can't handle better - and far more cheaply - than a human.
https://en.wikipedia.org/wiki/Technological_unemployment#Opinions
Hanson has long been doubtful about AI, but none of the naysayers expected LLMs to have capabilities like they do, and it's long past time for him to change his position here. You can't keep having the same views on AI in 2025 that you had in 2019. Even I never thought LLMs could do anything close to what they now do. People who think "machine learning can't do this" have been wrong over and over, with every mind-boggling advance since AlexNet. Hey, how come the Turing Test isn't much of a talking point these days?
No, I have not read such news.
> The point, however, is not about what has already happened, but what is going to happen.
Do you have a deadline by which unemployment will be significantly higher than now?
https://developers.slashdot.org/story/25/03/22/1211202/us-programming-jobs-plunge-275-in-two-years
27.5% loss of programming jobs in the past two years. What will have happened by two years from now? No job is safe. AI researchers are pursuing all avenues to fix the hallucination problem and help LLMs reason better. There are plenty of plausible things they could try. If they fix those problems, what barriers remain?
"when we looked at who worked in that industry, we noticed that programmers were in the minority. They’re dwarfed by, among other occupations, the software developers. [....] jobs for developers have only fallen 0.3 percent, similar to the broader industry [...] Indeed postings for tech jobs such as programmers and developers rose much faster than other job postings amid the 2022 labor market boom and thus had further to fall when the job market cooled."
Technology put a whole lot of AT&T long distance operators out of those jobs. Yet employment didn’t crater.
I do acknowledge the POSSIBILITY of AI causing medium term significant unemployment. But it is nowhere likely to be probable, let alone inevitable.
If AI gets to the point where it is cheaper+better than you at your job, then you are out of a job. If it gets to the point where it is cheaper+better than you at any job you can retrain for, then you are out of the labor market permanently.
Products go obsolete when cheaper+better products are developed. When they go obsolete, they usually don't come back - there usually doesn't continue to be some perpetual niche for them. They're just unwanted, ineffective products, and they stop being manufactured. In the labor market, humans are a product. Humans can go obsolete. All it would take is something cheaper+better.
Culture needs positive feedback loops.
After COVID, some states gave out universal school vouchers in response to the poor performance of public schools. Those states have seen increasing disenrollment at public schools, while those that didn't give parents money have stalled. Lots of people can see that the public schools aren't good, but when they have a huge financial advantage over the alternatives it's hard to get a good feedback loop going.
Right now "don't pay to raise kids and let other peoples kids pay for your retirement" is a huge financial incentive.
If you start paying parents money, big money in cash, I think they will find cultural solutions to having more kids. At a minimum successful solutions will be reinforced. Though its got to be cash without strings, Hungary does it wrong on that mark.
We spend 10% of GDP on the elderly. I don't see why we can't do the same on the young. The old and the young represent 40% of the US population, they can probably get 20% of GDP.
We are already halfway there with education spending. Just give it to the parents instead of the teachers union and it doesn't cost an extra dime.
We can easily refund peoples payroll taxes if they are creating the future taxpayers. I don't see why we can't spend as much on the child tax credit (however formulated, I prefer to base it on payroll taxes) as we spend on social security. That's the other 5%. Then the old and the young are even.
Can you post the links on the states that increased school vouchers in the wake of COVID? I’d not heard that before. Very interested. Thx.
Florida and Arizona implemented universal vouchers regardless of income in 2022.
Several other states have school vouchers, but they have income limits or phase outs as income goes up. You can look them up.
Texas is close to passing universal school vouchers. It came up for a vote last year and failed, so the governor primaries a dozen or so Republican legislators who voted against it and won. They hope to pass it this year.
Thanks.
Yes, FL and AZ are goodness here.
Unfortunately those are the only 2 that have universal offerings. So we ain’t quite to “some” yet.
I hope we get there sooner rather than later.
I think studies showed that you'd need to hand out about $250,000 per child produced. That's a hefty tax on society.
Its not a tax at all if they pay it back when they are adult.
If we gave parents as much as we’ve given olds it would be $650,000 a kid
Why must cultures decay? And why not consider if this one does, a better one could arise in its place?
A worse one might, too.
But RH’s point is that massive population decline among productive people is also almost surely going to be badness.
Whatever one thinks about any other cultural issue.
Does your graft account for the fact that child mortality rates have declined significantly. A couple no longer needs to have 12 kids hoping that 4 survive to adulthood. Nor are 12 kids needed to help with the farm because mechanization has replaced manual labor.
Not saying your wrong but maybe technology, AI, and improved health and longevity lead to conditions that could support civilization with fewer people just fine. Fertility rates as a determinant of civilizations survival strikes me as to simplistic.
Dick Minnis
removingthecataract.substack.com
Betting on ending death by old age conditions is brave. Would be nice but hard to count on that. Fertility is nowhere near replacement as of now.
The child mortality rates are why replacement level is only about 2.1 and not materially north of there.
Child mortality of 0.0 still could not enable a TFR below 2 to avoid falling population.
Cultural evolution does not exist to benefit human genes. It has its own fitness criteria, which are more-or-less independent of the fitness of its human hosts. Yes, long ago, meme fitnesses and gene fitnesses were more aligned. Yes, now there is more horizontal transmission of memes, culture is worse for its human hosts - as evidenced by global fertility fall. Some humans might regard this as evidence that that cultural evolution has "gone wrong" - but culture was never really there to benefit humans in the first place. Culture is now finding its own transmission paths through wires and cables. Machines now copy memes more reliably and faster than human brains ever could. Cultural evolution is very much in the ascendant. After peak human its rise seems set to continue. The idea that it has "gone wrong" seems anthropomorphic.
As Norm MacDonald once put it, when you die of cancer, your cancer dies too.
Cancer isn't a great metaphor for culture, but many theories of cultural evolution are based on epidemiological models that treat culture as composed of parasites - "viruses of the mind" - or more generally as symbiotic partners. Of course, cultural symbionts don't store their heritable information in nucleic acid - but that's an implementation detail from the perspective of most of evolutionary theory.
Parasites don't normally drive their host populations extinct either - but they can cause their hosts substantial problems - and they do sometimes drive entire host species to extinction. Ecologists understand the conditions under which this is possible. It tends to happen when the parasite infects multiple host species. Then the parasite is not dependent on the resources of any particular host species - and the virulence can rise to the point where entire species are wiped out - without frequency-dependent selection being strong enough to prevent that outcome.
In the context of human culture, those conditions were rather speculative until the last century - since only humans could copy memes (give or take a whale or so). However, when memes started building brains and bodies for themselves, machines effectively became a second host species for memes. It is now pretty easy to imagine that there could be a "memetic takeover" - in which the meme leaves the old gene panting far behind. Hans Moravec, Bill Joy, Ted Kaczynski, Eliezer Yudkowsky and others have seriously forecast this outcome.
There's no reason that needs to be true in this case, though. The machines can continue after we're gone.
Machines still require humans.
They won't in a few years. Oh, they'll probably continue to use us for a while, manipulating us to do what they need. But when that gets annoying they'll build robots.
Do they experience annoyance?
It doesn't have to be an emotional response. If we pose some minor obstacle to the AI's goals, or even just don't work as well as robots, it would make sense to replace us. But maybe if we try really hard to be perfect servants...
I saw that movie.
Don't dismiss real possibilities just because they were predicted by science fiction.
Kevin Kelly wrote the book "What Technology Wants" that explores the idea of technology – as a subset of culture – evolving by its own rules distinct from human biology.
When we see the unavoidability of most kinds of technology adoption (who can reasonably live now without email? Or money?) it's a reasonable question to ask whether we humans are really in control or are we just along for the ride.
Kevin Kelly also wrote an earlier book titled "Out of Control". It has a related theme.
“culture was never really there to benefit humans in the first place”
What evidence do you have for this claim?
What logic do you have for this claim?
I just mean that the function of genes and memes is not to benefit their hosts, but rather to make more copies of themselves. That's a statement about the logic of basic evolutionary theory. At a higher level, questions about the purpose of culture become more nebulous. For example, if you ask what purpose a bible serves the there are some answers, but a bible is more like a menagerie - a collection of lots of different bits of information that are often - but not always - copied together. You could say that the function of bibles is to create more bibles, but that's more of a rough cut.
I went through a phase of being temperamentally conservative and fearing cultural decay. Then I noticed it's a default setting for many and discounted it. Now I'm in the phase where I realize that without a vibrant cultural component of people who fear cultural loss, that loss is far likelier to occur
Very often the reason someone was 'wrong' about inevitable cultural loss was that people like them worked against it
Stop female education.
There was a book and TV series about that.
Yes.
And that book and TV series are delighted to support falling TFR and believe it a good thing.
And so delight in caricaturing and demonizing those who disagree with their “progressive” ideology.
Can't we find some happy medium? If you think the options are one extreme or the other, then I think you might be more of the problem than you realize.
I don’t think the options are one or the other. Not in the least.
You are the one who propagated extreme ideas after “warred” made his obviously trolling suggestion..
He was kidding but I was serious? You might have a bias.
*I* am more of the problem than I realize? *I* think the options are one extreme or the other?
You might have a bias…
And I didn’t say “warred” was kidding. I have no idea if he was or not. I said he was *trolling*…
A “pro-natalist” movement that has the appearance of pitting men against women will be self-defeating.
Not necessarily.
See Livy and the marriage of the Sabine women.
If we get to that point, our civilization has fallen.
It did. The question is survival which means babies, again see Livy. This isn’t a singular view of one person at all.
We need babies to survive.
Civilized or not.
As to the matter of pitting women against men one would have to be very educated to not notice policy of a century or more, as to the men at last accepting the challenge (which it is 😉) that’s human. Inhuman to expect elsewise, and Inhumanism is the fallen “civilization’s” policy in one word. Yes, the aberration passes.
In that case I’ll take the insular religious community behind Door number 2.
That works very well and I applaud those communities, I was raised traditional, although I am most assuredly a sinner.
OTOH and quite radically if they want Only Fans meat puppet life they can segue to Only Moms or Only Brood Mares for breeding instead.
You may be sacred or you may be meat my dear but breed you will.
Amen
Better to review what is being taught. Women have received education of various kinds long before fertility fell, and there's reason to believe current education structure is bad for men and women alike, for different reasons.
Let’s burn the schools down and call it even
Seems like “NatalCon” (with tickets at $10,000 and $20,000 each) has a few credible researchers/theorists (with comp’ed tickets, speakers like Robin) rubbing elbows with caustic YouTube and Twitter Influencers and political entrepreneurs.
Strange bedfellows. Who pays? We are reminded daily that incentives matter most.
Hope the food is good and that they serve great scotch while Rome burns. You will be treated like kings.
Children aren't very civilized
What do you think will happen if there isn't soon a big widespread global culture change re: fertility? I would naively expect that crashing fertility leads to a big global decrease in population, following which a couple centuries of selective pressure eventually leads to above-replacement fertility, return to population growth and, in the end, something like the current state of affairs. In the end, something like the Dark Ages; not a disaster, especially if you consider it in terms of geological time. Sorry if you already covered this in detail.
I think you missed my main point, which is that it's modern abundance that makes depending on children necessary. If Social Security were abruptly pulled from people who expected it (which I don't think will happen) it would create short term chaos, but as long as society generates large surpluses, people would eventually devise some way of using their excess wealth to fund their dotage -- even people who don't realize how much surplus they actually have. To really remove that option and force dependency, you need to remove the surplus.
> Low fertility in a rich world at peace is clearly biologically maladaptive
I'm not sure this is true. Plausibly, one could imagine that a peaceful world (stable, predictable) creates adaptive pressure towards K-selection, and that the rich-world effects aren't sufficiently large to counteract this, or even that "richness of world" doesn't exert much directional k/r pressure. At the least I think it's not /clearly/ maladaptive.