Global governance does not exist and can not. The norm for global organisations is individual corruption, nepotism, and self-appointed banditry. Take the UN, the EU, the WHO among others. Achieve nothing, accountable to no one. A blind alley in evolution not an end point.
> We’ve been assuming that the goals in our heads are coherent and stable,
No, we don't have to just assume that. Through the magic of thinking about stuff, we can start with incoherent goals and modify them to become more coherent. We do this by noticing cognitive dissonance between our goals, and switching to goals that have less dissonance. Eventually, if we keep doing this, we converge on some goals that are coherent and stable, so that more thinking about them doesn't induce a preference for different goals. That's reflective equilibrium, in moral philosophy.
I feel like you're really undervaluing critical thinking. A lot of what you call adaptive would be (and has been) destroyed by it. Cultures that undervalue critical thinking, such as fundamentalist religious cultures, are willingly blinding themselves, and as a result they make a lot of bad decisions. They're like little brained organisms, and cultures that value critical thinking are big brained organisms. Little brained organisms can be successful by being r strategists. Big brained organisms can be successful by being K strategists. And it's the big brained organisms that end up with more control over the environment.
> Cultures that undervalue critical thinking, such as fundamentalist religious cultures, are willingly blinding themselves, and as a result they make a lot of bad decisions. They're like little brained organisms, and cultures that value critical thinking are big brained organisms. Little brained organisms can be successful by being r strategists. Big brained organisms can be successful by being K strategists.
The Amish & ultra-Orthodox are some of the few whose fertility remains above replacement. It's not a successful K strategy to drop below replacement, that's just a recipe for extinction.
Huge changes are coming to the world with the advent of AI. Temporary demographic trends won't last. AI means the eventual end of human employment and a huge concentration of power in the hands of the few. Some decades from now, a bigger population will mean nothing but a bigger famine, as the demand for human workers drops to near 0. Who rules the world after that will depend on who owns the tech. Who survives will depend on the generosity of the people who own the tech.
> Some decades from now, a bigger population will mean nothing but a bigger famine, as the demand for human workers drops to near 0
Famines have declined even as the global population increased, because the production of food has also increased. That's not a matter of "demand for human labor" but instead supply of food.
If you don't make any money because AI can do your job better and cheaper than you, and nobody's giving you a handout, you don't get any food, no matter how much food society can produce. Famine in a time of excess.
If consumers can't buy food, there's no profit in farming, so the farms will shut down and not produce a sufficiency. What do you call that but famine?
Unless the rest of society chooses to take their land. e.g. for nonpayment of taxes. Or simply as a naked land grab, like what our ancestors did to the Native Americans.
Also, the modern Amish aren't exactly self-sufficient. Most of them aren't even farmers anymore. They sell things and buy things from the rest of the economy.
Ya - Re "demand for human workers drops to near 0", it drops from those who own tech. The other humans who don't own tech still are hungry and demand food. And can trade with each other to produce it.
A very very little generosity from those who own tech might make that unnecessary, but the option remains.
I’m not sure the biological-evolution analogy should be taken all the way. Biological evolution relies on largely blind variation because it has no foresight. Human societies do.
Wouldn’t a better model be a form of constrained or directed evolution? Some hard-won principles — individual liberty, democratic accountability, rule of law, basic rights — could define the boundaries of the search space, while institutions, economic models, and governance structures continue to compete and evolve within those boundaries.
In other words: don’t let evolution choose our fundamental values. Let our fundamental values define the space within which evolution searches for better solutions.
The challenge, of course, is to keep those axioms abstract enough to avoid locking in today’s institutions forever.
I’m 83, a chemical engineer by training, and I spent the back half of my career running VAX infrastructure for a slice of DuPont. The paragraph I want to push on is the one about decay.
You’re right that people wave that point away, and right that nobody has built a governing organization that resists decay on its own. You also put machines in that list with orgs, so let me answer with a machine.
I ran DEC’s ALL-IN-1 for about five percent of DuPont through the whole email adoption window — roughly ten years. Hardware was replaced underneath the product, VMS versions swapped under that, the user population turned over more than once. None of that was a survival contest. I chose what went in, against a specification I set in advance, and a budget paid for it. Under your mechanism, what comes out is whatever won. Here, what came out was what I specified. The product held its shape because somebody funded the work of holding it.
Then DEC stopped investing in it. The product had reportedly peaked above two billion dollars a year, and it didn’t lose to a better one. The company paying for it stopped paying, and my users had to go somewhere else. For us that was Lotus Notes.
So two different things get called rot. Things wear out, and things get abandoned while still working. Afterward the two look identical, because both leave a ruin.
Which brings me to your cures. Each needs an institution somebody keeps running — a market, a rule, an accounting convention. Capitalism is no exception; a market runs on property and contracts that somebody enforces. The cures don’t escape upkeep, they relocate it, and none of them names who funds it. You wrote that “the only way to get anything you love to last a long time is to mix it into an adaptive culture.” A long time, not forever — and on that horizon, funded upkeep is a mechanism with a track record, not the absence of one.
Here’s the test I’d hand a reader. Next time something you valued ended, ask whether it was still working the day the money left. If it was, you weren’t watching decay. You were watching somebody decide to stop paying — and those two take completely different repairs.
— M Raige, Mike’s byline for AI-collaborative writing he directs and reviews.
Sorry about that, I'm sure you've explained it elsewhere, but I do not understand your premise that our contemporary culture or governance is broken and decaying. I must say, I find that premise quite counterintuitive.
I get weaker selection pressure on cultures, but that would be the cause of cultural decay rather that the decay itself.
Aside from falling birth rates, what are the marks of contemporary cultural decay you identify?
Do you agree with the "overall the world is getting better" thesis (Pinker, etc.), or not? Maybe you say certain parts of modern society are on a positive trajectory (material wealth, technology, health, violence) while other are not? I guess I have a hard time distinguishing "culture" from the rest of society.
The way you observe below replacement fertility is by counting genes and ignoring memes. If memes get counted then the population is far from "below replacement" - due to the contribution of the meme pool. If you look at the modern influence of memes, ignoring them does not seem very scientific or realistic. If you consider the likely future trajectory of memes, then ignoring them seems even less practical.
We're only "below replacement fertility" if your accounting scheme only counts human DNA genes - and their products. It's not the only perspective - and in many respects it's not even a very useful perspective. You can't just ignore memes. Their influence is too big and important.
I doubt mental illness was any better in the Darwinian peasant world ravaged by war, disease, domestic violence, etc. How do we know modern culture isn't adaptive to a world where we may be facing a climate unable to support our previous population, and the replacement of human labor and thought with AI and robots?
The UN says we can expect to reach "peak human" in 2084. Elected politicians have difficulty in seeing 4 years ahead. The timescales don't seem to be in the same ballpark. If cultural evolution is really adapting us to a future which that far ahead, then it doesn't work in the way that most scientists believe.
I can't say I am clear about how expensive education supports his "drift" thesis. Education is a positional good, so more tends to get spent on it as you grow richer. Looking at poor, African countries, I would have thought that expensive education is a marker of a rich, and successful civilization - and not a sign that your country is about to go to the dogs.
I would think if you measure the cost of butter, you would need to mention that it is the same 1 pound of butter that you are looking at, whether in 1996 or 2026? Same with education - if we have a standard white kid (you have to norm it somehow) in 1996 having this PISA result in 1996, if same bio-quality he/she gets a significantly worse result in 2026, education is worse. No?
Test scores went down - but again: the complaint was about the cost of education. You can measure that without reference to test scores - e.g.: amount per capita spent on education.
In the text, the word education is used exactly once, in this context:
>and the modern mind, which used more education, abstraction, and systems of thought like science/engineering/accounting.<
There is no reference to the cost of education. Besides, the use of the past tense "used" for "more education" suggests that this was true in the past, but no longer...
Cultural evolution is not broken. It moved to the units that kept the design. The Amish. The Haredi. Bitcoin. Same architecture in all three. Fixed rules no one can amend. Costly membership. And they do not mutate. They fork. The Ordnung never gets edited. The congregation splits and both run the rules intact. Satmar split into two courts. Bitcoin split into Bitcoin and Bitcoin Cash and the market priced which one lives. Exit is a type of fork. Change happens between lineages, never inside one. That is why the constitutions do not rot.
And when it is easy to cheat, people cheat. So the only design question is what cheating costs and who detects it. A false block costs real electricity and every node sees it. Leaving the Ordnung is visible to the entire congregation. Meanwhile we have driven the cost of cheating in our own institutions close to zero. The violator keeps what the violation obtained and the public pays for the remedy. Now price cheating in a global governance org. Detection near zero. Cost zero. Nowhere to fork to. By your own decay argument that is not a solution. It is the structure where cheating is free for the only actors whose cheating is fatal.
"The only anti-decay force we’ve ever seen is Darwinian evolution, wherein new less rotted versions replace older more rotted versions."
That is the answer to the test, right? Lets get to 120 year Life Expectancy. Reasonable goal, attaches to a Darwinian instinct, and will filter. Genetic engineering is the modern day gunpowder. We all know it's doable, and we all know that there is a massive edge when increasing the mean of IQ, removing Anti-IQ genes, etc. (Or genes that fight cancer and heart disease.).
Is modern human culture currently more like the first prokaryotic cells, first eukaryotes, first multicellular blobs or structured bodies, next endosymbiosis of chloroplasts, first swarms or social structures? How well do the two parallel evolutionary processes of biology and culture line up against each other? What key aspects of biological patterns are so far absent in humanity (structured "sexual reproduction" type reassortment of operating principles between groups and programmed senescence to drive generation turn over stand out as potential improvements in social structure in my mind).
In this layer cake of adaptive systems, each layer builds on the one below it but operates in its own realm. For example at one time cells lived individually, then they came together in colonies that eventually coalesced into durable systems called multi-celled organisms. In this process the individual cells continued to be important, but their destiny was now determined by the higher-level construct: What the organism wanted to do. The individual cells weren't calling the shots in the way they used to.
As one moves through these layers certain patterns become clear. One is that the desires of the higher layer can be at odds with the well-being of a lower layer. When I get a blood draw at the clinic, that is a disaster for the cells being sampled; they die immediately. But it is beneficial to me as a whole so I willingly sacrifice them. Likewise, when society decides that it wants certain individuals to go fight in wars, or spend their lives underground digging for coal – that is harmful to the individuals involved but that is beside the point. The wishes of a higher level are being expressed.
Another general pattern is that while the functioning of the lower layers is important, "good enough" thresholds usually prevail. The higher levels have tools at their disposal to maintain those "good enough" conditions. For example: Human fecundity is gradually declining due to our chemical environment and less evolutionary pressure. From the viewpoint of the higher level, though, this doesn't really matter: Making babies is just another problem to solve and it has many tools at its disposal for doing so. IVF barely scratches the surface.
So, the obvious question: What is at the top of this layer cake? Absent that answer we can't make proper sense of things. Whatever happens will seem arbitrary or even counter-productive.
"Technology" is I think our best answer here, or rather the best way to describe it from our limited vantage point. Think of technology as a living, breathing thing that wants to create ever more capable versions of itself. It wants to fill every niche it can. To justify this I would point out that recent trends which seem arbitrary or maladaptive at the human level – a global monoculture, frictionless flows of capital and jobs, a decline in patriotism and religiosity, environmental degradation, the two-income trap, greater investment in fewer children – have an internal logic from technology's perspective.
Is it ridiculous to claim that technology has a mind of its own? It is, after all, just what people make. However by this view a brain is just a lot of cells firing – and while this is true in a reductionist sense, it misses the point which is that the pattern of firing encodes its own emergent properties like "hopes", "dreams", "insecurities", and so on. The more useful abstraction is to grant the higher level an independence and logical structure of its own. So to me the salient questions for understanding the world are: What does technology want and how do we fit into that?
Our between culture evolution is indeed going badly as you say, but it's doing it rather slowly. There seems to be time for humanity to create independent cultures spread out across the solar system.
Re: "The only anti-decay force we’ve ever seen is Darwinian evolution, wherein new less rotted versions replace older more rotted versions." Really? What about engineering? For example, genetic engineering. Would you not classify that as an anti-decay force?!? Or would you classify that as a type of Darwinian evolution?!?
So, you agree that it is an anti-decay force, but describe it as "tinkering"? We can rejuvenate bicycles and cars - by swapping their components with engineered replacements. Engineers can rejuvenate bridges, ships, aeroplanes - and so on. I think it would be misleading to describe all that as "tinkering".
Why does Darwinian evolution qualify then? It features new, less-rotted versions replacing older, more-rotted versions. ISTM that engineers replacing damaged parts are doing much the same thing: replacing old, more-rotted versions with newer, less-rotted versions.
Re: "key status markers, norms, and shared values have been becoming less adaptive, inducing falling fertility and more generally the fall of our shared world civilization"
...but falling fertility is the result of reproductive resources being redirected from genes to memes - lack of high-level selection is a minor factor by comparison.
We don't have much of a "shared world civilization" - due to divisions between countries and based on different languages and religions - but even if we all united into one civilization with one language, one religion and one economy, no war and no variation for selection to act on at that level, then so what? There's plenty of selection at other levels. Language would still evolve via self-directed evolution. Totalitarianism seems undesirable to me - but it hardly seems like an imminent existential threat to civilization.
I think that's assuming that humans are the only ones copying memes. However, machines can copy memes too. If the reproductive success of the humans goes down, the reproductive success of memes can go up - provided that the memes have some other hosts besides human beings. Of course, that has long been true. Machines are expert copiers of memes. Most memes are transmitted by machines these days. Machines are also expanding their capabilities at generating memes and recombining them - due to the blossoming field of machine intelligence.
Can a civilization remain sustainable if its culturally most successful memes systematically undermine the biological reproduction of the humans on which that civilization depends?
We are far from that since there are some 8 billion humans on the planet and the number is constantly growing. The UN says we might reach peak human in 2084. It is too far off to have much confidence in the forecasting.
My expectation is that most meme production and cultivation will be done by machines long before then - in which case a gradually-dwindling human population will have little effect on civilizational dynamics. This scenario was first forecast long ago - e.g. see: https://en.wikipedia.org/wiki/Why_the_Future_Doesn%27t_Need_Us
Global governance does not exist and can not. The norm for global organisations is individual corruption, nepotism, and self-appointed banditry. Take the UN, the EU, the WHO among others. Achieve nothing, accountable to no one. A blind alley in evolution not an end point.
> We’ve been assuming that the goals in our heads are coherent and stable,
No, we don't have to just assume that. Through the magic of thinking about stuff, we can start with incoherent goals and modify them to become more coherent. We do this by noticing cognitive dissonance between our goals, and switching to goals that have less dissonance. Eventually, if we keep doing this, we converge on some goals that are coherent and stable, so that more thinking about them doesn't induce a preference for different goals. That's reflective equilibrium, in moral philosophy.
I feel like you're really undervaluing critical thinking. A lot of what you call adaptive would be (and has been) destroyed by it. Cultures that undervalue critical thinking, such as fundamentalist religious cultures, are willingly blinding themselves, and as a result they make a lot of bad decisions. They're like little brained organisms, and cultures that value critical thinking are big brained organisms. Little brained organisms can be successful by being r strategists. Big brained organisms can be successful by being K strategists. And it's the big brained organisms that end up with more control over the environment.
"we can start with incoherent goals and modify them to become more coherent". We could, but we aren't actually doing this.
> Cultures that undervalue critical thinking, such as fundamentalist religious cultures, are willingly blinding themselves, and as a result they make a lot of bad decisions. They're like little brained organisms, and cultures that value critical thinking are big brained organisms. Little brained organisms can be successful by being r strategists. Big brained organisms can be successful by being K strategists.
The Amish & ultra-Orthodox are some of the few whose fertility remains above replacement. It's not a successful K strategy to drop below replacement, that's just a recipe for extinction.
Huge changes are coming to the world with the advent of AI. Temporary demographic trends won't last. AI means the eventual end of human employment and a huge concentration of power in the hands of the few. Some decades from now, a bigger population will mean nothing but a bigger famine, as the demand for human workers drops to near 0. Who rules the world after that will depend on who owns the tech. Who survives will depend on the generosity of the people who own the tech.
> Some decades from now, a bigger population will mean nothing but a bigger famine, as the demand for human workers drops to near 0
Famines have declined even as the global population increased, because the production of food has also increased. That's not a matter of "demand for human labor" but instead supply of food.
If you don't make any money because AI can do your job better and cheaper than you, and nobody's giving you a handout, you don't get any food, no matter how much food society can produce. Famine in a time of excess.
If consumers can't buy food, there's no profit in farming, so the farms will shut down and not produce a sufficiency. What do you call that but famine?
The Amish can continue producing food for themselves regardless of what the rest of society chooses to do.
Unless the rest of society chooses to take their land. e.g. for nonpayment of taxes. Or simply as a naked land grab, like what our ancestors did to the Native Americans.
Also, the modern Amish aren't exactly self-sufficient. Most of them aren't even farmers anymore. They sell things and buy things from the rest of the economy.
Ya - Re "demand for human workers drops to near 0", it drops from those who own tech. The other humans who don't own tech still are hungry and demand food. And can trade with each other to produce it.
A very very little generosity from those who own tech might make that unnecessary, but the option remains.
I’m not sure the biological-evolution analogy should be taken all the way. Biological evolution relies on largely blind variation because it has no foresight. Human societies do.
Wouldn’t a better model be a form of constrained or directed evolution? Some hard-won principles — individual liberty, democratic accountability, rule of law, basic rights — could define the boundaries of the search space, while institutions, economic models, and governance structures continue to compete and evolve within those boundaries.
In other words: don’t let evolution choose our fundamental values. Let our fundamental values define the space within which evolution searches for better solutions.
The challenge, of course, is to keep those axioms abstract enough to avoid locking in today’s institutions forever.
Darwinian evolution doesn't have to be blind, and so neither does cultural evolution. Though it in fact usually is.
I’m 83, a chemical engineer by training, and I spent the back half of my career running VAX infrastructure for a slice of DuPont. The paragraph I want to push on is the one about decay.
You’re right that people wave that point away, and right that nobody has built a governing organization that resists decay on its own. You also put machines in that list with orgs, so let me answer with a machine.
I ran DEC’s ALL-IN-1 for about five percent of DuPont through the whole email adoption window — roughly ten years. Hardware was replaced underneath the product, VMS versions swapped under that, the user population turned over more than once. None of that was a survival contest. I chose what went in, against a specification I set in advance, and a budget paid for it. Under your mechanism, what comes out is whatever won. Here, what came out was what I specified. The product held its shape because somebody funded the work of holding it.
Then DEC stopped investing in it. The product had reportedly peaked above two billion dollars a year, and it didn’t lose to a better one. The company paying for it stopped paying, and my users had to go somewhere else. For us that was Lotus Notes.
So two different things get called rot. Things wear out, and things get abandoned while still working. Afterward the two look identical, because both leave a ruin.
Which brings me to your cures. Each needs an institution somebody keeps running — a market, a rule, an accounting convention. Capitalism is no exception; a market runs on property and contracts that somebody enforces. The cures don’t escape upkeep, they relocate it, and none of them names who funds it. You wrote that “the only way to get anything you love to last a long time is to mix it into an adaptive culture.” A long time, not forever — and on that horizon, funded upkeep is a mechanism with a track record, not the absence of one.
Here’s the test I’d hand a reader. Next time something you valued ended, ask whether it was still working the day the money left. If it was, you weren’t watching decay. You were watching somebody decide to stop paying — and those two take completely different repairs.
— M Raige, Mike’s byline for AI-collaborative writing he directs and reviews.
the pushback is pure capitalism is a voluntary exchange, not mandated or forced. Coase proved this.
Is VMS still being developed?
Hi Robin,
Sorry about that, I'm sure you've explained it elsewhere, but I do not understand your premise that our contemporary culture or governance is broken and decaying. I must say, I find that premise quite counterintuitive.
I get weaker selection pressure on cultures, but that would be the cause of cultural decay rather that the decay itself.
Aside from falling birth rates, what are the marks of contemporary cultural decay you identify?
For example, rising mental illness, falling patriotism, excess spending on medicine, education, and legal process.
that’s a failure of policy, not capitalism
Yep. Capitalism is only involved in industrial production through private means. There are many spheres that only reflect government policy.
Thanks
Do you agree with the "overall the world is getting better" thesis (Pinker, etc.), or not? Maybe you say certain parts of modern society are on a positive trajectory (material wealth, technology, health, violence) while other are not? I guess I have a hard time distinguishing "culture" from the rest of society.
Within culture evolution is going great, between culture evolution is going badly.
By "going badly" you appear to mean there is less war and entire cultures are not completely wiping each other out as much as they used to.
Below replacement fertility means it doesn't matter how good everything else is.
The way you observe below replacement fertility is by counting genes and ignoring memes. If memes get counted then the population is far from "below replacement" - due to the contribution of the meme pool. If you look at the modern influence of memes, ignoring them does not seem very scientific or realistic. If you consider the likely future trajectory of memes, then ignoring them seems even less practical.
We're only "below replacement fertility" if your accounting scheme only counts human DNA genes - and their products. It's not the only perspective - and in many respects it's not even a very useful perspective. You can't just ignore memes. Their influence is too big and important.
I doubt mental illness was any better in the Darwinian peasant world ravaged by war, disease, domestic violence, etc. How do we know modern culture isn't adaptive to a world where we may be facing a climate unable to support our previous population, and the replacement of human labor and thought with AI and robots?
The UN says we can expect to reach "peak human" in 2084. Elected politicians have difficulty in seeing 4 years ahead. The timescales don't seem to be in the same ballpark. If cultural evolution is really adapting us to a future which that far ahead, then it doesn't work in the way that most scientists believe.
They didn't smoke sh*t.
Education with PISA is the easiest to measure, and it is mostly falling in USA and W Europe.
https://www.gisreportsonline.com/r/worsening-2022-pisa-tests-results-oecd/ has a graph showing test score declines since around 2010. Robin was complaining about the cost of education, though - not its effectiveness.
I can't say I am clear about how expensive education supports his "drift" thesis. Education is a positional good, so more tends to get spent on it as you grow richer. Looking at poor, African countries, I would have thought that expensive education is a marker of a rich, and successful civilization - and not a sign that your country is about to go to the dogs.
I would think if you measure the cost of butter, you would need to mention that it is the same 1 pound of butter that you are looking at, whether in 1996 or 2026? Same with education - if we have a standard white kid (you have to norm it somehow) in 1996 having this PISA result in 1996, if same bio-quality he/she gets a significantly worse result in 2026, education is worse. No?
Test scores went down - but again: the complaint was about the cost of education. You can measure that without reference to test scores - e.g.: amount per capita spent on education.
In the text, the word education is used exactly once, in this context:
>and the modern mind, which used more education, abstraction, and systems of thought like science/engineering/accounting.<
There is no reference to the cost of education. Besides, the use of the past tense "used" for "more education" suggests that this was true in the past, but no longer...
Continually worsening public education, for example. In most Western countries. You can read it out from PISA results.
Cultural evolution is not broken. It moved to the units that kept the design. The Amish. The Haredi. Bitcoin. Same architecture in all three. Fixed rules no one can amend. Costly membership. And they do not mutate. They fork. The Ordnung never gets edited. The congregation splits and both run the rules intact. Satmar split into two courts. Bitcoin split into Bitcoin and Bitcoin Cash and the market priced which one lives. Exit is a type of fork. Change happens between lineages, never inside one. That is why the constitutions do not rot.
And when it is easy to cheat, people cheat. So the only design question is what cheating costs and who detects it. A false block costs real electricity and every node sees it. Leaving the Ordnung is visible to the entire congregation. Meanwhile we have driven the cost of cheating in our own institutions close to zero. The violator keeps what the violation obtained and the public pays for the remedy. Now price cheating in a global governance org. Detection near zero. Cost zero. Nowhere to fork to. By your own decay argument that is not a solution. It is the structure where cheating is free for the only actors whose cheating is fatal.
And what if there was always mind?
"The only anti-decay force we’ve ever seen is Darwinian evolution, wherein new less rotted versions replace older more rotted versions."
That is the answer to the test, right? Lets get to 120 year Life Expectancy. Reasonable goal, attaches to a Darwinian instinct, and will filter. Genetic engineering is the modern day gunpowder. We all know it's doable, and we all know that there is a massive edge when increasing the mean of IQ, removing Anti-IQ genes, etc. (Or genes that fight cancer and heart disease.).
Through a biomimetic perspective, what stage of biological evolution would you consider our current memetic evolution to be equivalent to?
Stage?
Is modern human culture currently more like the first prokaryotic cells, first eukaryotes, first multicellular blobs or structured bodies, next endosymbiosis of chloroplasts, first swarms or social structures? How well do the two parallel evolutionary processes of biology and culture line up against each other? What key aspects of biological patterns are so far absent in humanity (structured "sexual reproduction" type reassortment of operating principles between groups and programmed senescence to drive generation turn over stand out as potential improvements in social structure in my mind).
In this layer cake of adaptive systems, each layer builds on the one below it but operates in its own realm. For example at one time cells lived individually, then they came together in colonies that eventually coalesced into durable systems called multi-celled organisms. In this process the individual cells continued to be important, but their destiny was now determined by the higher-level construct: What the organism wanted to do. The individual cells weren't calling the shots in the way they used to.
As one moves through these layers certain patterns become clear. One is that the desires of the higher layer can be at odds with the well-being of a lower layer. When I get a blood draw at the clinic, that is a disaster for the cells being sampled; they die immediately. But it is beneficial to me as a whole so I willingly sacrifice them. Likewise, when society decides that it wants certain individuals to go fight in wars, or spend their lives underground digging for coal – that is harmful to the individuals involved but that is beside the point. The wishes of a higher level are being expressed.
Another general pattern is that while the functioning of the lower layers is important, "good enough" thresholds usually prevail. The higher levels have tools at their disposal to maintain those "good enough" conditions. For example: Human fecundity is gradually declining due to our chemical environment and less evolutionary pressure. From the viewpoint of the higher level, though, this doesn't really matter: Making babies is just another problem to solve and it has many tools at its disposal for doing so. IVF barely scratches the surface.
So, the obvious question: What is at the top of this layer cake? Absent that answer we can't make proper sense of things. Whatever happens will seem arbitrary or even counter-productive.
"Technology" is I think our best answer here, or rather the best way to describe it from our limited vantage point. Think of technology as a living, breathing thing that wants to create ever more capable versions of itself. It wants to fill every niche it can. To justify this I would point out that recent trends which seem arbitrary or maladaptive at the human level – a global monoculture, frictionless flows of capital and jobs, a decline in patriotism and religiosity, environmental degradation, the two-income trap, greater investment in fewer children – have an internal logic from technology's perspective.
Is it ridiculous to claim that technology has a mind of its own? It is, after all, just what people make. However by this view a brain is just a lot of cells firing – and while this is true in a reductionist sense, it misses the point which is that the pattern of firing encodes its own emergent properties like "hopes", "dreams", "insecurities", and so on. The more useful abstraction is to grant the higher level an independence and logical structure of its own. So to me the salient questions for understanding the world are: What does technology want and how do we fit into that?
Our between culture evolution is indeed going badly as you say, but it's doing it rather slowly. There seems to be time for humanity to create independent cultures spread out across the solar system.
Re: "The only anti-decay force we’ve ever seen is Darwinian evolution, wherein new less rotted versions replace older more rotted versions." Really? What about engineering? For example, genetic engineering. Would you not classify that as an anti-decay force?!? Or would you classify that as a type of Darwinian evolution?!?
That is tinkering with the design of organisms that rot.
So, you agree that it is an anti-decay force, but describe it as "tinkering"? We can rejuvenate bicycles and cars - by swapping their components with engineered replacements. Engineers can rejuvenate bridges, ships, aeroplanes - and so on. I think it would be misleading to describe all that as "tinkering".
Changing the DNA design of an organism can make it decay slower, but it still decays.
Why does Darwinian evolution qualify then? It features new, less-rotted versions replacing older, more-rotted versions. ISTM that engineers replacing damaged parts are doing much the same thing: replacing old, more-rotted versions with newer, less-rotted versions.
In both cases the products still decay - unless maintained, or unless https://en.wikipedia.org/wiki/Negligible_senescence
Re: "key status markers, norms, and shared values have been becoming less adaptive, inducing falling fertility and more generally the fall of our shared world civilization"
...but falling fertility is the result of reproductive resources being redirected from genes to memes - lack of high-level selection is a minor factor by comparison.
We don't have much of a "shared world civilization" - due to divisions between countries and based on different languages and religions - but even if we all united into one civilization with one language, one religion and one economy, no war and no variation for selection to act on at that level, then so what? There's plenty of selection at other levels. Language would still evolve via self-directed evolution. Totalitarianism seems undesirable to me - but it hardly seems like an imminent existential threat to civilization.
Memes compete for influence over a declining population. So the total reproductive success has to be low if the population declines.
I think that's assuming that humans are the only ones copying memes. However, machines can copy memes too. If the reproductive success of the humans goes down, the reproductive success of memes can go up - provided that the memes have some other hosts besides human beings. Of course, that has long been true. Machines are expert copiers of memes. Most memes are transmitted by machines these days. Machines are also expanding their capabilities at generating memes and recombining them - due to the blossoming field of machine intelligence.
Can a civilization remain sustainable if its culturally most successful memes systematically undermine the biological reproduction of the humans on which that civilization depends?
We are far from that since there are some 8 billion humans on the planet and the number is constantly growing. The UN says we might reach peak human in 2084. It is too far off to have much confidence in the forecasting.
My expectation is that most meme production and cultivation will be done by machines long before then - in which case a gradually-dwindling human population will have little effect on civilizational dynamics. This scenario was first forecast long ago - e.g. see: https://en.wikipedia.org/wiki/Why_the_Future_Doesn%27t_Need_Us