The world’s dominant civilization today seems on track to decline and be replaced in a few centuries by what are now small insular fertile subcultures, like the Amish and Haredim.
"five good LLMs agree that this habitual elite deference to price estimates usually disappears when topics turn to key norms and values"
I've seen this phrase in a lot of your posts lately (X good LLMs agree that...), and it strikes me as bad/tainted/dirty epistemology. Even the "good" llms are sycophantic and can hallucinate, and a lot of people are going off the deep end talking to them.
There are no unflawed sources but LLMs are text generators. I use them all day and am pro AI, but citing them like that is exactly like citing the top 5 google search results. Maybe worthy of consideration, maybe a good barometer for public opinion, but not more than that.
IMO the standard practice should be to ask the SOTA LLM to generate and then separately rate questions for neutrality. Also of course avoid memory/standard web interface that leaks info about the user (try asking ChatGPT what coffee shop you are in!)
Hallucinations are way down at the SOTA but sycophancy and user bias is way more subtle and difficult to root out.
Social Darwinism is usually understood as “let the weak die so the strong can rise.” If you mean something different by it, you should speak up.
The “let the weak die” approach historically has come against more humanist cultures more than once in direct cultural confrontation and been left behind. (See the Axial
Age adaptations in Greece, India, Israel, China, and Persia; Christianity vs Roman Paganism, Christianity vs Vikings, Nazis vs The West, even early 1900s Progressives vs everyone else.) Social Darwinism is not adaptive.
Social Darwinism doesn’t require the weak to die. It’s enough for them to have sub-replacement fertility, which they already do. The challenge is to encourage the strong into higher fertility.
That’s marginally better. But in the fray of cultural evolution, it’s been the societies that care MORE for the weak that have won the day. Encouraging social Darwinism seems to be empirically pushing in an anti-fitness anti-adaptive cultural direction.
There’s an important distinction to be made. In-group altruism increases a group’s fitness because it reduces the risk of wipeouts—there’s solid math and evolutionary modeling showing how risk-sharing within a group improves survival odds.
But inter-group competition is a very different dynamic. Evolutionarily, it’s closer to a zero-sum game: in pre-industrial societies, groups and cultures couldn’t simply coexist in the same territory, because they were competing for the same finite resources.
>All of this lowers my already low estimate that we can turn the ship of world culture before it hits an iceberg.
I agree, but I don't think that the first most likely iceberg looks like "decline and be replaced in a few centuries", some sort of sharp crisis will definitely come sooner, and what the culture will look after that is anyone's guess.
It seems to me like AI will cause such a huge shift in society (doesn’t matter if good or bad for this argument) that it makes thinking about the future very difficult. For example I doubt that « number of humans in a culture » will matter as much as it does now. Even problems like « low fertility » might be solved or become unimportant. How does the advent of ASI alter your model?
By your theories, if we get to EMs, again what will matter is who gets the EMs, not who has the bigger population.
I've put a lot of work into understanding what a world of AIs would look like, and think that is in fact possible. Everything I say is in the context of my best guesses about that.
While I agree with your overall point, no our society is not going to be replaced by Amish or Chareidim. Both face a major challenge with apostacy so their headline high fertility rates are very deceptive. I'm not as familiar with the Amish, but Chareidi society is also to some degree eroding due to the limitations on external influences just not being sustainable, it's on the way out.
A much better example would be Israel as a whole which has a remarkably high overall fertility rate in a far more open society
I posted a Note about an issue that applies to your message. Please help spread the word. Here’s my Note:
For those of us who reference the so-called elite in our writings, let us say “financial elite,” not “elite.” Why? A quick example. Nikola Tesla was an elite; Elon Musk is not an elite. Not even close. Elon Musk is a “financial elite.” Big difference. The “elite” are genius, the “financial elite” are wealthy. Let’s not automatically associate excessive wealth with superior intelligence, and let’s not give credit where none is due. (Elon also had a rich dad helping him out, Nikola did not.) Please help spread the word on this point, the difference between the elite and the financial elite.
I see two long term (decadal scale) processes that could make a difference:
1. Educational choice: With legislation already passed, more than half of all students in the US have access to Educational Scholarship Accounts (ESAs). This will result in considerably growth of both religious schools and homeschooling. Even the secular schooling options will be much more likely to reflect values such as family, character, hard work, etc. (e.g. normal American cultural values). I see this as an end run around the elite at scale. The 1.0 version, for the next decade, will not be so different from regular education, but the 2.0 and 3.0 versions in coming decades will increasingly reflect the values that most people outside of academic care about. Implicit in this assessment is that most people care about children and grandchildren, and the bizarre ideological priorities integrated with secular hedonism that currently dominate our elites will naturally be marginalized as we escape the influence of the whacko types that currently dominate universities (and therefore teacher training and state schooling curricula). End the coercive rule of university elites and then the coming generations gradually become more normal.
2. Jurisdictional choice: As new jurisdictions become more common (Prospera is very much a beta test of a future 1.0 version, far from what we'll see a decade from now), many of them will reject the state control over everything that has led to so many declines due to subsidizing harmful behaviors. Herbert Spencer's approach to cultural norms that are likely to survive will make a come back.
What time scales might these market forces make an impact? Depending on what one means by impact, it might be 40, 50, 60, 70 years or so. But on these longer term scales, I predict that we will have more, better, and diverse cultural options beyond the Amish and Orthodox Jews.
There was debate starting around the 1970s in evolutionary biology about at what level selection acted - at what "level" it was useful to talk about adaptation. The candidates were roughly, the gene, the individual, groups, and species. There was fierce criticism of the idea that selection worked at the level of groups and species. Instead, critics argued that "genes" and "individuals" were more suitable candidates for the X in "selection benefits entity X" - and that selection at those levels would frequently override selection at higher levels. The latter sometimes being ridiculed as "naive" group selection. The idea is sometimes characterized as "a herd of fleet deer" - rather than "a fleet herd of deer".
It seems generally plausible that similar dynamics apply to cultural evolution - with memes and memeplexes being the primary focus of adaptation and larger entities - such as, say Germany or France being largely the result of selection acting at lower levels - as opposed to high-level selective forces - such as historical wars of conquest between European countries.
We are all driven, in order, by religion, philosophy, culture and the law (I paraphrase L Strauss). Western Culture (us) has drifted from its Christian roots (I throw Aristotle in) of enlightened altruism. My question is, "where do we point our lifeboats?" Confucian, Hindu and Islamic societies don't look very welcoming.
Yes, elites can start with understanding that Effective Altruism is, evolutionarily speaking, the least effective altruism whereby the potential future benefit to themselves and their extended kin is minimised while the cost to themselves is maximised. Your own book TEiTB promoted EA once upon a time though I certainly couldnt hold it against you, we are a very different world now.
they will lead the adaptation or vanish. Or at least get on board.
We’re coming out of the decline here 🇺🇸 now , it’s the young, The Don is just setting up the country for success by clearing away the debris human and bureaucratic.
See Iron Mountain.
The Elites names aren’t yet known, but I bet one of them is nick named “Big Balls.”
This means America was and is and will be worth fighting for…
I don't know enough bout Amish, but Haredi culture is somewhat centralized. It would require first identifying what values are most critical, and what you want to see them adopt. Then it would be about influencing their elites (which is probably much easier than targeting the global world's elites)
> Even if we fail this time, what we learn may help the next great cycle of civilization to deal better when it’s their turn.
Will there even be useful records? The shift from physical to digital publications seems to give a lower shelf-life to writing, particularly if our civilization declines.
I recall a geologist writing that over the very long term, only something optical will be viable, because old formats will be forgotten. Miniaturizing writing will at least allow future microscopes to see it.
Yes optical storage can last long. But you'd also want to bury readers of those drives if you are to help people before civs revive to levels where they can make their own readers.
"five good LLMs agree that this habitual elite deference to price estimates usually disappears when topics turn to key norms and values"
I've seen this phrase in a lot of your posts lately (X good LLMs agree that...), and it strikes me as bad/tainted/dirty epistemology. Even the "good" llms are sycophantic and can hallucinate, and a lot of people are going off the deep end talking to them.
If I said five relevant human experts, those could still easily be flawed. There are no unflawed sources.
There are no unflawed sources but LLMs are text generators. I use them all day and am pro AI, but citing them like that is exactly like citing the top 5 google search results. Maybe worthy of consideration, maybe a good barometer for public opinion, but not more than that.
It’s more like “five human experts, but they all pay their rent by sucking your dick”.
There are no unflawed sources, but as far as I know, a lot less people are going off the deep end talking to human experts than to LLMs
IMO the standard practice should be to ask the SOTA LLM to generate and then separately rate questions for neutrality. Also of course avoid memory/standard web interface that leaks info about the user (try asking ChatGPT what coffee shop you are in!)
Hallucinations are way down at the SOTA but sycophancy and user bias is way more subtle and difficult to root out.
Social Darwinism is usually understood as “let the weak die so the strong can rise.” If you mean something different by it, you should speak up.
The “let the weak die” approach historically has come against more humanist cultures more than once in direct cultural confrontation and been left behind. (See the Axial
Age adaptations in Greece, India, Israel, China, and Persia; Christianity vs Roman Paganism, Christianity vs Vikings, Nazis vs The West, even early 1900s Progressives vs everyone else.) Social Darwinism is not adaptive.
There was a link next to that phrase which elaborates on its relevant meaning here.
Social Darwinism doesn’t require the weak to die. It’s enough for them to have sub-replacement fertility, which they already do. The challenge is to encourage the strong into higher fertility.
That’s marginally better. But in the fray of cultural evolution, it’s been the societies that care MORE for the weak that have won the day. Encouraging social Darwinism seems to be empirically pushing in an anti-fitness anti-adaptive cultural direction.
There’s an important distinction to be made. In-group altruism increases a group’s fitness because it reduces the risk of wipeouts—there’s solid math and evolutionary modeling showing how risk-sharing within a group improves survival odds.
But inter-group competition is a very different dynamic. Evolutionarily, it’s closer to a zero-sum game: in pre-industrial societies, groups and cultures couldn’t simply coexist in the same territory, because they were competing for the same finite resources.
Weak have super-replacement fertility. They also have an increasing obesity problem. No way this is social darwinist.
>All of this lowers my already low estimate that we can turn the ship of world culture before it hits an iceberg.
I agree, but I don't think that the first most likely iceberg looks like "decline and be replaced in a few centuries", some sort of sharp crisis will definitely come sooner, and what the culture will look after that is anyone's guess.
Perhaps religious revival is necessary
It seems to me like AI will cause such a huge shift in society (doesn’t matter if good or bad for this argument) that it makes thinking about the future very difficult. For example I doubt that « number of humans in a culture » will matter as much as it does now. Even problems like « low fertility » might be solved or become unimportant. How does the advent of ASI alter your model?
By your theories, if we get to EMs, again what will matter is who gets the EMs, not who has the bigger population.
I've put a lot of work into understanding what a world of AIs would look like, and think that is in fact possible. Everything I say is in the context of my best guesses about that.
While I agree with your overall point, no our society is not going to be replaced by Amish or Chareidim. Both face a major challenge with apostacy so their headline high fertility rates are very deceptive. I'm not as familiar with the Amish, but Chareidi society is also to some degree eroding due to the limitations on external influences just not being sustainable, it's on the way out.
A much better example would be Israel as a whole which has a remarkably high overall fertility rate in a far more open society
Since you're not as familiar with the Amish, you should read https://westhunt.wordpress.com/2012/12/23/boiling-off/
I posted a Note about an issue that applies to your message. Please help spread the word. Here’s my Note:
For those of us who reference the so-called elite in our writings, let us say “financial elite,” not “elite.” Why? A quick example. Nikola Tesla was an elite; Elon Musk is not an elite. Not even close. Elon Musk is a “financial elite.” Big difference. The “elite” are genius, the “financial elite” are wealthy. Let’s not automatically associate excessive wealth with superior intelligence, and let’s not give credit where none is due. (Elon also had a rich dad helping him out, Nikola did not.) Please help spread the word on this point, the difference between the elite and the financial elite.
I see two long term (decadal scale) processes that could make a difference:
1. Educational choice: With legislation already passed, more than half of all students in the US have access to Educational Scholarship Accounts (ESAs). This will result in considerably growth of both religious schools and homeschooling. Even the secular schooling options will be much more likely to reflect values such as family, character, hard work, etc. (e.g. normal American cultural values). I see this as an end run around the elite at scale. The 1.0 version, for the next decade, will not be so different from regular education, but the 2.0 and 3.0 versions in coming decades will increasingly reflect the values that most people outside of academic care about. Implicit in this assessment is that most people care about children and grandchildren, and the bizarre ideological priorities integrated with secular hedonism that currently dominate our elites will naturally be marginalized as we escape the influence of the whacko types that currently dominate universities (and therefore teacher training and state schooling curricula). End the coercive rule of university elites and then the coming generations gradually become more normal.
2. Jurisdictional choice: As new jurisdictions become more common (Prospera is very much a beta test of a future 1.0 version, far from what we'll see a decade from now), many of them will reject the state control over everything that has led to so many declines due to subsidizing harmful behaviors. Herbert Spencer's approach to cultural norms that are likely to survive will make a come back.
What time scales might these market forces make an impact? Depending on what one means by impact, it might be 40, 50, 60, 70 years or so. But on these longer term scales, I predict that we will have more, better, and diverse cultural options beyond the Amish and Orthodox Jews.
There was debate starting around the 1970s in evolutionary biology about at what level selection acted - at what "level" it was useful to talk about adaptation. The candidates were roughly, the gene, the individual, groups, and species. There was fierce criticism of the idea that selection worked at the level of groups and species. Instead, critics argued that "genes" and "individuals" were more suitable candidates for the X in "selection benefits entity X" - and that selection at those levels would frequently override selection at higher levels. The latter sometimes being ridiculed as "naive" group selection. The idea is sometimes characterized as "a herd of fleet deer" - rather than "a fleet herd of deer".
It seems generally plausible that similar dynamics apply to cultural evolution - with memes and memeplexes being the primary focus of adaptation and larger entities - such as, say Germany or France being largely the result of selection acting at lower levels - as opposed to high-level selective forces - such as historical wars of conquest between European countries.
We are all driven, in order, by religion, philosophy, culture and the law (I paraphrase L Strauss). Western Culture (us) has drifted from its Christian roots (I throw Aristotle in) of enlightened altruism. My question is, "where do we point our lifeboats?" Confucian, Hindu and Islamic societies don't look very welcoming.
Yes, elites can start with understanding that Effective Altruism is, evolutionarily speaking, the least effective altruism whereby the potential future benefit to themselves and their extended kin is minimised while the cost to themselves is maximised. Your own book TEiTB promoted EA once upon a time though I certainly couldnt hold it against you, we are a very different world now.
Value or 💀
they will lead the adaptation or vanish. Or at least get on board.
We’re coming out of the decline here 🇺🇸 now , it’s the young, The Don is just setting up the country for success by clearing away the debris human and bureaucratic.
See Iron Mountain.
The Elites names aren’t yet known, but I bet one of them is nick named “Big Balls.”
This means America was and is and will be worth fighting for…
… I think we will.
"We are on a ship heading for an iceberg; our main options are to turn the ship, or get off onto lifeboats. I’d rather turn our ship."
Why don't we just bomb the fu*k out of the Iceberg?
If only we could.
Why not inject haredi and amish with the values you want? Seems easiest
How?
I don't know enough bout Amish, but Haredi culture is somewhat centralized. It would require first identifying what values are most critical, and what you want to see them adopt. Then it would be about influencing their elites (which is probably much easier than targeting the global world's elites)
> Even if we fail this time, what we learn may help the next great cycle of civilization to deal better when it’s their turn.
Will there even be useful records? The shift from physical to digital publications seems to give a lower shelf-life to writing, particularly if our civilization declines.
The key question is what sort of computer tech will last long to read long lasting USB drives etc. Seems a noble cause to create and bury such things.
I recall a geologist writing that over the very long term, only something optical will be viable, because old formats will be forgotten. Miniaturizing writing will at least allow future microscopes to see it.
Yes optical storage can last long. But you'd also want to bury readers of those drives if you are to help people before civs revive to levels where they can make their own readers.