"The riddle of the Amish."
"The Amish have much to teach us. It may seem strange, even surreal, to turn to one of America's most traditional groups for lessons about living in a hyper-tech world -- especially a horse-driving people who have resisted 'progress' by snubbing cars, public grid power, and even high school education. Nonetheless, the Amish deserve a hearing."
"The key to unlock the riddle, in Donald Kraybill's view, is to realize that the Amish are negotiating with modernity in three ways: they accept, they reject, they bargain. They might reject 'radios, televisions, high school, church buildings, and salaried ministers,' Kraybill notes, but accept 'small electronic calculators and artificial insemination of cows. And more recently...LED lights on buggies, and battery-powered hand tools'. Then, there also is bargaining. One way they bargain with modernity is by 'Amishizing' certain technologies or techniques, for example 'neutering' computers (stripping WiFi, video games, internet access, etc.), or creating a school curriculum that reflects Amish values and traditions. Another form of bargaining is found in the Amish distinction between access and ownership of certain technologies. An Amish person may pay someone for a car-ride to work, but may not own a car."
"The Amish, arguably more than any other group in America, have tried to domesticate technology so that its potent force does not overwhelm or cripple their culture."
"The Amish aren't anti-technology; they are pro-community."
Towards the end of the article, it mentions fertility rates.
"Amish population continues to grow (a 116% increase since the year 2000) and communities are spreading into new states (including Colorado, Nebraska, New Mexico, South Dakota, Vermont, and Wyoming). Since the Amish don't tend to proselytize, this growth is in large part organic and natural, through new births and an extremely high retention rate that has increased to around 90% at a time when other Christian denominations are shedding members at record numbers."
To get a sense of what that means, I looked up the fertility rate of the US as a whole and found 1.84 children per woman. I tried looking it up for the Amish, but most people just said something like 6-8 or 6-9. I found two websites that said 7, so I decided to just plug in 7.0 and run with it.
If we start with a society that's 50% regular Americans and 50% Amish, after 1 generation it'll be 29% regular Americans and 71% Amish, after 2 generations it'l be 19% regular Americans, 81% Amish, and ...
At this point I decided to plug in the actual population numbers. For the US, I got 307.205 million, and for the Amish... again accurate numbers were difficult to come by but I found someone who said 350,000, so I decided to run with that number.
After 1 generation: 99.6% non-Amish, 0.4% Amish
After 2 generations: 98.8% non-Amish, 1.2% Amish
After 3 generations: 96.0% non-Amish, 4.0% Amish
After 4 generations: 87.8% non-Amish, 12.2% Amish
After 5 generations: 68.8% non-Amish, 31.2% Amish
After 6 generations: 43.0% non-Amish, 57.0% Amish
After 7 generations: 24.9% non-Amish, 75.1% Amish
After 8 generations: 17.2% non-Amish, 82.8% Amish
After 9 generations: 14.7% non-Amish, 85.3% Amish
After 10 generations: 13.9% non-Amish, 86.1% Amish
After 11 generations: 13.7% non-Amish, 86.3% Amish
After 12 generations: 13.6% non-Amish, 86.4% Amish
If we assume a "generation" is about 25 years, then what this means is that in 300 years, the population of the US will be 86.4% Amish.
If you're wondering what happens after generation 12, the simulation stabilizes on 13.6% regular non-Amish Americans and 86.4% Amish. It doesn't go up to 99% Amish because of the 10% defection rate per generation. But you'll note that most non-Amish Americans in that time frame will be Amish defectors, not descendants of people who today are non-Amish.
In reality, it could happen faster than in my simulation because generation times for the non-Amish are actually longer than 25 years, and generation times for the Amish are about 20 years.
Another implication of this model is that the shift to "progressive" or "left" values that we are seeing in the upcoming generations ("gen Z" and "gen alpha") are probably not going to last more than 3 generations (about 75 years). After that, we'll see a shift towards conservative, or more precisely Amish, values.
Another thing that might make the model different from reality is that we don't account for other high-fertility groups, in particular the Muslims. Muslims have a lower fertility rate than the Amish but they have the advantage of in-migration from the outside world, where the global Muslim population also has a high fertility rate.
This brings me to my central thesis in all this: What is the ultimate cause of low fertility rates? Those of you who've been hanging around me for any length of time probably have some idea what my answer is. My answer is: It's technology, or more specifically, when technology maximally complements humans, fertility rates are maximized, and when technology maximally competes against humans, fertility rates are minimized.
This underlying driver is, in my view, obscured by many things: people point out how cities have lower fertility rates than rural areas and attribute low fertility rates to "urbanization"; people argue about the role of the feminism movement, birth control, porn, social media, dating apps, and so on; people looking for competition against humans by machines focus on unemployment numbers instead of fertility numbers -- but people don't react to competition from machines by increasing the unemployment numbers, they do it by spending more time in school (which accounts for the declines we've seen in total labor force participation) and delaying marriage and childbirth.
If you look at the long arc of human history, humans spent 95+% of our existence as a species as hunter/gatherers. During this time, the human population barely increased.
We didn't see substantial increases in the human population until the agricultural revolution. The agricultural revolution started more or less immediately after the end of the last ice age, 11,600 years ago, but took a long time to get started -- what we think of as the "agricultural revolution" didn't really get going until around 5,000 years ago. Even then, the rate of human population growth was modest.
Eventually, what really got the human population growing fast was inventions that dramatically increased agricultural production. Heavy plows, like the wheeled moldboard plow, allowing for deeper plowing, turning over more soil, improved scythes, iron tools replacing wooden tools, horse collars, horse harnesses, and horseshoes, watermills and windmills, crop rotation, rotating grains with legumes, etc, and 3-field systems, with spring planting, autumn planting, and fallow, drainage and irrigation techniques, composting techniques, and selective breeding of both plants and animals.
Eventually, the industrial revolution happened. Human, ox, and horse power got replaced with steam power and gasoline. The Haber-Bosch process of industrial-scale nitrogen fixation and fertilizer production was invented. Industrial-scale production of fertilizer with phosphorus, sulfur, electrolytes (sodium, potassium, calcium, magnesium), and micronutrients were invented. Industrial-scale pesticide production began. Genetically engineered crops supplemented selective breeding for high-yielding crops. All these industrial-scale developments taken together brought us the "green revolution", and this is when you really saw the human population explode -- from under 1 billion to more than 8 billion today.
The thing that first clued me in that machines were a life form that competed against humans was a book called "Cosmic Evolution by Eric Chaisson. It showed how a measurement in physics, free energy rate density, correlated with subjective "complexity" or "structure" or "negative entropy". More complex animals have higher free energy rate density than less complex animals, and within human beings, human brain turns out to have the highest free energy rate density of any part of any animal. But the interesting thing is that CPUs surpassed the human brain on this measure around 2005 or so. So by this measure, the smartest "species" on the planet is actually computers, and actually has been since about 2005.
The interesting thing about that is that in 2005, computers were sort of idiot savants. They could do billions of arithmetic calculations per second -- more than any human -- and without any mistakes, too. But they couldn't see or understand language or manipulate objects in the physical world. Now computers are starting to take on these "human" non-idiot-savant abilities -- computers have vision and can generate images, can understand and generate language, and, well, still can't manipulate objects in the physical world. But those abilities are increasing. We don't know when but they'll be on par with humans at some point.
If we imagine a future where computers have completely taken over the job market, does that mean humans are just going to die off? All but the rich, who can survive on their investment incomes? No, there is another option -- subsistence farming. And that's what the Amish are doing. The world a hundred years into the future will consist of machines who run the world economy -- creating goods and services for other machines to purchase, with humans largely out of the market due to lack of labor income, but some will have lots of income from their investments -- and humans who survive outside the labor market through subsistence farming.
The key to becoming the latter group is the ability to resist technology. Worldwide we see the groups with the highest fertility rates are religious fundamentalists. And they don't need to be Christian -- the Heredim, also know as Ultra-Orthodox Jews, have high fertility rates. Within Islam, the most fundamentalist groups, like the Salafis in Egypt, have the highest fertility rates. What I find interesting about the Amish is that they are not fundamentalists. Rather than resisting technology as a side-effect of extreme adherence to a fundamentalist belief system, they resist technology deliberately, as an objective in and of itself. And they prove it can be done. And since they have proven it can be done, it seems reasonable to assume high fertility rates will continue into the future -- or will drop only modestly. In my model I assumed no change in fertility rates, but it seems to be the fertility rates of the non-Amish population is actually likely to continue dropping. Fertility rates for the US population as a whole have dropped faster than anyone expected and there doesn't seem to be any floor. If we look around the world we can see countries like South Korea with even lower fertility rates like South Korea (1.12 children per woman) and they seem to just continue going down.
So, I think the world we have to look forward to in a couple hundred years is: machines are the dominant species and control the world economically and militarily, humans survive within that world as subsistence farmers, mostly Amish and fundamentalist Muslims and fundamentalist sects of other religions.
This is the point where people usually chime in and say, there's going to be universal basic income. The thing is, I have the research paper from the UBI study funded by Sam Altman, and it's 147 pages, and I'm only a few pages in. So I really can't comment on it right now -- that will have to wait until another time.
My feeling for a long time has been that UBI is politically unfeasible, but people have told me, it will become politically feasible once a large enough percentage of the population is affected, and affected severely enough. If that's your view, then you can consider my projection to be what the world of the future will look like in the absence of UBI.
Steel-manning the Amish: The wisdom of communal discernment - Front Porch Republic
