Astral Codex Ten Podcast

Ribbonfarm likes to talk about refactoring, a conceptual change in how you see the world. I'm not totally sure I understand it, but I think it means things like memetics – where you go from the usual model of people deciding what ideas they want, to a weird and inside-out (but not objectively wrong) model of ideas competing to colonize people. Here is a refactoring I think about a lot: imagine a world where people considered culture the fourth branch of government. Imagine that civics textbook writers taught high school students that the US government had four branches: executive, legislative, judicial, and cultural. I think about this because I have a bias to ignore anything that isn't nailed down and explicit. Culture isn't nailed down. But if it were in the Constitution in nice calligraphy right beside the Presidency and the Supreme Court, why, then it would be as explicit as it gets. Like many other people, I was hopeful that nation-building Iraq (or Afghanistan, or…) would quickly turn it into a liberal democracy (in my defense, I was eighteen at the time). Like many other people, I was disappointed and confused when it didn't. The people in the world that considers culture the fourth branch of government weren't confused. Bush forgot to nation-build an entire branch of government. If he'd given Iraq a western-style Supreme Court, marble facade and all, but left their executive and legislature exactly how they were before, that would be a recipe for conflict, confusion, and eventually nothing getting done. So why should westernizing their executive, legislature, and courts – but not their culture – work any better? The world that considers culture the fourth branch of government doesn't get all confused calling hunter-gatherers or peasant villagers "primitive communism" or "anarchism" or "ruled by elders" or things like that. Those people's governments have a cultural branch but not much else. Why should we be surprised? Medieval Iceland had onlylegislative and judicial branches; medieval Somalia only had a judiciary; some dictatorships run off just an executive. Each branch of government enforces rules in its own way. The legislature passes laws. The executive makes executive orders. The judiciary rules on cases. And the culture sets norms. In our hypothetical world, true libertarians are people who want less of all of these. There are people who want less of the first three branches but want to keep strong cultural norms about what is or isn't acceptable – think Lew Rockwell and other paleoconservatives who hope that the retreat of central government will create strong church-based communities of virtuous citizens. These people aren't considered libertarians. They might be considered principled constitutionalists, the same way as people who worry about the "imperial presidency" and its use of executive orders. But in the end, what they want to strengthen some branches of government at the expense of others. The real libertarians also believe that cultural norms enforced by shame and ostracism are impositions on freedom, and fight to make these as circumscribed as possible.

A recent discussion: somebody asked why people in Silicon Valley thought that only high-tech solutions to climate change (like carbon capture or geoengineering) mattered, and why they dismissed more typical solutions like international cooperation and political activism. Another person cited statements from the relevant Silicon Valley people, who mostly say that they think political solutions and environmental activism were central to the fight against climate change, but that we should look into high-tech solutions too. This is a pattern I see again and again. Popular consensus believes 100% X, and absolutely 0% Y. A few iconoclasts say that X is definitely right and important, but maybe we should also think about Y sometimes. The popular consensus reacts "How can you think that it's 100% Y, and that X is completely irrelevant? That's so extremist!" Some common forms of this: Reversed moderation of planning, like in the geoengineering example. One group wants to solve the problem 100% through political solutions, another group wants 90% political and 10% technological, and the first group thinks the second only cares about technological solutions. Reversed moderation of importance. For example, a lot of psychologists talk as if all human behavior is learned. Then when geneticists point to experiments showing behavior is about 50% genetic, they get accused of saying that "only genes matter" and lectured on how the world is more complex and subtle than that. Reversed moderation of interest. For example, if a vegetarian shows any concern about animal rights, they might get told they're "obsessed with animals" or they "care about animals more than humans". Reversed moderation of certainty. See for example my previous article Two Kinds Of Caution. Some researcher points out a possibility that superintelligent AI might be dangerous, and suggests looking into this possibility. Then people say it doesn't matter, and we don't have to worry about it, and criticize the researcher for believing he can "predict the future" or thinking "we can see decades ahead". But "here is a possibility we need to investigate" is a much less certain claim than "no, that possibility definitely will not happen". I can see why this pattern is tempting. If somebody said the US should allocate 50% of its defense budget to the usual global threats, and 50% to the threat of reptilian space invaders, then even though the plan contains the number "50-50" it would not be a "moderate" proposal. You would think of it as "that crazy plan about fighting space reptiles", and you would be right to do so. But in this case the proper counterargument is to say "there is no reason to spend any money fighting space reptiles", not "it's so immoderate to spend literally 100% of our budget breeding space mongooses". "Moderate" is not the same as "50-50" is not the same as "good". Just say "Even though this program leaves some money for normal defense purposes, it's stupid". You don't have to deny that it leaves anything at all.

Donald Trump has been called a setback for many things. America. The global community. The environment. Civil service. Civil society. Civility. Civilization. The list goes on. One might think he has at least been useful to his own cause. That he could at least claim to have benefited the ideas of populism, nationalism, immigration control, and protectionism. That if anything could avoid being devastated by Trump, it would be Trumpism. But here are some polls from the past few years. They're all on slightly different things, but I think together they tell an interesting story: Support for global free trade mysteriously spiked around 2016. So did moral support for immigrants. …and, less clearly but still there, support for increasing the number of immigrants (though see here for an apparently contrary source). …and opposition to deporting illegal immigrants. So did belief in racial discrimination as a major cause of inequality, according to this chart with a completely unbiased title which is willing to let readers decide how to think about this issue for themselves. And so did trust in the New York Times and other mainstream media sources. The clearest example I can find of this effect doesn't come from the US at all. It's Minkus, Deutschmann & Delhey (2018). They find that a large European poll asked the same question about support for the EU the week before and after Trump's election. Just after the election, there was a giant spike in support for the EU, "considerable in size, roughly equivalent to three years of education". They conclude that: The election of Trump as a right-wing nationalist with a declared aversion to supranational institutions including the EU — did not trigger a domino effect in the same direction in Europe. To the contrary, a rally effect occurred, in which Europe moved closer together, rallying around the EU's "flag." This indicates that an event that may at first sight appear to be a global victory for nationalism can immediately trigger measurable sentiments of resistance in another part of the world, actually leading to new impetus for supranationalism.

One interesting thing I took from Evolutionary Psychopathology was a better understanding of the diametrical theory of the social brain. There's been a lot of discussion over whether schizophrenia is somehow the "opposite" of autism. Many of the genes that increase risk of autism decrease risk of schizophrenia, and vice versa. Autists have a smaller-than-normal corpus callosum; schizophrenics have a larger-than-normal one. Schizophrenics smoke so often that some researchers believe they have some kind of nicotine deficiency; autists have unusually low smoking rates. Schizophrenics are more susceptible to the rubber hand illusion and have weaker self-other boundaries in general; autists seem less susceptible and have stronger self-other boundaries. Autists can be pathologically rational but tend to be uncreative; schizophrenics can be pathologically creative but tend to be irrational. The list goes on. I've previously been skeptical of this kind of thinking because there are many things that autists and schizophrenics have in common, many autistics who seem a bit schizophrenic, many schizophrenics who seem a bit autistic, and many risk factors shared by both conditions. But Del Giudice, building on work by Badcock and Crespipresents the "diametrical model": schizophrenia and autism are the failure modes of opposing sides of a spectrum from high functioning schizotypy to high functioning autism, ie from overly mentalistic cognition to overly mechanistic cognition. Schizotypy is a combination of traits that psychologists have discovered often go together. It's classified as a personality disorder in the DSM. But don't get too caught up on that term – it's a disorder in the same sense as narcissistic or antisocial tendencies, and like those conditions, some schizotypals do very well for themselves. Classic schizotypal traits include tendency toward superstition, disorganized communication, and nonconformity (if it sounds kind of like "schizophrenia lite", that's not really a coincidence).

[Content note: eating disorders] Anorexia has a cultural component. I'm usually reluctant to assume anything is cultural – every mediocre social scientist's first instinct is always to come up with a cultural explanation which is simple, seductive, flattering to all our existing prejudices, and wrong. But after seeing enough ballerinas and cheerleaders who became anorexic after pressure to lose weight for the big competition, even I have to throw up my hands and admit anorexia has a cultural component. But nobody ever tells you the sequel. That ballerina who's losing weight for the big competition at age 16? At age 26, she's long since quit ballet, worried it would exacerbate her anorexia. She's been in therapy for ten years; for eight of them she's admitted she has a problem, that her anorexia is destroying her life. Her romantic partners – the ones she was trying to get thin to impress – have long since left her because she looks skeletal and weird. She understands this and would do anything to cure her anorexia and be a normal weight again. But she finds she isn't hungry. She hasn't eaten in two days and she isn't hungry. In fact, the thought of food sickens her. She goes to increasingly expert therapists and dieticians, asking them to help her eat more. They recommend all the usual indulgences: ice cream, french fries, cookies. She tries all of them and finds them inexplicably disgusting. Sometimes with a prodigious effort of will she will manage to finish one cookie, and congratulate herself, but the next day she finds the task of eating dessert as daunting as ever. Finally, after many years of hard work, she is scraping the bottom end of normal weight by keeping to a diet so regimented it would make a Prussian general blush. And nobody ever tells you about all the people who weren't ballerinas. The young man who stops eating because it gives him a thrill of virtue and superiority to be able to demonstrate such willpower. The young woman who stops eating in order to show her family how much their neglect hurts her. If they pursue their lack of appetite far enough, they end up the same way as the ballerina – admitting they have a problem, admitting they need to eat more, hiring all sorts of doctors and dieticians to find them a way to eat more, but discovering themselves incapable of doing so.

I. Evolutionary psychology is famous for having lots of stories that make sense but are hard to test. Psychiatry is famous for having mountains of experimental data but no idea what's going on. Maybe if you added them together, they might make one healthy scientific field? Enter Evolutionary Psychopathology: A Unified Approach by psychology professor Marco del Giudice. It starts by presenting the theory of "life history strategies". Then it uses the theory – along with a toolbox of evolutionary and genetic ideas – to shed new light on psychiatric conditions. Some organisms have lots of low-effort offspring. Others have a few high-effort offspring. This was the basis of the old r/k selection theory. Although the details of that theory have come under challenge, the basic insight remains. A fish will lay 10,000 eggs, then go off and do something else. 9,990 will get eaten by sharks, but that still leaves enough for there to be plenty of fish in the sea. But an elephant will spend two years pregnant, three years nursing, and ten years doing at least some level of parenting, all to produce a single big, well-socialized, and high-prospect-of-life-success calf. These are two different ways of doing reproduction. In keeping with the usual evolutionary practice, del Giudice calls the fish strategy "fast" and the elephant strategy "slow". To oversimplify: fast strategies (think "live fast, die young") are well-adapted for unpredictable dangerous environments. Each organism has a pretty good chance of randomly dying in some unavoidable way before adulthood; the species survives by sheer numbers. Fast organisms should grow up as quickly as possible in order to maximize the chance of reaching reproductive age before they unpredictably die. They should mate with anybody around, to maximize the chance of mating before they unpredictably die. They should ignore their offspring, since they expect most offspring to unpredictably die, and since they have too many to take care of anyway. They should be willing to take risks, since the downside (death without reproducing) is already their default expectation, and the upside (becoming one of the few individuals to give birth to the 10,000 offspring of the next generation) is high. Slow strategies are well-adapted for safer environments, or predictable complex environments whose intricacies can be mastered with enough time and effort. Slow strategy animals may take a long time to grow up, since they need to achieve mastery before leaving their parents. They might be very picky maters, since they have all the time in the world to choose, will only have a few children each, and need to make sure each of those children has the best genes possible. They should work hard to raise their offspring, since each individual child represents a substantial part of the prospects of their genetic line. They should avoid risks, since the downside (death without reproducing) would be catastrophically worse than default, and the upside (giving birth to a few offspring of the next generation) is what they should expect anyway.

I. The Mind Illuminated is a guide to Buddhist meditation by Culadasa, aka John Yates, a Buddhist meditation teacher who is also a neuroscience PhD. At this point I would be more impressed to meet a Buddhist meditation teacher who wasn't a neuroscience PhD. If I ever teach Buddhist meditation, this is going to be my hook. "Come learn advanced meditation techniques with Scott Alexander, whose lack of a neuroscience PhD gives him a unique perspective that combines ancient wisdom with a lack of modern brain science." I think the world is ready for someone to step into this role. But Culadasa is not that person, and The Mind Illuminated is not that book. I am trying not to read too many books on spiritual practices until I'm ready to practice some spirituality. I made an exception for TMI because lots of people recommended it to me for its description of how the brain works. This seems like the sort of thing that Buddhist meditation teachers who are also neuroscientists could have insight on, so I decided to check it out. Tradition divides meditation into two parts: concentration meditation, where you sharpen and control your focus, versus insight meditation, where you investigate the nature of perception and reality. TMI follows a long tradition of focusing on concentration meditation, with the assumption that insight meditation will become safer and easier once you've mastered concentration, and maybe partly take care of itself. Its course divides concentration meditation into ten stages. Early stages contain basic tasks like setting up a practice, focusing on the breath, and overcoming distractability. Later stages are more interesting; the ninth stage is learning how to calm the intensity of your meditative joy; apparently without special techniques "overly intense joy" becomes a big problem. I usually hate meditation manuals, because they sound like word salad. "One attains joy by combining pleasure with happiness. Pleasure is a state of bliss which occurs when one concentrates focus on the understanding of awareness. Happiness is a state of joy that occurs when one focuses concentration on the awareness of understanding. By focusing awareness on bliss, you can increase the pleasure of understanding, which in turn causes concentration to be pleasant and joy to be blissful, and helps you concentrate on understanding your awareness of happiness about the bliss of focus." At some point you start thinking "Wait, were all the nouns in that paragraph synonyms for each other?"

[This post was up a few weeks ago before getting taken down for complicated reasons. They have been sorted out and I'm trying again.] Is scientific progress slowing down? I recently got a chance to attend a conference on this topic, centered around a paper by Bloom, Jones, Reenen & Webb (2018). BJRW identify areas where technological progress is easy to measure – for example, the number of transistors on a chip. They measure the rate of progress over the past century or so, and the number of researchers in the field over the same period. For example, here's the transistor data: This is the standard presentation of Moore's Law – the number of transistors you can fit on a chip doubles about every two years (eg grows by 35% per year). This is usually presented as an amazing example of modern science getting things right, and no wonder – it means you can go from a few thousand transistors per chip in 1971 to many million today, with the corresponding increase in computing power. But BJRW have a pessimistic take. There are eighteen times more people involved in transistor-related research today than in 1971. So if in 1971 it took 1000 scientists to increase transistor density 35% per year, today it takes 18,000 scientists to do the same task. So apparently the average transistor scientist is eighteen times less productive today than fifty years ago. That should be surprising and scary. But isn't it unfair to compare percent increase in transistors with absolute increase in transistor scientists? That is, a graph comparing absolute number of transistors per chip vs. absolute number of transistor scientists would show two similar exponential trends. Or a graph comparing percent change in transistors per year vs. percent change in number of transistor scientists per year would show two similar linear trends. Either way, there would be no problem and productivity would appear constant since 1971. Isn't that a better way to do things? A lot of people asked paper author Michael Webb this at the conference, and his answer was no. He thinks that intuitively, each "discovery" should decrease transistor size by a certain amount. For example, if you discover a new material that allows transistors to be 5% smaller along one dimension, then you can fit 5% more transistors on your chip whether there were a hundred there before or a million. Since the relevant factor is discoveries per researcher, and each discovery is represented as a percent change in transistor size, it makes sense to compare percent change in transistor size with absolute number of researchers. Anyway, most other measurable fields show the same pattern of constant progress in the face of exponentially increasing number of researchers. Here's BJRW's data on crop yield: The solid and dashed lines are two different measures of crop-related research. Even though the crop-related research increases by a factor of 6-24x (depending on how it's measured), crop yields grow at a relatively constant 1% rate for soybeans, and apparently declining 3%ish percent rate for corn. BJRW go on to prove the same is true for whatever other scientific fields they care to measure. Measuring scientific progress is inherently difficult, but their finding of constant or log-constant progress in most areas accords with Nintil's overview of the same topic, which gives us graphs like …and dozens more like it. And even when we use data that are easy to measure and hard to fake, like number of chemical elements discovered, we get the same linearity:

[Content warning: scrupulosity] I. "There is no ethical consumption under late capitalism". I hear this from a bunch of people. Sometimes it is taken to its conclusion; no currently living person is morally acceptable. People who aren't activists reorienting their entire lives around acknowledging and combating the evils of the world aren't even on the scale. And people who are such activists are (in the words of one of my friends who is close to that community) "only making comfortable sacrifices that let them think of themselves as a good person within their existing comfortable moral paradigm, instead of confronting the raw terrible truth." IE "If you think you're one of the good ones, you're wrong". I have heard this sentiment raised by animal rights activists. The average meat-eater isn't even on the scale. The average vegetarian still eats milk and cheese, and so is barely even trying. Even most vegans probably use some medical product with gelatin, or something tested on lab rats, or are just benefitting from animal suffering in some indirect way. And I have heard it raised by environmentalists. The average SUV driver isn't even on the scale. The average conscientious liberal might think they're better because they bike to work and recycle, but they still barely think about how they're using electricity generated by coal plants and eating food grown with toxic pesticides. Everyone could be doing more. And I have heard raised by labor activists. Most of us use stuff made in sweatshops. Even if you avoid sweatshops, you probably use stuff made at less than a living wage. Even if you avoid that, are you doing everything you can to help and support workers who earn less than you do? Even if you aren't an animal rights activist, environmentalist, or labor advocate, do you believe in anything? Are you a Christian, a social justice advocate, or rationalist? Do you know anyone who really satisfies you as being sinless, non-racist, and/or rational? Then perhaps you too believe nobody is good.

I. A lot of people pushed back against my post on preschool, so it looks like we need to discuss this in more depth. A quick refresher: good randomized controlled trials have shown that preschools do not improve test scores in a lasting way. Sometimes test scores go up a little bit, but these effects disappear after a year or two of regular schooling. However, early RCTs of intensive "wrap-around" preschools like the Perry Preschool Program and the Abecedarians found that graduates of those programs went on to have markedly better adult outcomes, including higher school graduation rates, more college attendance, less crime, and better jobs. But these studies were done in the 60s, before people invented being responsible, and had kind of haphazard randomization and followup. They were also small sample sizes, and from programs that were more intense than any of the scaled-up versions that replaced them. Modern scaled-up preschools like Head Start would love to be able to claim their mantle and boast similar results. But the only good RCT of Head Start, the HSIS study, is still in its first few years. It's confirmed that Head Start test score gains fade out. But it hasn't been long enough to study whether there are later effects on life outcomes. We can expect those results in ten years or so. For now, all we have is speculation based on a few quasi-experiments. Deming 2009 is my favorite of these. He looks at the National Longitudinal Survey of Youth, a big nationwide survey that gets used for a lot of social science research, and picks out children who went to Head Start. These children are mostly disadvantaged because Head Start is aimed at the poor, so it would be unfair to compare them to the average child. He's also too smart to just "control for income", because he knows that's not good enough. Instead, he finds children who went to Head Start but who have siblings who didn't, and uses the sibling as a matched control for the Head Starter. This ensures the controls will come from the same socioeconomic stratum, but he acknowledges it raises problems of its own. Why would a parent send one child to Head Start but not another? It might be that one child is very stupid and so the parents think they need the extra help preschool can provide; if this were true, it would mean Head Starters are systematically dumber than controls, and would underestimate the effect of Head Start. Or it might be that one child is very smart and the so the parents want to give them education so they can develop their full potential; if this were true, it would mean Head Starters are systematically smarter than controls, and would inflate the effect of Head Start. Or it might be that parents love one of their children more and put more effort into supporting them; if this meant these children got other advantages, it would again inflate the effect of Head Start. Or it might mean that parents send the child they love more to a fancy private preschool, and the child they love less gets stuck in Head Start, ie the government program for the disadvantaged. Or it might be that parents start out poor, send their child to Head Start, and then get richer and send their next child to a fancy private preschool, while that child also benefits from their new wealth in other ways. There are a lot of possible problems here.