Astral Codex Ten Podcast

Discover the intriguing world of psychiatry through a host's journey in managing a database of mental health information. Explore the delicate balance between drug benefits and potential addiction, prompted by listener feedback about real-life experiences. Delve into the muddled waters of medical misinformation and the consequences of overly cautious advice. Reflect on the roles of expertise and trust in public health, revealing the ethical dilemmas faced by professionals as they navigate a complex landscape of information.

Thanks to everyone who has waited patiently for more information on this. I planned a book review contest for last summer, which I didn't get to do because of my unexpected hiatus. I currently have 31 entries, none of which I've read yet. My plan is to give the rest of you until March 1 to send in reviews. Send them to scott[at]slatestarcodex[dot]com. I originally wanted ones that you hadn't already posted somewhere else first, but if you posted it over the last ~year because you didn't know if I was coming back, I guess that's fair and I'll make an exception this time. Sometime after March 1 I'll read all of them, publish the top ~5 as posts here, and let people vote on the winners using some method that combines and weighs their votes and mine. First place will get at least $1000, second place $500, third place $250 - I might increase these numbers later on. Some winners may also get an invitation to pitch me any other pieces they have that they think would make good SSC posts. I may also release non-finalist entries somewhere else so people can read them – if you strongly object to me making your entry public, let me know. I previously said to send me entries in .txt format with hand-written HTML. If you've already done that, you're fine and don't need to change. If you're submitting a new entry, you can either do that or send me a normal Word/Google document type thing - I think Substack's interface should be able to handle it.

[Previously in sequence: Taxometrics] I. Imagine Alice has a chronic disease. Luckily, as long as she has a job, she will have health insurance. And health insurance provides her with a treatment. Every day she takes the treatment, her health will go up one point on a 0-100 scale; every day she misses the treatment, it will go down one point. If her health ever gets below 75, she will be too ill to work. Mathematicians would call this a dynamical system with three variables: does she have a job or not, does she have insurance or not, and her health level. We know from the rules above that j always equals i, and that j is 1 as long has h is 75 or higher. And every day i = 1, h goes up one; every day i = 0, h goes down 1. Alice starts with a job and health of 100. Since j = 1 and i = 1, health goes up one each day, but it's maxed out at 100 so it just says there. This state is perfectly stable; as long as the system follows the rules above, it will never change. Suppose Alice gets a mild cold which knocks her health down to 90. She keeps working, her health keeps going up 1 each day, and she eventually gets back to 100. Suppose Alice gets a medium flu which knocks her health down to 80. She keeps working, her health keeps going up 1 each day, and she eventually gets back to 100. But suppose Alice gets a serious pneumonia which knocks her health down to 70. Now she can't work, she loses insurance, her health starts going down 1 each day, and she eventually goes down to 0. This is another stable state; as long as the system keeps evolving according to the rules, it will never change.

I was driving down to LA when the cops pulled me over. "You have to turn back sir, the Sphinx here eats any traveler who can't answer her riddle." "I've trained my whole life for this" I said, and stepped on the gas. Soon I saw a Sphinx lounging in the middle of the road. When she spotted me, she asked: "What has braces, crowns, and retainers, but is not teeth?" "A medieval king in armor. My turn. What has pupils, irises, and whites, but is not an eye?" "A gardening class during apartheid. How is a river like the Federal Reserve?" "It maintains liquidity despite rushes on the banks. What has wings, but cannot fly - fins, but cannot swim - and heels, but cannot walk?" "Helsinki General Hospital." The Sphinx licked her lips. "But tell me, how is Lord Nelson like a cigar?" "They both have one I and won sea. How is Mary Mary Quite Contrary like the Norse god Odin?"

Prediction markets are the future. They're a type of trustless, decentralized expertise that often equals or outperforms official sources. But they're not quite the present. Right now I only know of three prediction markets, and none of them live up to their potential. As usual, it's the government's fault: betting on prediction markets is technically gambling, which makes it mostly illegal (of course, you can still buy all the Gamestop stock you want). Each of the three big prediction markets tries to solve this in a different way. PredictIt struck a special deal with the government where they can run a limited number of predictions as long as nobody is allowed to bet very much. They probably have the most users, but most of them are dumb money, and the restrictions prevent smart money from coming in and replacing them. Also, the market fritters away its limited number of predictions on dumb political horse race stuff. I think of this as a missed opportunity. Augur is a decentralized Ethereum-based prediction market. Since it's crypto, it ought to be able to violate laws with impunity, and in theory this should make it the leader of the pack. In practice it's either nonfunctional or so minimally functional as to be useless. The team involved seems pretty dedicated and competent, and I assume they have some good reason for pretending their product currently exists instead of replacing the whole thing with a COMING SOON banner. I hope this will one day be our savior. But like the real Messiah, it's taking its sweet time. Metaculus solves the regulatory problem by using fake Internet points instead of money. This is a disappointing solution; it limits the user base to Internet obsessives instead of (say) investment bankers. Still, there are a lot of Internet obsessives. And the team running it is really top-notch, interested in pushing the limits of what prediction markets can do, and trying to focus on some of the most important questions. I want to raise awareness of prediction markets, and right now Metaculus seem like the best people to raise awareness of. So welcome to Metaculus Mondays, where I make you listen to reports of how the prediction markets did this week and what they're predicting for later.

Glen Weyl posted a reply to my post criticizing his essay on technocracy, and kindly agreed to let me elevate it into a top-level post. (consider this a standing offer to anyone else I write a post criticizing to do the same) I've very slightly edited some parts to adjust for differences in how the code works. You can read more from Glen Weyl on his website, his Twitter, or by buying his book. I am grateful for your taking the time to respond. There is a lot there to respond to and in general I think the exchange speaks for itself. However, I think there are few points where clarification is important for the exchange to be productive, which I'll briefly address here. 1. Let me start with concessions. There are many points where @slatestarcodex correctly highlights various areas where my grasp of beliefs and facts are limited or wrong, especially in the depth of my grasp of the views of the rationalist community. I freely admit that there are serious limits to how much I've been able to research the views of people in this community and I certainly hope they are not as I characterized them, though as I will point out below many elements of Scott's response confirm my concerns.

https://astralcodexten.substack.com/p/contra-weyl-on-technocracy I. I am not defending technocracy. Nobody ever defends technocracy. It's like "elitism" or "statism". There is no Statist Party. Nobody holds rallies demanding more statism. There is no Citizens for Statism Facebook page with thousands of likes and followers. Yet for some reason libertarians don't win every single national election. Strange, isn't it? Maybe it's one of those Russell conjugations - "I am firm, you are obstinate". I support rule of law, you're a statist. I want checks and balances on mob rule, you're an elitist. I like evidence-based policy, you're a technocrat. I am not defending technocracy. But I do like evidence-based policy. So I read with interest Glen Weyl's Why I Am Not A Technocrat. It starts with a short summary of Seeing Like A State. It ties this into modern "evidence-based policy" and "mechanism design". It talks about how technocrats will always have their own insular culture and biases and paradigms, which prevent them from seeing the real world in its full complexity. Therefore, we should be careful about supposedly "objective" policies, and make sure they are always heavily informed by real people's real knowledge. Then it draws on vague rumors of the "rationalist community" and a shadowy figure named "Eliezer Yudkowsky" to create a completely fictional reimagination of us as a group of benighted people who don't understand any of these things, and just go around saying "hurr durr top-down systems are great, no way there could possibly be anything our models don't capture."

https://astralcodexten.substack.com/p/ontology-of-psychiatric-conditions [reposted from here, with edits] I. Taxometrics is the study of whether psychiatric conditions are categorical or dimensional. Something is categorical if it neatly, objectively separates into different groups. For example, consider humans and rabbits. If we take a mixed group containing some humans and some rabbits, and graph them along some variable like weight, it would probably look like this: There's one big obvious group around 3 lbs (weight of the average rabbit) and another around 140 lbs (weight of the average human). Not a lot of subtletly here. If we used some other graphable variable – height, lifespan, IQ – we'd probably get something similar. Maybe the biggest rabbit in the world is bigger than the smallest human. That doesn't mean they're not two obvious categories. It just means they're two obvious categories with a tiny overlap. It happens. If we wanted to be clever, we could create a multivariate distance measure that combines weight, height, lifespan, IQ, and lots of other ways humans and rabbits could differ, into a 0 – 1 variable where 0 is "most rabbity" and 1 is "most humanish". Probably these scores wouldn't overlap at all – if they did, it would mean there's some human who's more like a rabbit than some rabbit is, which would be pretty surprising. But even if this were true, it wouldn't change the fundamental finding that humans and rabbits are pretty different. Or to put it some other way, there's a fundamental hidden generator producing differences between humans and rabbits (in this case, the species difference). By contrast, something is dimensional if it's just a spectrum and there's no obvious place to separate it into different groups. For example, consider tall people vs. short people. We take a general cross-section of the population, and graph them by height, and it would probably look like this: There's no clear point where short people stop and tall people begin. Some people are a little taller than others, and other people taller still, and so on until you're at Yao Ming.

In the 1950s, a shady outfit called Obetrol Pharmaceuticals made a popular over-the-counter diet pill called Obetrol. If you're familiar with any of: the 1950s, shady pharma, or diet pills, your next question will be "did it contain amphetamines?" and the answer is yes, loads of them. Obetrol was a mix of four different amphetamine salts: racemic amphetamine sulfate, dextroamphetamine sulfate, methamphetamine saccharate, and methamphetamine hydrochloride. Why did they need four different kinds of speed? I'm not sure. The uncharitable explanation is: for the same reason Dr. Nick's Cure-All Home Remedy has twelve different herbs, ie customers think things with more ingredients are better. By the 1970s, people figured out meth was bad, so Obetrol replaced their two methamphetamine salts with two more kinds of non-methylated amphetamine. But the FDA continued to crack down, and although the historical paper trail goes kind of dark, it looks like Obetrol had disappeared by the 1980s.

Logistics Don't worry, there will be real posts next week. Jan 22 Substack First, thanks for following me to Substack. I know some of you are skeptical. I was too at first, but Substack has gone above and beyond in allaying my concerns. They've let me test out a "no popup telling you to subscribe" feature. They've changed the comment section to be more like WordPress. We've agreed I'm here for a year, but if it goes badly I can leave in 2022 with no hard feelings. And I know some of you are concerned about the risk of corporate deplatforming. My weak answer is that so far Substack has been great at resisting calls for this, I think it's worth rewarding them with my business, and I'm proud to contribute to companies that share my values. My strong answer is that if I start feeling too constrained, I'll leave. The past six months weren't fun, but at least I credibly signaled willingness to destroy everything and start over when needed. I'm saving the Substack mailing list regularly to my hard drive, and if I go somewhere else I'll let you all know. And I know Substack is supposed to be an email newsletter thing, but consider reading it online instead so you can participate in the comments. If you miss the old layout, the anonymous author of Applied Divinity Studies has made a Chrome extension thing you can use to convert the Substack layout to the older one.