Astral Codex Ten

Wuhan

153 min read

Wuhan

Article

Wuhan is a recurring place in the Astral Codex Ten archive, appearing 10 times across 10 issues between July 07, 2021 and July 01, 2025. The archive places it in contexts such as “If, the moment COVID had been reported in Wuhan”; “American citizens who had recently visited Wuhan”; “The original Wuhan strain was probably around 2.5”. It most often appears alongside China, US, Taiwan.

Metadata

  • Category: Places
  • Mention count: 10
  • Issue count: 10
  • First seen: July 07, 2021
  • Last seen: July 01, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

Lockdown Effectiveness: Much More Than You Wanted To Know
July 07, 2021 · Original source
If, the moment COVID had been reported in Wuhan, other countries had closed their borders tightly, that would have prevented the pandemic (at least for a while). In that sense, lockdowns definitely could have worked.
Adumbrations Of Aducanumab
August 05, 2021 · Original source
So everyone sat on their defective FDA-approved coronavirus tests, and their excellent high-quality non-FDA approved coronavirus tests that they were banned from using, and didn’t test anyone for coronavirus. Meanwhile, American citizens who had recently visited Wuhan or other COVID hotspots started falling sick and asking their doctors or health departments whether they had COVID. Since the FDA had essentially banned testing, those departments told their citizens that they couldn’t help and they should just use their best judgment. Most of those people went out and interacted and spread the virus, and incidence started growing exponentially. By March 1, China was testing millions of people a week, South Korea had tested 65,000 people, and the USA had done a grand total of 459 coronavirus tests. The pandemic in these three countries went pretty much how you would expect based on those numbers.
MM: Omicron Variant
November 29, 2021 · Original source
R0 is a measure of how quickly a disease spreads under certain ideal conditions. The original Wuhan strain was probably around 2.5, and the Delta variant was probably around 5. So if this number is higher than 5, it’s more transmissible than Delta. The community prediction is 7.31, so Metaculus predicts it will be significantly more transmissible than Delta.
Inline links: original Wuhan strain, the Delta variant
Your Book Review: Viral
July 30, 2022 · Original source
As a non-fiction book on current events, an unavoidable weakness of Viral is that it does not include recent developments that have come out after the book’s publication. At least one of these developments is important enough for me to mention in this review. In February 2022, three scientific pre-prints [1, 2, 3] were released, related to the spread of SARS-CoV-2 in the Huanan seafood market in the early stage of the pandemic. The Huanan seafood market, located in Wuhan, is thought by natural origins proponents to have been the source of the first zoonotic spillover (or possibly, two separate spillovers) into humans. Advocates of this hypothesis have taken these pre-prints as further confirmation of a zoonotic origin in the market. However, proponents of the lab leak hypothesis have pointed out that they never denied that an early superspreader event occurred in the market – they just think the virus was brought there by an infected human, and spread to others in the crowded and enclosed space. They point to the fact that all of the market animals that were tested for COVID came up negative. Fence-sitters, like Chan, say that the pre-print findings appear to be consistent with both hypotheses.
Inline links: 1, 2, 3, or possibly, two separate spillovers
A lot of these technical points are over my head, so I encourage you to read the pre-prints (as well as the critiques of them) yourself. Here are some more sources to check out about these recent pre-prints and the debate around them [1, 2, 3].
Inline links: 1, 2, 3
Smallpox escaped from research labs in the UK three times from 1966-1978. In fact, the last ever case of smallpox occurred after it had already been eradicated, when it escaped from a medical laboratory in 1978 and infected a medical photographer, who eventually died from the illness. These are only a few of many examples. According to the US Federal Select Agent Program, which oversees the possession and handling of dangerous biological agents and toxins, there were 219 accidental releases of these “select agents” in 2019. So, while accidental lab leaks are uncommon, they’re not unheard of. When it comes to the COVID-19 pandemic, it still makes sense to have a strong prior in favor of the natural origins hypothesis, but the idea that a pathogen can be accidentally released from a lab isn’t some wild, ridiculous idea like believing in alien abductions or Bigfoot or something. 3. The outbreak location in Wuhan appears to be relevant There’s a famous psychology experiment [1] in which participants were told to wait in a room, and their reactions were recorded as the room gradually filled with smoke. In some cases, participants waited alone, while in other cases they waited with a group of people who, unbeknownst to the participant, were actors who had been instructed to ignore the smoke. Of the participants who waited alone, 75% reported the smoke. However, of the participants who waited with the group, only 10% reported the smoke. Photograph of the famous Latané and Darley experiment, cerca 1968. So, what could those participants have been thinking? Maybe something like: Hmm, why’s the room filling up with smoke? Is this a problem? *looks around the room* Well nobody else seems to care, so I guess not. Looking back at the early stages of the COVID-19 pandemic, I think maybe this is why so many of us didn’t think twice about the location of the initial outbreak. Hmm, is it kinda suspicious that this virus broke out near a major virology institute that works on bat coronaviruses? Should we maybe look into that? *looks around* Well nobody else seems to think so, so I guess not. I can’t speak for everyone else, but this was at least my mindset. I had vaguely heard something about how there was a virology research institute close to where the pandemic broke out, and that some conspiracy theorists were claiming it was the source of the virus. I looked around and noticed that nobody was really taking this idea seriously, so I figured I didn’t need to take it seriously either. Also, I was thinking something like: Eh, probably every major city has labs and research institutes doing this kind of research. And I’ll bet they purposely built the virology institute close to where these viruses occur in nature, to give them easy access for sampling. Well, it turns out both of these things are wrong. The type of research conducted at the Wuhan Institute of Virology (WIV) is pretty rare and specialized. It includes things like creation of chimeric coronaviruses [1, 2], infecting humanized mice with bat coronaviruses, and other types of gain of function research, which Chan and Ridley devote a chapter to. The WIV is one of only a few institutions in the world doing this type of research. It’s not the case, as I had assumed, that every major university has a couple labs doing similar work. So it does seem like a pretty remarkable coincidence that the outbreak happened in Wuhan. But maybe they purposely built the Wuhan Institute of Virology close to where these viruses are found in nature? Well, this also turns out to be wrong. The areas where viruses most similar to SARS-CoV-2 are found in nature are Yunnan province and Laos, which are more than a thousand kilometers away from Wuhan. The authors put this distance in perspective by noting that it’s more than the distance between Orlando and NYC. Image source: https://www.bloomberg.com/news/features/2020-12-30/china-is-making-it-harder-to-solve-the-mystery-of-how-covid-began If SARS-CoV-2 originated in an animal somewhere around the Yunnan / Laos area, how did it make it all the way to Wuhan without leaving a trail along the way? 4. The story of RaTG13 Although I enjoyed the book, I do have one pretty major criticism. The authors repeatedly make the claim that a virus called RaTG13, which was being studied at the WIV before the pandemic, is the closest known genetic match to SARS-CoV-2. But this claim is outdated and no longer correct. In September 2021 researchers identified a virus called BANAL-52 in Laos that’s a 96.8% match to SARS-CoV-2, closer than RaTG13’s 96.2% match. (Important note: a 96.8% match is still a long way off in genomic space, and does not imply that this is the same virus as SARS-CoV-2, or even necessarily a progenitor.) At first I thought maybe the authors didn’t mention BANAL-52 because it was discovered after the book was published, but this isn’t the case – Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published. Although I’m writing an overall-positive review here, I don’t want to go easy on the book where serious criticism is warranted. It’s completely unacceptable that BANAL-52 wasn’t mentioned. Even if it would have been inconvenient from a publishing standpoint, the authors should have rewritten the RaTG13 chapter, or at least included an addendum about the discovery of BANAL-52. With that being said, I think the story of RaTG13 is still interesting and important, so I’ll give a quick summary here. At the start of the pandemic in 2020, SARS-CoV-2 was quickly sequenced, and the full genome sequence was published by Dr. Shi Zhengli’s team at the WIV. In this paper, they also briefly mentioned that the genome was a 96.2% match with another bat coronavirus called RaTG13 – the closest known match at the time. Oddly, the mention of RaTG13 did not include any reference, footnote, or link to any previously published sequence. Although the WIV didn’t provide details on this mysterious RaTG13 virus, a group of internet volunteers, including both amateurs as well as professional scientists working in their free time, began to investigate. This loose collection of open-source researchers, called DRASTIC, uncovered a medical thesis describing an outbreak of a mysterious disease in 2012. Six men who had been working in a bat-infested mine in Mojiang County, China, fell ill and were admitted to a hospital with symptoms including dry coughs, shortness of breath, fevers, muscle aches, headaches, and fatigue. Three of the men eventually died of this mysterious illness. In the years following this incident, teams of researchers (including a team led by Dr. Shi Zhengli of the WIV) were sent to investigate the cause of this illness and collect samples from the Mojiang mine. This sampling led to the discovery of a novel SARS-like coronavirus in 2013, and a part of its genomic sequence was published under the name BtCoV/4991 in 2016. The DRASTIC researchers discovered that RaTG13 was genetically identical to the BtCoV/4991 sequence from the Mojiang mine – it was the same virus, and had just been renamed for some reason, without any public record of the change. They also discovered that at least eight other closely related coronaviruses were also sampled from this mine and brought to the WIV. Although unhelpful throughout the investigation, the WIV eventually verified these facts when pressed on them, and an addendum was added to the original paper confirming DRASTIC’s account of the origin of RaTG13. So what should we make of this? Well, as I mentioned before, RaTG13 is no longer the closest known genetic match to SARS-CoV-2, so maybe the whole story is less important as it pertains to the origin of the pandemic. But the discovery of BANAL-52 doesn’t really resolve things either [2]. Laos is very far away from Wuhan (actually even further than Yunnan), so we’re left with the same question as before – how did SARS-CoV-2 make it all the way to Wuhan from such a distant natural reservoir without leaving a trail along the way? 5. Lack of institutional transparency and competence A lot of the book is devoted to criticizing the Chinese government’s lack of transparency during the pandemic. Some brief examples: In the early days of the initial outbreak in Wuhan, hundreds of people were investigated and punished for the crime of “spreading rumors”. This included whistleblowing doctors who attempted to warn others [3] about the spread of the disease and its human-to-human transmission, which was being denied by the Chinese government at the time.
Inline links: eventually died from the illness., psychology experiment, https://substackcdn.com/image/fetch/$s_!mhjk!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Fc8199936-a484-4951-b90e-8ea42a0a4ebc_1600x758.jpeg, 1, 2, infecting humanized mice with bat coronaviruses, https://substackcdn.com/image/fetch/$s_!6khv!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Fac0a8f26-8f26-4712-b245-797d478ae385_1246x642.png, https://www.bloomberg.com/news/features/2020-12-30/china-is-making-it-harder-to-solve-the-mystery-of-how-covid-began, virus called BANAL-52, published by Dr. Shi Zhengli’s team at the WIV., addendum
If The Media Reported On Other Things Like It Does Effective Altruism
November 25, 2022 · Original source
8. Some epidemiologists are worrying that a new virus from Wuhan could become a wider catastrophe. Their message is infecting people around the world with fear and xenophobia, spreading faster than any plague. Perhaps they should consider that in some sense, they themselves are the global pandemic. OR DID I JUST BLOW YOUR MIND? SERIOUSLY, DID YOU SEE WHAT I DID THERE? IN CASE YOU WANT TO SEND ME PULITZER PRIZES FOR THAT AMAZING CONCLUSION, MY ADDRESS IS . . .
Highlights From The Comments On Housing Density And Prices
May 10, 2023 · Original source
Kangbashi, China’s most famous ghost city. What are housing prices like in the ghost city? Again from Bloomberg: Sitting on the southern outskirts of Inner Mongolia’s Ordos City (population 2.2 million), Kangbashi was the archetypal ghost city 10 years ago, with barren boulevards and empty buildings standing forlornly in the desert. Local officials are adamant that things have changed. They say 91% of homes in the district are occupied. In fact, after a yearslong construction freeze, the government approved six housing projects in 2020 and expects 3,000 homes to be built by the end of this year. Apartments in a new development are selling for 9,500 yuan per square meter, and downtown they go for 15,000 to 16,000 yuan, according to Liu Yueyue, 28, a salesman at a new residential development in the district’s northeast. “Would houses in a ghost town sell at such high prices?” asks Liu. Half of his customers come from outside Kangbashi, and most are parents who want to send their children to the well-regarded local schools, he says. Looking at this list of real estate prices across Chinese cities, Kangbashi seems squarely in the middle - for example, Wuhan and Xian are also in the 15,000 - 16,000 range. I claim this supports my argument: surely twenty years ago, houses in this particular deserted corner of Inner Mongolia would have been dirt cheap (if any even existed). But if you build a city there, it becomes just as expensive as any other city! Here it’s very obvious that the density caused the high prices instead of the other way around. Still, the Chinese housing market is weird, with significant vacancies even in expensive, well-developed cities. Paul Botts: No official vacancy rates are published in China and no specific definition of it exists there. Various think tanks and researchers both within that country and elsewhere have published estimates ranging from as low as 11 percent to as high as 24 percent. Those estimates have been for varying samples of Chinese cities, have used various definitions of housing vacancy rate, etc. The best (as in most systematic) estimate yet produced has come from researchers at a university in Liaoning. They used night-time urban lightsheds captured by a new (2018 launch) Chinese satellite having a new level of light sensing technology which allows separating out light from parks and plazas. They covered a large sample (49 cities), and made their sample representative of city type, city size, regions within China, etc. They also crossed-referenced with local housing data to ensure accurate balancing of their sample and to confirm that the satellite was successfully identifying light coming from housing blocks. They found vacancy rates of just under 20 percent in China's Tier 1 cities, and found rates above 20 percent in 40 of the 49 cities. They found the highest vacancy rates in western and northeastern cities, which are also the newest ones; that finding is consistent with the hypothesis of significant numbers of recently-built ghost cities. https://www.researchgate.net/publication/345092218_Housing_Vacancy_Rate_in_Major_Cities_in_China_Perspectives_from_Nighttime_Light_Data And Phil H (author of the blog Tang Poetry) writes: The price of housing in China has skyrocketed over the past few decades, as all those extra apartments have been built. I live in a pleasant but unremarkable southern city, and I paid London prices (about 4.5m yuan/$650k for a 1,300 sq ft flat). That seems to match Scott's hypothesis that high density leads to high prices. House prices here have risen much faster than incomes. They've risen in rural areas, too, but the increases in price in cities have been stratospheric. 4. Comments Accusing Me Of Not Considering Tokyo, Even Though I Included A Section In The Post On Why I Didn’t Think Tokyo Was Relevant I won’t name and shame people, but for example: You excluded Tokyo from your dataset. Tokyo has much higher density than SF and much lower price per sqft. Tokyo just kills this. Tokyo is bigger than New York and has significantly lower rent because they build more housing! This is in a wealthy country with even lower interest rates than the US. I don't think you have justified excluding non-US metros, like Tokyo, or Auckland. Doesn't this lead to the natural conclusion that there is a sufficient level of housing to build, and that the problem is that the USA's many metros are structured to prevent housing? It seems like you're just arguing that US metros are bad at building housing, which is also what Matt Yglesias is arguing. "Change my mind about housing, but don't mention Tokyo" is like saying "Change my mind about gun possession, but don't mention Switzerland." You can't test the effect of allowing new housing unless you're willing to look at cities that do, in fact, allow it. Tokyo and NYC both attract tons of new residents But Tokyo's housing rents have been stable, while NYC rents keep rising. Why? Tokyo has permissive housing construction laws. NYC makes building new housing almost illegal. Yes, dense cities are attractive, and that makes them get more dense over time. But it only makes them more expensive if you forbid new housing to keep up with the new residents. Tokyo! But I’m like the 10th person to bring it up… As I wrote on the original post (not even edited in! it’s been there the whole time!): I worry someone will bring up Tokyo as a counterexample. But I think Tokyo managed to build its way to low housing prices in the context of the rest of Japan also having good housing policy. Even if that isn’t true, Tokyo on its own is a quarter of the Japanese market, so it might be able to exhaust the entire pool of Japanese house-seekers by itself! That is, yes, you’re all correct that cities are only expensive in the context of more demand for city housing than the (NIMBY-constrained) city housing market can currently supply. You are all correct that if this problem were solved at the national level, then city housing would be cheap, and every additional city house would make it cheaper. My claim is that marginal changes - like Oakland building an extra 10,000 units, but everyone else staying the same - will most likely increase Oakland prices. Yes, if Oakland unilaterally built 50 million units, that would soak up the entire excess demand and probably lower prices everywhere (including Oakland). Yes, if the entire US switched to good housing policy at the same time, that would probably lower prices everywhere (including Oakland). But if we don’t do any of that stuff, and just build another 10,000 houses in Oakland, I think it would probably increase prices in Oakland. Some other people brought up that Japan has a declining population, and it’s much easier to have low house prices when your population is declining (compared to some previous time when number of houses presumably matched number of people), but ddd pointed out that people continue to migrate from the Japanese countryside to Tokyo, so its population continues to increase. Also, Mike (I’m stitching together two comments here): In a country with a declining population, you would expect that fewer homes are being built per capita because there's little to no competition for existing homes. But it's exactly the opposite! Japan builds far more homes per capita than the US does, despite their declining population […] As a result, the average Japanese home is very new and the average house is torn down and replaced after a relatively short 30 years. They're living in nice new homes for cheaper. 5. Comments Accusing Me Of Not Understanding Economics Maximum Limelihood Estimator writes: I think you're making a very common mistake here of confusing supply/demand with *quantity* supplied or quantity demanded. (This is very common! we teach students about this in micro 101 because it's so easy to make!) What you're seeing is that the quantity supplied is correlated with housing prices (true!). But this is very different from establishing that the supply curve--i.e. the amount of housing that would be produced at any given price, and what moves up/down when we regulate/deregulate supply--is positively correlated with price. Figuring out what supply curves look like is a lot less intuitive and requires some high-grade econometrics, which is why economists had to set up a whole commission just to study this particular problem (the Cowles Commission). In terms of resources for understanding how these concepts are different, a micro 101 textbook will cover this distinction. For the econometrics side of this, I've heard good things about Scott Cunningham's *Causal Inference Mixtape*, although I haven't personally used it. My claim is that increasing density within a city shifts the demand curve for housing within that city, because of increasing desirability. MLE later gets more on point: The effect you're discussing here is kind of real in a sense. When the marginal utility of housing increases for *other* people, density arguably becomes more desirable for me, which is kind of like the demand curve shifting up. These are called bandwagon goods and discussed here: http://econfac.bsu.edu/research/workingpapers/bsuecwp200804gisser.pdf In theory, the bandwagon effect could be so strong that parts of the demand curve are upward-sloping. Solutions like this are not, technically, prohibited by the laws of mathematics, just the laws of economics. (And arguably of physics--see paper for conditions where these kinds of bandwagon effects imply the amount of housing in the city would have to be negative). In practice, this effect exists but just can't overcome the normal, non-weird economics that says "making more of a good makes the prices fall." Again, I claim the existence of Manhattan vs. Conanicut shows that sometimes it does. I cannot find the words “housing”, “real estate”, or “land value” anywhere in that paper. Alex Poterack writes: There's two things going on here: confusing shifts in demand with movement along the demand curve, and getting causation backwards. You're assuming density causes prosperity, rather than prosperity causing density. There are ways the former can happen, but the bigger thing is that, for a wide range of historical reasons, you can make a lot of money in NYC and SF, so lots of people want to live there, so they get very dense. This is the prosperity shifting demand right, so at any given price, more people want to live there; this drives prices up, and they go higher the more fixed supply is. If you built a bunch of housing in Oakland, lots of people would move there because it's cheaper, which is movement along the demand curve; it's still the same number of people who want to live there at any price. Now, it's possible that the increased number of people living there makes the city more prosperous (this is the phenomenon of induced demand), which would shift demand right, but there are way more differences between NYC/SF and Oakland than just the density, so I don't think it would shift demand enough to offset this. In particular, if it's just a small increase in small, it's also a small increase in density, so there's almost no shift in demand (but there is movement along the curve). I still think this is missing my point, but I present it here in case anyone else is enlightened by it and wants to try further to convince me I’m making this mistake. 6. Comments By Famous People Who Potentially Have Good Opinions Scott Sumner is an economist and blogger; he writes: It is certainly the case that building more housing can make a city more desirable, and that this effect could be so strong that it overwhelms the price depressing impact of a greater quantity supplied. But studies suggest that this is not generally the case. Texas provides a nice case study. Among Texas’s big metro areas, Austin has the tightest restrictions on building and Houston is the most willing to allow dense infill development. Even though Houston is the larger city, house prices are far higher in Austin: Houston pretty much describes the “Oakland with more housing” outcome that Alexander views as somewhat far-fetched. Only in this case, it’s Austin with more housing. Alexander seems too quick to accept the, “If you build it they will come” idea—that you can build more housing and thereby boost demand so much that prices actually rise. I started the post with a graph of about 50 cities, showing a positive correlation between density and price. I’m having trouble seeing how Sumner’s point isn’t just “if you remove 48 of those cities and cherry-pick two, the relationship is negative”. My attempt to place Austin and Houston on the original graph, using Sumner’s data plus a few other things available online. Why weren’t they on there already? Maybe because the graph is metro areas and Sumner was talking about Austin and Houston as cities, but I’m not sure and agree this is confusing. Everyone knows Austin is more expensive than Houston because Austin is a trendy tech and culture hub and Houston isn’t (and relatedly, because Austin’s median family income is 50% higher than Houston’s). Unless someone wants to claim that its failure to build housing helped turn it into a trendy tech and culture hub, I don’t think there’s much point to this comparison. It’s true that Houston’s bigger size didn’t let it leapfrog over Austin to become a trendy tech and culture hub, which goes against some of what I claimed in the first part of this post. But I never claimed there would be a perfect 1-1 correlation between city size and trendiness, or that you could never find a pair of cities where one was bigger but the other was more trendy. Just that there would be a correlation. Moving on: Here’s the problem with this argument. It mixes up population change due to economic effects such as the benefits of agglomeration, with population changes due to regulatory changes such as less strict zoning. If you look at things this way, then the stylized facts work against Alexander’s argument. Over the past 50 years, increasingly strict zoning has reduced housing construction on big cities like New York and San Francisco. As a result, their populations have increased by less than in cities with less strict zoning, such as Houston. If Alexander were correct, then the price gap between the tightly controlled cities on the coast and the more laissez-faire cities of Middle America should have shrunk over time. Instead, the price gap has widened. New York and San Francisco were always more expensive than other cites, but with tighter zoning and less new construction the gap has become far wider. During the last fifty years, there was also deindustrialization and demographic sorting. This is just the Austin vs. Houston story all over again. Alexander is implicitly viewing this outcome as a “problem” for the city that builds more housing. They must sacrifice so that the rest of the country can gain. But in his scenario, Oakland is better off. Indeed if it were not better off, then why would more people choose to live in Oakland? In order for it to be true that building more housing boosts housing prices, it must also be true that the quality of existing houses (including neighborhood effects) rises by more than enough to offset the increase in supply. That means the new housing construction must make Oakland such a desirable place to live that the amenity effect overwhelms the quantity effect [...] Of course, economic change always has winners and losers. Here’s how I would describe the impact of allowing more housing construction in Oakland, in the unlikely event that this did raise housing prices: 1. America would benefit. 2. Oakland would benefit. 3. Poor people in America would benefit, in aggregate. 4. Affluent people in America would benefit, in aggregate. 5. Homeowners in Oakland would benefit. 6. Some renters in Oakland would benefit (from a more economically dynamic city.) 7. Some renters in Oakland would suffer from higher rents. In the much more likely case where new housing construction would lower prices, the impact described in #5 and #7 might reverse. Either way, there is no defensible argument for not building more housing in Oakland, regardless of the impact on price. If building more housing reduces its price, then there is a strong argument for allowing more housing construction. If building more housing raises its price, then the argument for more construction is even stronger. I agree with all this. Jeremiah Johnson is a co-founder of the Center for New Liberalism, host of the Neoliberal Podcast, and a YIMBY activist (not to be confused with Jeremiah “Liver-Eating” Johnson, who killed 300 Native Americans and ate their livers). He writes: Here's why you're wrong in a single sentence: Demand causes high prices, not new units. Prices are high in SF and NYC because those are desirable places to live for a huge number of people. People all over the country and the world would live there if they could, and prices reflect that. The fact that the densest cities are the most expensive is true. But the high prices are not caused by density - rather, the density and the high prices are both a consequence of crushingly high demand […] There's a feedback loop, but what matters here is the elasticity, which is less than one. We can measure this empirically. New housing lowers prices via the mechanism of adding supply, which is basic economics and how we expect markets to work. New housing could raise prices if it also made the city a more desirable place to live and shifted people's preferences, such that there was more demand to live there after the new housing is built. If you think it's unclear which of these effects would dominate, luckily we have empirical data that over and over and over shows adding housing supply does indeed lower prices on a local level. This is a fairly well established result that replicates well. edit: I'm actually thinking about drawing out the weighted DAG graphs here to make the conceptual stuff easier, but it would be pretty long. I'd love to do this as a guest post. I’m skeptical of the empirical results because they don’t match the much stronger “Manhattan vs. Conanicut island” empirical results, and if I try to think about why, the best explanation I can think of is that the Manhattan experiment has been going on longer (ie long enough for Manhattan’s extra residents to found businesses and institutions that attract new people). I’ve told him he can try pitching this guest post to me; in either case, I would be interested in seeing the graphs. Several other people also posted this graph that Johnson helped make famous: Hopefully by now you can predict my objection: the places in the southeast corner are mostly unfashionable red state Sun Belt cities; the places in the northwest corner are mostly trendy liberal coastal cities. My conclusion is that trendy liberal coastal cities are both more NIMBY and more desirable, and if you use this to draw any conclusions about housing policy you’ll just end up confused. But maybe I should take this same lesson to heart myself. Dense cities are mostly trendy liberal coastal cities; uninhabited tundra in North Dakota isn’t. Maybe the demand is just for trendy liberal coastal cities, and once you attain that status, extra density doesn’t matter that much. Maybe Oakland has already maxed out its “trendy liberal coastal city” status, and even if it became Manhattan-sized, it wouldn’t get any trendier, or would get trendier only with a long time lag. There are a few very trendy small coastal villages in California (think eg Sea Ranch); maybe these (rather than North Dakota) are the natural control group for San Francisco. I think they are still cheaper than SF, but maybe not by very much. Cameron Murray is a housing economist whose work some other commenters recommended; he also writes the blog Fresh Economic Thinking. He very kindly showed up and wrote: I think you are in general right that agglomeration effects are real, which is why bigger cities have higher value to residents. I agree that people move locations. But I think you can go a step further. If one city is growing faster and densifying, surely those people are not demanding homes in other cities and those cities build slower. This is part of the spatial equilibrium story that further makes claims about “build density and get cheap homes” less plausible. 7. My Final Thoughts + Poll Thanks to everyone who commented on this post and helped me refine my thoughts. I’m willing to concede the following points: It might be that only attracting the sort of educated people who found companies, universities, etc will make housing prices go up. Less educated people will take more jobs than they create and not ratchet up the city’s desirability level. (I’d previously told commenters talking about “gentrification” that it was irrelevant to the mechanism I was talking about here, but maybe it isn’t - maybe “gentrifiers” are the people creating more jobs and institutions than they consume, and so homes that attract them in particular will increase demand more than they increase supply? Maybe this discussion does reduce to the gentrification discussion?)
Inline links: this list of real estate prices across Chinese cities, Paul Botts, https://www.researchgate.net/publication/345092218_Housing_Vacancy_Rate_in_Major_Cities_in_China_Perspectives_from_Nighttime_Light_Data, Tang Poetry, writes, ddd pointed out, Mike, Maximum Limelihood Estimator, gets more on point, http://econfac.bsu.edu/research/workingpapers/bsuecwp200804gisser.pdf, writes, writes:, far higher in Austin, https://substackcdn.com/image/fetch/$s_!tAF8!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F03f2b338-da84-4d85-b71c-0ce2e02e10e1_685x559.png, https://substackcdn.com/image/fetch/$s_!mLwn!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9cf942da-5716-49e8-a177-20140681e7fc_573x421.png, Center for New Liberalism, the Neoliberal Podcast, Jeremiah “Liver-Eating” Johnson, Johnson helped make famous, https://substackcdn.com/image/fetch/$s_!tg5t!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fc490a60a-3e97-4b73-9806-79bb967b5172_608x550.png, Sea Ranch, some other commenters, Fresh Economic Thinking, wrote
Practically-A-Book Review: Rootclaim $100,000 Lab Leak Debate
March 28, 2024 · Original source
Eric Stansifer, an applied mathematician with a PhD from MIT and experience in mathematical virology. …both of whom received $5,000 as payment for their ~1001 hours of work, paid by the two contestants along with their $100,000 table stakes2. The format would be three sessions, each consisting of hour-and-a-half arguments by both sides, then three hours for the debaters to answer questions from the judges and each other. II. The Debate Below, I’ve included the videos from each session, plus my (long) summary if you prefer text. In the second session (on viral genetics) biotech entrepreneur and lab leak expert Yuri Deigin stood in for Saar; Peter continued to represent himself. Session 1: Epidemiology Peter: The first officially confirmed COVID case was a vendor at the Wuhan wet market. So were the next four, and half of the next 40. A heat map of early cases is obviously centered on the wet market, not on the lab. The wet market and the lab are about 6 miles away as the crow flies, or a 15 mile / half hour drive. Location of COVID cases in December 2020. Source: NYT, slightly edited. A map of cases at the wet market itself shows a clear pattern in favor of the very southwest corner: The southwest corner is where most of the wildlife was being sold. Rumor said that included a stall with raccoon-dogs, an animal which is generally teeming with weird coronaviruses, and is a plausible intermediate host between humans and bats: Awwww, come on, you can’t stay mad at this little guy. China said this rumor was false and refused to release any information. Scientists were finally able to confirm the existence of the raccoon-dog shop in the funniest possible way: a virologist had visited Wuhan in 2014, saw the awful conditions in the shop, and took a picture as an example of the kind of place that a future pandemic might start. Source: NPR. To be fair, we have only the scientist’s word that this is why he had the picture. But he definitely did have it. People say it would be a surprising coincidence if a zoonotic coronavirus pandemic just so happened to start in a city with a big coronavirus research lab, and this is true. But it would be an even more surprising coincidence if a lab-leak coronavirus pandemic just so happened to first get detected at a raccoon-dog stall in a wet market! Saar: It’s not clear that the first case was at the wet market; a certain Mr. Chen, with no connection to the market, seems to have fallen sick on December 8. An SCMP article suggested there were 92 previously-undetected cases suspicious for COVID as far back as November. And even if half of the first forty universally-agreed-upon cases had market connections that means another half didn’t. There was a bias towards detecting cases at the market: because authorities thought the market was the origin, and because everyone was thinking about zoonosis after SARS1, they only screened/diagnosed people with a market connection. One of the few non-market-connected COVID cases detected during this period was only detected because he was the relative of a hospital worker; the worker noticed the signs and insisted they go to the hospital despite the lack of a wet market connection. Although the map of positive samples and cases at the market was centered near the raccoon-dog stall, that could be because that area was sampled more; it’s also close to the mahjong room, where visitors and vendors at the market would go and unwind in a tight, poorly ventilated area. The next session will focus more on the WIV, but the short version is that they were doing lots of gain of function research. So one story compatible with the evidence is that a worker at WIV got infected with their modified coronavirus and passed it to his contacts. COVID started spreading quietly a few weeks to months before the first market-related case was detected. This accounts for the 92 earlier cases, Mr. Chen’s case, and the half of officially-detected cases with no wet market association. Then an infected person went to the market, causing a super-spreader event. Some of the infected market patrons went to the hospital, where doctors traced it back to the market and told other doctors to be on the lookout for wet market patrons coming in with weird viral pneumonias. They found some, declared victory, and the few anomalies - like the hospital worker’s relative - were forgotten, or assumed to have wet market connections that nobody could find. China quashed all evidence of the lab research (as was done in previous lab leak cases, eg the USSR) so all we have is the apparent wet market links that Peter found so convincing. Peter: The supposed pre-wet-market cases are confirmed fakes. Yes, the WHO did an investigation of whether there might have been COVID cases circulating before the wet market, and identified 92 unusual pneumonias that merited further review. But their final investigation, which included testing samples from these people after good tests became available, found that none of these people really had COVID. As for Mr. Chen, he said in an interview that he was hospitalized for dental issues on December 8, caught COVID in the hospital on December 16, and then was erroneously reported as “hospitalized for COVID on December 8”. The December 16 date is after the first wet market cases. Further, it seems epidemiologically impossible for COVID to have been circulating much before the first cases were officially detected December 11. The COVID pandemic doubles every 3.5 days. So if the first infection was much earlier - let’s say November 11 - we would expect 256x as much COVID as we actually saw. Even if the first couple of cases were missed because nobody was looking for them, the number of hospitalizations, deaths, etc, in January or whenever were all consistent with the number of people you’d expect if the pandemic started in early December - and not consistent with 256x that many people. So probably we should just accept that the first reported case - a wet market vendor, December 11 - was very early in the pandemic. She wasn’t literally the first case - that would most likely have been someone who worked at the raccoon-dog shop, whose case might (like 95% of COVID cases) have been mild enough not to come to medical attention. But she was certainly very early. Although authorities eventually decided COVID spread through a wet market and started deliberately looking for wet market connections, this only happened on December 30. So the earliest cases - including the 40 very earliest cases where half came from the wet market - weren’t biased (at least not through that particular route). So the claim that “the first case, and half of the first 40 cases, had wet market connections” stands as real and convincing evidence. Although the exact center of the map of positive COVID samples in the wet market was the mahjong room, the samples taken from the mahjong room were not, themselves, positive (cf: although a low-resolution population density map of New York might show Central Park in the exact center of the population density gradient, Central Park does not itself have population). There was no real “super-spreader event” at the wet market. There was a slow burn - one case the first day, a few more the next day, a few more the day after that. It’s hard to see how a single visit from an infected lab worker could do that. So the only way it could possibly be a lab leak is if the lab leaked sometime in late November, infected exactly one lab worker, that worker went straight to the wet market, infected a vendor, then went home, quarantined, recovered, and all other cases were downstream of that first infected wet market vendor. This is unparsimonious. Saar: The only source saying that Mr. Chen got sick early was an anonymous interview. And even if he was later than the first wet market cases, nobody was able to find any wet market connections. This means that whoever infected him was earlier than the index case and not linked to the wet market. Peter argued that COVID couldn’t have been more than a few weeks old when the first wet market cases were detected. But this was based on its known doubling rate. If pre-discovery COVID had a slower doubling time than known COVID, it could have been around longer. And post-lockdown serology suggested numbers that were larger than claimed at the time. So contra Peter’s claims, the infection could have been going on longer, which wouldn’t require the first lab worker to go straight to the market. It could have been weeks. Dr. Jesse Bloom’s investigation of the wet market samples, considered the final and most conclusive, failed to find a clear connection between COVID and raccoon-dogs or any other animals. Although the concentration of positive samples seemed highest near the raccoon dog stall, if you do a formal statistical analysis of which animals’ DNA was found near COVID samples most often, raccoon dogs are near the bottom. The top is wide-mouth bass, which can’t get COVID. This is obviously contamination, probably from infected humans touching wide-mouth bass tanks or something. Although the Chinese data included a negative sample from a mahjong table, it included a mention of poultry being sold nearby, which might mean this wasn’t the mahjong room itself, but some other mahjong table at a poultry shop elsewhere in the market, and (dry) mahjong tables might not hold the virus well anyway. Peter: Raccoon-dogs were sold in various cages at various stalls, separated by air gaps big enough to present a challenge for COVID transmission, and there’s no reason to think that one raccoon-dog would automatically pass it to all the others. The statistical analysis just proves there were many raccoon-dogs who didn’t have COVID. But you only need one. The raccoon dog shop and the drain leading out of the raccoon dog shop had some of the highest positive sample rates, which is more interesting than a statistical analysis which everyone agrees must be wrong (since it favors bass). It’s unclear why the negative mahjong sample says something about poultry, but based on the stated location, it’s definitely the one in the mahjong room. Session 1.5: Lineages This was technically part of Session 2, but formed enough of a discrete topic that I found it confusing to intermix it with all the other viral genetics points. I’m spinning it out into a separate summary, but the videos are all in the next session. Yuri: The coronavirus eventually mutated into many different strains. But the first big split, seen in some of the earliest samples, is between two different sub-strains called Lineage A and Lineage B, which differ by two mutations. In these two mutations, Lineage A is the same as BANAL-52, a bat virus which is the closest-known relative of COVID, but Lineage B is different. Since COVID probably evolved from something like BANAL-52, Lineage A must have come first, spread for a while, and then gotten two new mutations, turning it into Lineage B. All of the cases at the wet market, including the first detected case, were Lineage B. Lineage A wasn’t discovered until about a week later, and none of the Lineage A patients had been to the wet market. Lineage A (left) was used by the Minoan Cretans, but has never been deciphered. Lineage B (right) was used by the Mycaeneans for lists of palace goods. This matches Saar’s story above. The lab leaked to somewhere else in Wuhan, not the wet market. The virus spread undetected in the population for a while. During this time, it mutated to Lineage B. Then one of the people with Lineage B went to the wet market and started a superspreader event. The authorities sampled the patients, found Lineage B, then started looking elsewhere. Later they detected some of the earlier Lineage A cases. The market is unlikely to be the origin of the pandemic, because the original Lineage A strain wasn’t found there. Peter: Although Lineage A is evolutionarily older, Lineage B started spreading in humans first. We know this because Lineage B is more common. Throughout the early pandemic, until the D614G variant drove all other strains extinct, a consistent 2/3 of the cases were B, compared to 1/3 A. Both strains spread at the same rate, so the best explanation is that B started earlier than A. Since COVID doubles every 3-4 days, probably Lineage B started 3-4 days earlier than Lineage A, which explains why it’s always been twice as many cases. But also, Lineage B also has more internal genetic diversity than Lineage A. In general, older viruses have more genetic diversity (the “molecular clock”). This is further evidence that B started spreading first. Pekar 2022 and Pipes 2021 do analyses with known parameters for spread rate and diversity, and find 90%+ odds that Lineage B was the first one in humans. Why did the older strain start spreading later? Probably the virus crossed from bats into raccoon-dogs on some raccoon-dog farm out in the country. It spread in the raccoon-dogs for a while, racking up mutations, including the (less mutated) Lineage A strain and the (slightly more mutated) Lineage B strain. Then several raccoon-dogs were taken to Wuhan for sale, including one with Lineage A and another with Lineage B. The one with Lineage B passed its virus to humans earlier. Then 3-4 days later, the Lineage A one passed its virus to humans. Lineage A was first found in a Wuhan neighborhood right next to the wet market (closer to the wet market than 97% of Wuhan’s population). Again, it would be a bizarre coincidence if a lab leak pandemic was first detected at a wet market. But it would be an even more bizarre coincidence if a lab leak pandemic separated into two strains, and both were first detected at a wet market! Although no known wet market cases were Lineage A, a positive Lineage A environmental sample was found at the wet market, and everyone agrees most cases went undetected. So maybe the Lineage B raccoon-dog spread its virus to a vendor, and that sub-strain mostly stayed in the market. But the Lineage A raccoon-dog spread its virus to a customer, who went back to his house nearby, and that strain spread in the neighborhoods next to the market. This is the only story that explains the evolutionary precedence of A, the greater spread and older molecular clock of B, and the fact that both strains were first found very close to the wet market. Yuri/Saar: Lineage B could be more common and diverse because it got the advantage of a super-spreader event in the wet market. There are a few scattered cases of intermediates between A and B, and a few other scattered cases of lineages that seem even more ancestral (ie closer to the bat virus) than either. This doesn’t make sense in a double spillover hypothesis. But it does make sense if the lineages separated in human transmission somewhere between the lab and the first super-spreader event at the wet market. Peter: Again, the wet market wasn’t a super-spreader event. COVID spread in the wet market at exactly its normal spread rate, doubling about once every 3.5 days. Stop calling the wet market a super-spreader event. The scattered cases of “intermediates” are sequencing errors. They were all found by the same computer software, which “autofills” unsequenced bases in a genome to the most plausible guess. Because Lineage B was already in the software, depending on which part of a Lineage A virus you sequenced, you might get one half or the other autofilled as Lineage B, which looked like an “intermediate”. We know this because all the supposed “intermediates” were partial cases sequenced by this particular software. We can confirm this by noting that there are too many intermediates! That is, where Lineage A is (T/C) and Lineage B is (C/T), the software found both (T/T) “intermediates” and (C/C) “intermediates”. But obviously there can only be one real intermediate form, and we have to dismiss one or the other. But in fact we can dismiss both, because they were both caused by the same software bug. The scattered “progenitor” cases - those closer to the ancestral bat virus than either A or B - are reversions, ie cases where a new mutation in the virus happened to hit an already-mutated base and shift it back towards the ancestral virus. We know this because all of these “progenitors” were scattered cases found months after the pandemic started, often in entirely different countries from Wuhan. If these were real progenitor viruses, they would have either fizzled out or exploded into a substantial portion of all cases, not be found one time in one guy in Malaysia. Given the number of mutations the virus developed over the course of the pandemic, it’s inevitable that some of them would be mutations that bring it closer to the original bat virus, and in fact we find the number of “progenitors” found very nicely matches the number of progenitor-appearing viruses we would expect by chance. And in many cases, we know the “progenitors” are newer than the original lineages, because they also have some of the later mutations that Lineage A or B picked up along the way, alongside their apparent ancestral-bat-virus-like mutations. Session 2: Viral Genetics Yuri: Two years before COVID, scientists at the Wuhan Institute of Virology, together with colleagues at the University of North Carolina, sent in a grant proposal for the DEFUSE program. This program, intended to locate and better understand potential future pandemic viruses, involved going into bat caves and collecting new coronaviruses. Once they had them, they would do gain-of-function: specifically, they would add a furin cleavage site to make them more infectious and see what happened. (quick interlude: COVID’s spike protein has two sections: one binds to human cells through the ACE2 receptor, the other helps fuse with the cell after binding. In order to avoid the immune system, it hides both of these into one spike. But when it reaches a cell, it needs to separate them again. It takes advantage of a human respiratory enzyme, furin, to do the separation - this also ensures that it only infects its primary target, human respiratory cells. The part of COVID that lets it get separated by furin is called the “furin cleavage site”. COVID’s bat-virus ancestors were gastrointestinal viruses; the addition of a furin cleavage site was what made them respiratory viruses.) We’ve found two close relatives of COVID: bat viruses called RATG-13 and BANAL-52. In particular, COVID looks more or less like BANAL-52 plus a furin cleavage site. There are 1500 sarbecoviruses, members of the family of viruses that includes SARS and SARS2/COVID. None of them except COVID have furin cleavage sites. BANAL-52, COVID’s closest ancestor, doesn’t even have anything resembling one that could mutate into a functional furin cleavage site like COVID’s. Instead, COVID - which mostly just resembles BANAL-52 with a few scattered single-point mutations - has twelve completely new nucleotides in a row - a fully formed furin cleavage site that came out of nowhere. There is nowhere else in the genome that COVID differs from BANAL-52 in such a profound way. It’s just BANAL-52 plus a little bit of random mutation plus a fully-formed furin cleavage site that came out of nowhere. Further, the furin cleavage site is weird. It uses the protein arginine twice. But instead of the nucleotides coding for arginine in the usual viral way, both times it uses the codons CGG - the way that higher animals code for arginine. This works fine - it’s just not how viruses do it. So the obvious conclusion is that WIV, which said in 2018 that it was going to find viruses and add furin cleavage sites to them, found a close relative of BANAL-52 and added a furin cleavage site. Since they were humans, and most familiar with the human way of encoding arginine, they added it as CGG both times. COVID seemed surprisingly optimized for infecting humans. Of fifty animals it was tested in, including the usual coronavirus intermediate hosts (pangolins, raccoon-dogs, etc), it was best at infecting human cells. Further, a virus that enters a new species will usually show a burst of mutations as it “figures out” the best way to adapt to that species’ unique biology. But COVID has had a pretty constant mutation rate in humans, from the beginning of the pandemic to the end. That suggests it was already adapted to humans. This could be because the lab screened for viruses with existing adaptations, because they passed it through humanized mice in the lab, or because it adapted in the hundreds of undetected cases that happened between the lab and detection in the wet market. Usually, research with potentially dangerous coronaviruses is done in BSL-3 or 4, ie high to very-high security. But WIV was irresponsibly doing it in BSL-2, ie medium security. The researchers weren’t even required to wear masks. In general, about 1/500 labs will leak any given pathogen they’re working on (?!). But because WIV was researching such an infectious virus in such an irresponsible way, the odds of a leak were much higher. The most likely explanation for all these facts is that WIV went ahead and did the gain-of-function research they said they were going to do (the particular DEFUSE grant proposal we know about got rejected, but it proves that Wuhan wanted to do this, and they could easily have gotten funding somewhere else, or done it out of their regular budget). They found a close relative of BANAL-52 and added a furin cleavage site as a simple twelve-nucleotide insertion, using the human method of encoding arginine that their genetic engineers were familiar with. Then it leaked, spread for a while in the general Wuhan population, and eventually made it to the wet market where it got detected. Peter: As mentioned earlier, the DEFUSE grant was rejected. Further, the grant said that the Wuhan Institute of Virology was responsible for finding the viruses, and the University of North Carolina would do all the gain-of-function research. This was a reasonable division of labor, since UNC was actually good at gain-of-function research, and WIV mostly wasn’t. They had done a few very simple gain-of-function projects before, but weren’t really set up for this particular proposal and were happy to leave it for their American colleagues. Even if WIV did try to create COVID, they couldn’t have. As Yuri said, COVID looks like BANAL-52 plus a furin cleavage site. But WIV didn’t have BANAL-52. It wasn’t discovered until after the COVID pandemic started, when scientists scoured the area for potential COVID relatives. WIV had a more distant COVID relative, RATG-13. But you can’t create COVID from RATG-13; they’re too different. You would need BANAL-52, or some as-yet-undiscovered extremely close relative. WIV had neither. Are we sure they had neither? Yes. Remember, WIV’s whole job was looking for new coronaviruses. They published lists of which ones they had found pretty regularly. They published their last list in mid-2019, just a few months before the pandemic. Although lab leak proponents claimed these lists showed weird discrepancies, this was just their inability to keep names consistent, and all the lists showed basically the same viruses (plus a few extra on the later ones, as they kept discovering more). The lists didn’t include BANAL-52 or any other suitable COVID relatives - only RATG-13, which isn’t close enough to work. Could they have been keeping their discovery of BANAL-52 secret? No. Pre-pandemic, there was nothing interesting about it; our understanding of virology wasn’t good enough to point this out as a potential pandemic candidate. WIV did its gain-of-function research openly and proudly (before the pandemic, gain-of-function wasn’t as unpopular as it is now) so it’s not like they wanted to keep it secret because they might gain-of-function it later. Their lists very clearly showed they had no virus they could create COVID from, and they had no reason to hide it if they did. COVID’s furin cleavage site is admittedly unusual. But it’s unusual in a way that looks natural rather than man-made. Labs don’t usually add furin cleavage sites through nucleotide insertions (they usually mutate what’s already there). On the other hand, viruses get weird insertions of 12+ nucleotides in nature. For example, HKU1 is another emergent Chinese coronavirus that caused a small outbreak of pneumonia in 2004. It had a 15 nucleotide insertion right next to its furin cleavage site. Later strains of COVID got further 12 - 15 nucleotide insertions. Plenty of flus have 12 to 15 nucleotide insertions compared to other earlier flu strains. Sometimes insertions happen because of a mistake in viral replication. Other times the virus gets confused between its own RNA and its host’s, and splices a bit of the host RNA into the virus. This would neatly explain why the insertion used the unusual coding CGG for arginine, which is common in animals but rare in viruses. On the other hand, it’s not that rare in viruses - COVID uses CGG for arginine about 3% of the time. And human engineers don’t necessarily use it any more than that - Peter was able to find one example of humans adding arginine to a virus, and 0 out of the 5 arginines added were CGG. COVID’s furin cleavage site is a mess. When humans are inserting furin cleavage sites into viruses for gain-of-function, the standard practice is RRKR, a very nice and simple furin cleavage site which works well. COVID uses PRRAR, a bizarre furin cleavage site which no human has ever used before, and which virologists expected to work poorly. They later found that an adjacent part of COVID’s genome twisted the protein in an unusual way that allowed PRRAR to be a viable furin cleavage site, but this discovery took a lot of computer power, and was only made after COVID became important. The Wuhan virologists supposedly doing gain-of-function research on COVID shouldn’t have known this would work. Why didn’t they just use the standard RRKR site, which would have worked better? Everyone thinks it works better! Even the virus eventually decided it worked better - sometime during the course of the pandemic, it mutated away from its weird PRRAR furin cleavage site towards a more normal form. Further, COVID’s furin cleavage site was inserted via what seems to be a frameshift mutation - it wasn’t a clean insertion of the amino acids that formed the site, it was an insertion of a sequence which changed the context of the surrounding nucleotides into the amino acids that formed the site. This is a pointless too-clever-by-half “flourish” that there would be no reason for a human engineer to do. But it’s exactly the kind of weird thing that happens in the random chance of evolution. COVID is hard to culture. If you culture it in most standard media or animals, it will quickly develop characteristic mutations. But the original Wuhan strains didn’t have these mutations. The only ways to culture it without mutations are in human airway cells, or (apparently) in live raccoon-dogs. Getting human airway cells requires a donor (ie someone who donates their body to science), and Wuhan had never done this before (it was one of the technologies only used at the superior North Carolina site). As for raccoon-dogs, it sure does seems suspicious that the virus is already suited to them. The claim that COVID is uniquely adapted to humans is false. The paper that claimed that defined how well COVID was adapted to different animals by those animals’ difference (on the relevant cell receptors) from humans. So in its methodology, humans came out #1 by default. If you don’t do that, COVID is better-adapted to many other animals. It’s not necessarily true that viruses see a burst of mutations when they enter a new host. COVID spread to deer and mink, and in neither case was there a burst of mutations. COVID has a pretty simple job of infecting respiratory cells and is already very good at it, regardless of species. In Yuri’s model, Wuhan Institute of Virology picked up a discarded grant and decided to do the gain-of-function half allotted to a different university, despite their relative inexperience. They skipped over all the SARS-like viruses they were supposed to work on, and all the standard gain-of-function model backbones, in favor of BANAL-52, a virus which would not be discovered for another two years, but which they somehow had samples of, which they had for some reason decided to keep secret despite its total lack of interestingness. Then they would have had to eschew all usual gain-of-function practices in favor of inserting a weird furin cleavage site that shouldn’t have worked according to the theory they had at the time, via a frameshift mutation. Then they would have had to culture it, a technique beyond their limited capabilities. Then it would have had to leak, and magically show up again in front of the raccoon-dog stall at a wet market. Yuri: WIV wouldn’t have needed to keep BANAL-52 “secret” in some kind of sinister way. Plenty of researchers have backlogs of work they haven’t published yet. Probably they a found BANAL relative in one of their normal sampling trips, did some preliminary studies on it, and planned to publish it later once they cleaned up their data. Everyone works like this. The part of DEFUSE saying that they would only work on viruses that were 95% similar to SARS is unclear and might mean something else. It looks more like they say they’ll start with those viruses, but also do some work on novel viruses. BANAL-52 could have been one of the novel viruses. The furin cleavage site is weird, but the researchers might have done that on purpose, to make the virus easier to keep track of, or to test different furin cleavage sites. Depending on the exact BANAL-52 relative they used, it might not even be a frameshift; there’s a particular way to spell serine that would make the insertion more natural. The claims that COVID can’t be cultured in normal media are based on speculative original research by Peter and might not hold up. Peter: WIV did most of its virus-gathering in a trip to a Yunnan cave between 2010 and 2015. All those viruses have long since been processed and added to the database. There’s no sign that they made more trips to Yunnan caves, and no reason for them to keep that secret. So the idea that they might just have some new viruses they didn’t publish doesn’t hold up. But suppose they did make more trips. Given the amount of time between the DEFUSE proposal and COVID, if they kept to their normal virus-collection rate, they would have gotten about thirty new viruses. What’s the chance that one of those was BANAL-52? There are thousands of bat viruses, and BANAL-52 is so rare that it wasn’t found until well after the pandemic started and people were looking for it very hard. So the chance that one of their 30 would be BANAL-52 is low. Also, they said in DEFUSE that they planned to go back to the same Yunnan cave. But BANAL-52 was found far away from that cave, so unless it ranged over a wide area, they probably couldn’t have found it even if they got very lucky. Session 3: Closing Arguments This third debate was supposed to be about “inference”, ie how much Bayesian evidence was provided by each of the facts given so far, and how to fit them into the Rootclaim probabilistic model. I’m going to relegate my summary of the more probabilistic half to the next section of this post, and just include the closing arguments here. Saar: Peter’s case hinges on the idea that it’s very improbable that a lab leak pandemic would first show up at a wet market. But this isn’t necessarily improbable. The Huanan Seafood Market had several factors that made it a likely location for a superspreader event. It was busy, with over 10,000 visitors a day. Many of the people there (eg the 1,000 vendors) came back daily, letting them reinfect each other. It had poor ventilation, especially in the high-positivity area near the raccoon-dog stall. It had cold wet surfaces on which the virus could survive for long periods. It was indoors, which prevented UV light from killing the virus. Given a small amount of sporadic COVID going around Wuhan, it’s not surprising for the first place it started spreading en masse to be a wet market. In fact, we have several examples of this. When China was COVID Zero, there would occasionally be small outbreaks that the authorities would have to contain. Most of these were at wet markets. For example, the big COVID outbreak in Beijing started at Xinfadi Market, their local seafood market. This couldn’t be an animal spillover, because there were no raccoon-dogs or other weird wildlife there. So it must be that wet markets are natural places for superspreader events. There are several other examples, which make up about half of the total outbreaks in Zero COVID era China, plus others in Singapore and Thailand. Since COVID clusters concentrate in wet markets even when there is no animal spillover, we should accept this as a property of the virus, and not attribute any significance to the fact that this happened in Wuhan too. Peter: About 1/10,000 citizens of Wuhan was a wet market vendor. So there’s a 1/10,000 chance that the first known COVID case should be a wet market vendor by chance alone. Weibo lists the most popular places for people to check in to their network on their phones, and the wet market was the 1600th most popular place in Wuhan, meaning that if you weight locations by busy-ness, there’s a less than 1/1600 chance that the first cases would be in the wet market. Yes, the wet market is indoors, has mediocre ventilation, has repeat visitors, etc. So do thousands of other places in Wuhan, like schools, hospitals, workplaces, places of worship. The wet market isn’t special in any way. And again, it wasn’t a superspreader event! COVID spread at the same rate in the wet market as it does everywhere else: doubling once per 3.5 days. It doesn’t matter what kinds of arguments you can come up with for why the wet market should have been the perfect superspreader event location, we can look at it and see that it wasn’t. It’s an environment that spreads COVID at exactly the normal rate. Zero COVID era Chinese outbreaks were concentrated in wet markets because they received infected animal products. We know why there was an outbreak in the Xinfadi Market in Beijing: it was because the seafood stall got frozen fish from some non-Zero-COVID country, the fish had COVID particles on it, and the vendor got infected and spread it to everyone else. Something like this is true for the other Chinese wet market based outbreaks we know about it. So this makes the opposite point you think it does: wet markets start outbreaks because there are infected goods being sold there. Then the virus spreads through the wet market at a completely normal rate. Saar: The Weibo list of 1600 places bigger than the wet market is likely inaccurate, because it's based on check-in data and people don't check in to seafood markets. Most of those 1600 places aren't amenable to superspread. The 70 markets supposedly bigger than Huanan are irrelevant, because they're supermarkets, open air markets, etc. Huanan is the largest seafood market in central China, and a more likely place for the first cluster of cases to be noticed. Markets weren't a common spillover location in SARS1, so the zoonosis hypothesis hasn't "called" this event in a way that should give them a high Bayes factor. And there’s still plenty of evidence for isolated (though not super-spreading) pre-market cases. A British expatriate in Wuhan, Connor Reed, says he got sick in November, three weeks before the first wet market case. Later the hospital tested his samples and said it was COVID. Another paper reports 90 cases before the first wet market one. Peter: Connor Reed was lying. The case wasn’t reported in any peer-reviewed paper. It was reported in the tabloid The Daily Mail, months after it supposedly happened. He also told the Mail that his cat died of coronavirus too, which is rare-to-impossible. Also, to get a positive hospital test, he would have had to go to the hospital, but he was 25 years old and almost no 25-year-olds go to the hospital for coronavirus. His only evidence that it was COVID was that two months later, the hospital supposedly “notified” him that it was. The hospital never informed anyone else of this extremely surprising fact which would be the biggest scientific story of the year if true. So probably he was lying. Incidentally, he died of a drug overdose shortly after giving the Mail that story; while not all drug addicts are liars, given all the other implausibilities in his story, this certainly doesn’t make him seem more credible. And in any case, he claimed he got his case at a market “like in the media” The other 90 cases are also fake. A lab leak guy found a paper that mentioned 90 more cases than other papers, and made up a conspiracy theory where the author was trying to secretly communicate that there had been 90 secret cases before any of the confirmed cases, even though there was nothing about this in the text of the paper. But actually that paper just counted cases differently than other papers, and they were referring to normal cases after the pandemic officially started. Again, I’ll come back to the discussion about inference later, but for now, here’s a table of both sides’ reasoning. This exact presentation comparing both analyses is mine3, but you can see Saar’s version here, and Peter’s starting at 45:33 of this video. Slightly made up; the two sides didn’t express their probabilities in the same way and I had to make editorial decisions to match them. Note that these aren't entirely comparable because Peter is being laxer about out-of-model probability than Saar. Although Saar's final odds here are 533-to-1, this just the central estimate. Rootclaim’s real final probability is 94% lab leak. You can see their analysis here. And The Winner Is . . . … … … … … Peter and the zoonosis hypothesis. This was a decisive victory. There were two judges, who each gave separate verdicts (or were allowed to declare a draw). Both judges decided in favor of Peter. You can see the judges’ own summary of their reasoning here (Will, Eric) Manifold agreed with the judges. There was a prediction market on who would win. It started out 70-30 in favor of lab leak. As the videos came out, zoonosis started doing better and better. I don’t want to take the exact final numbers too seriously, since I think some of the later price increases involved hints from the participants’ behavior. But it’s clear which way viewers thought the wind was blowing4. Around the same time, the Good Judgment Project - Philip Tetlock’s group studying superforecasters - put out a report on the lab leak hypothesis. After studying it in depth, his forecasters ended up 75-25 in favor of zoonosis. The Rootclaim debate was one of ten sources they said they found especially interesting. And also around the same time, and unrelated to any of this, the Global Catastrophic Risks Institute surveyed experts (“168 virologists, infectious disease epidemiologists, and other scientists from 47 countries”) and found the same thing (though see here for some potential problems with the survey): For what it’s worth, I was close to 50-50 before the debate, and now I’m 90-10 in favor of zoonosis. III. The Math And The Aftermath The third debate session was about “inference”, how to put evidence together. I put this part off until after disclosing the winner, because I wanted to talk about some of these issues at more length. The Math: Judges Both judges included a probabilistic analysis in their written decision. Here’s the same table as above, expanded to add the judges: I shoehorned the judges’ factors into the categories I already had; some of them were actually subtly different from Peter’s, Saar’s, and each other’s. The “priors” category is especially a mess here. We’ll go over these later, but I get the impression that they both thought of probabilistic analyses as an afterthought. For example, Judge Eric wrote 30,000 words about which considerations moved him, and only then includes the analysis, saying: I am not convinced that this Bayesian calculation is even an appropriate way to estimate the relative posterior probability of Z and LL; it just seemed fair that after criticizing Rootclaim’s calculations at length I should make an attempt at it myself. Judge Will’s decision ran to 10,000 words. He said he independently tried both reasoning it out intuitively, and running the Bayesian analysis, and was relieved when these two methods returned the same result. He said: I am skeptical that the Bayesian decision making/evaluation methods are any more "objective" than [intuitive reasoning]. I think they maximize legibility, not objectivity, and tend to hide the intuitive/heuristic portion in the data inclusion step and values, where it’s harder to see . . . I am not skilled in the Bayesian method, and I am sure I made significant mistakes. More time and practice would improve and refine my estimates. At the fundamental rules of the universe level, Bayesian analysis must be the best way to evaluate evidence. However, I am unsure that it’s a good strategy for a human given our cognitive limitations, and doubly unsure it’s truly being used (in the dispassionate sense) where the outcome is social desirability/fame/Twitter likes. I’m focusing on this because Saar’s opinion is that the debate went wrong (for his side) because he didn’t realize the judges were going to use Bayesian math, they did the math wrong (because Saar hadn’t done enough work explaining how to do it right), and so they got the wrong answer. I want to discuss the math errors he thinks the judges made, but this discussion would be incomplete without mentioning that the judges themselves say the numbers were only a supplement for their intuitive reasoning. That having been said, let’s look deeper into some of Saar’s concerns. The Math: Extreme Odds Saar complained that Peter’s odds were too extreme. For example, Peter said there was only a 1/10,000 chance that a lab leak pandemic would first show up at a wet market. Peter’s argument went something like: obviously a zoonotic pandemic would start at a site selling weird animals. But a lab leak pandemic - if it didn’t start at the lab - could show up anywhere. 1/10,000 Wuhan citizens work at the wet market. So if a lab leak was going to show up somewhere random, the wet market was a 1/10,000 chance. Saar had specific arguments against this, but he also had a more general argument: you should rarely see odds like 1/10,000 outside of well-understood domains. In his blog post, he gave this example: A prosecutor shows the court a statistical analysis of which DNA markers matched the defendant and their prevalence, arriving at a 1E-9 probability they would all match a random person, implying a Bayes factor near 1E9 for guilty. But if we try to estimate p(DNA|~guilty) by truly assuming innocence, it is immediately evident how ridiculous it is to claim only 1 out of a billion innocent suspects will have a DNA match to the crime scene. There are obviously far better explanations like a lab mistake, framing, an object of the suspect being brought by someone to the scene, etc. So the real p(wet market|lab leak) isn’t the 1/10,000 chance a pandemic arising in a random place hits the wet market, but the (higher?) probability that there’s something wrong with Peter’s argument. Then Saar tried to show specific things that might be wrong with Peter’s argument. I didn’t find his specific examples convincing. But maybe the question shouldn’t be whether I agreed with him. It should be whether I’m so confident he’s wrong that I would give it 10,000-to-1 odds. This makes total sense, it’s absolutely true, and I want to be really, really careful with it. If you take this kind of reasoning too far, you can convince yourself that the sun won’t rise tomorrow morning. All you have to do is propose 100 different reasons the sunrise might not happen. For example: The sun might go nova.
Inline links: 1, 2, first officially confirmed COVID case, https://substackcdn.com/image/fetch/$s_!k6BI!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9e076d66-9a5b-47c1-8b7f-2d06252571da_626x426.png, NYT, https://substackcdn.com/image/fetch/$s_!Waxw!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F7276f6a2-1021-4fe7-952a-4492447cb26f_906x635.png, https://substackcdn.com/image/fetch/$s_!M2v4!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F148d0da1-bf00-47cb-bc25-9035280588e7_1280x960.jpeg, https://substackcdn.com/image/fetch/$s_!xaM6!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fa05681e0-d6d4-42e3-aa4c-a5c6620dd5e5_617x405.png, NPR, found that, Mr. Chen, failed to find a clear connection between COVID and raccoon-dogs, https://substackcdn.com/image/fetch/$s_!8Led!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F85e46d3a-c19e-407b-88b7-1d8ea1489df7_1190x503.png, Pekar 2022, Pipes 2021, says, 3, here, this video, https://substackcdn.com/image/fetch/$s_!8aU2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F815fe32d-d7ea-401b-b3a2-d8cd25b52ee8_490x780.png, https://substackcdn.com/image/fetch/$s_!0Tm_!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd0492f69-7b7e-4611-9d76-64ef8d7f59d5_511x511.png, Will, Eric, agreed, 4, put out a report on the lab leak hypothesis, https://substackcdn.com/image/fetch/$s_!g7k2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F37f1b493-b556-41ec-925e-03f9d8bc26cb_1456x849.webp, surveyed experts, see here, https://substackcdn.com/image/fetch/$s_!Zejl!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fe9c88e87-b6ca-4c6d-840e-24da726f50b7_975x365.png, https://substackcdn.com/image/fetch/$s_!T5rV!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F4983e2cd-4151-42de-9685-08037ef7a8e8_635x788.png
Location of COVID cases in December 2020. Source: NYT, slightly edited. A map of cases at the wet market itself shows a clear pattern in favor of the very southwest corner: The southwest corner is where most of the wildlife was being sold. Rumor said that included a stall with raccoon-dogs, an animal which is generally teeming with weird coronaviruses, and is a plausible intermediate host between humans and bats: Awwww, come on, you can’t stay mad at this little guy. China said this rumor was false and refused to release any information. Scientists were finally able to confirm the existence of the raccoon-dog shop in the funniest possible way: a virologist had visited Wuhan in 2014, saw the awful conditions in the shop, and took a picture as an example of the kind of place that a future pandemic might start. Source: NPR. To be fair, we have only the scientist’s word that this is why he had the picture. But he definitely did have it. People say it would be a surprising coincidence if a zoonotic coronavirus pandemic just so happened to start in a city with a big coronavirus research lab, and this is true. But it would be an even more surprising coincidence if a lab-leak coronavirus pandemic just so happened to first get detected at a raccoon-dog stall in a wet market! Saar: It’s not clear that the first case was at the wet market; a certain Mr. Chen, with no connection to the market, seems to have fallen sick on December 8. An SCMP article suggested there were 92 previously-undetected cases suspicious for COVID as far back as November. And even if half of the first forty universally-agreed-upon cases had market connections that means another half didn’t. There was a bias towards detecting cases at the market: because authorities thought the market was the origin, and because everyone was thinking about zoonosis after SARS1, they only screened/diagnosed people with a market connection. One of the few non-market-connected COVID cases detected during this period was only detected because he was the relative of a hospital worker; the worker noticed the signs and insisted they go to the hospital despite the lack of a wet market connection. Although the map of positive samples and cases at the market was centered near the raccoon-dog stall, that could be because that area was sampled more; it’s also close to the mahjong room, where visitors and vendors at the market would go and unwind in a tight, poorly ventilated area. The next session will focus more on the WIV, but the short version is that they were doing lots of gain of function research. So one story compatible with the evidence is that a worker at WIV got infected with their modified coronavirus and passed it to his contacts. COVID started spreading quietly a few weeks to months before the first market-related case was detected. This accounts for the 92 earlier cases, Mr. Chen’s case, and the half of officially-detected cases with no wet market association. Then an infected person went to the market, causing a super-spreader event. Some of the infected market patrons went to the hospital, where doctors traced it back to the market and told other doctors to be on the lookout for wet market patrons coming in with weird viral pneumonias. They found some, declared victory, and the few anomalies - like the hospital worker’s relative - were forgotten, or assumed to have wet market connections that nobody could find. China quashed all evidence of the lab research (as was done in previous lab leak cases, eg the USSR) so all we have is the apparent wet market links that Peter found so convincing. Peter: The supposed pre-wet-market cases are confirmed fakes. Yes, the WHO did an investigation of whether there might have been COVID cases circulating before the wet market, and identified 92 unusual pneumonias that merited further review. But their final investigation, which included testing samples from these people after good tests became available, found that none of these people really had COVID. As for Mr. Chen, he said in an interview that he was hospitalized for dental issues on December 8, caught COVID in the hospital on December 16, and then was erroneously reported as “hospitalized for COVID on December 8”. The December 16 date is after the first wet market cases. Further, it seems epidemiologically impossible for COVID to have been circulating much before the first cases were officially detected December 11. The COVID pandemic doubles every 3.5 days. So if the first infection was much earlier - let’s say November 11 - we would expect 256x as much COVID as we actually saw. Even if the first couple of cases were missed because nobody was looking for them, the number of hospitalizations, deaths, etc, in January or whenever were all consistent with the number of people you’d expect if the pandemic started in early December - and not consistent with 256x that many people. So probably we should just accept that the first reported case - a wet market vendor, December 11 - was very early in the pandemic. She wasn’t literally the first case - that would most likely have been someone who worked at the raccoon-dog shop, whose case might (like 95% of COVID cases) have been mild enough not to come to medical attention. But she was certainly very early. Although authorities eventually decided COVID spread through a wet market and started deliberately looking for wet market connections, this only happened on December 30. So the earliest cases - including the 40 very earliest cases where half came from the wet market - weren’t biased (at least not through that particular route). So the claim that “the first case, and half of the first 40 cases, had wet market connections” stands as real and convincing evidence. Although the exact center of the map of positive COVID samples in the wet market was the mahjong room, the samples taken from the mahjong room were not, themselves, positive (cf: although a low-resolution population density map of New York might show Central Park in the exact center of the population density gradient, Central Park does not itself have population). There was no real “super-spreader event” at the wet market. There was a slow burn - one case the first day, a few more the next day, a few more the day after that. It’s hard to see how a single visit from an infected lab worker could do that. So the only way it could possibly be a lab leak is if the lab leaked sometime in late November, infected exactly one lab worker, that worker went straight to the wet market, infected a vendor, then went home, quarantined, recovered, and all other cases were downstream of that first infected wet market vendor. This is unparsimonious. Saar: The only source saying that Mr. Chen got sick early was an anonymous interview. And even if he was later than the first wet market cases, nobody was able to find any wet market connections. This means that whoever infected him was earlier than the index case and not linked to the wet market. Peter argued that COVID couldn’t have been more than a few weeks old when the first wet market cases were detected. But this was based on its known doubling rate. If pre-discovery COVID had a slower doubling time than known COVID, it could have been around longer. And post-lockdown serology suggested numbers that were larger than claimed at the time. So contra Peter’s claims, the infection could have been going on longer, which wouldn’t require the first lab worker to go straight to the market. It could have been weeks. Dr. Jesse Bloom’s investigation of the wet market samples, considered the final and most conclusive, failed to find a clear connection between COVID and raccoon-dogs or any other animals. Although the concentration of positive samples seemed highest near the raccoon dog stall, if you do a formal statistical analysis of which animals’ DNA was found near COVID samples most often, raccoon dogs are near the bottom. The top is wide-mouth bass, which can’t get COVID. This is obviously contamination, probably from infected humans touching wide-mouth bass tanks or something. Although the Chinese data included a negative sample from a mahjong table, it included a mention of poultry being sold nearby, which might mean this wasn’t the mahjong room itself, but some other mahjong table at a poultry shop elsewhere in the market, and (dry) mahjong tables might not hold the virus well anyway. Peter: Raccoon-dogs were sold in various cages at various stalls, separated by air gaps big enough to present a challenge for COVID transmission, and there’s no reason to think that one raccoon-dog would automatically pass it to all the others. The statistical analysis just proves there were many raccoon-dogs who didn’t have COVID. But you only need one. The raccoon dog shop and the drain leading out of the raccoon dog shop had some of the highest positive sample rates, which is more interesting than a statistical analysis which everyone agrees must be wrong (since it favors bass). It’s unclear why the negative mahjong sample says something about poultry, but based on the stated location, it’s definitely the one in the mahjong room. Session 1.5: Lineages This was technically part of Session 2, but formed enough of a discrete topic that I found it confusing to intermix it with all the other viral genetics points. I’m spinning it out into a separate summary, but the videos are all in the next session. Yuri: The coronavirus eventually mutated into many different strains. But the first big split, seen in some of the earliest samples, is between two different sub-strains called Lineage A and Lineage B, which differ by two mutations. In these two mutations, Lineage A is the same as BANAL-52, a bat virus which is the closest-known relative of COVID, but Lineage B is different. Since COVID probably evolved from something like BANAL-52, Lineage A must have come first, spread for a while, and then gotten two new mutations, turning it into Lineage B. All of the cases at the wet market, including the first detected case, were Lineage B. Lineage A wasn’t discovered until about a week later, and none of the Lineage A patients had been to the wet market. Lineage A (left) was used by the Minoan Cretans, but has never been deciphered. Lineage B (right) was used by the Mycaeneans for lists of palace goods. This matches Saar’s story above. The lab leaked to somewhere else in Wuhan, not the wet market. The virus spread undetected in the population for a while. During this time, it mutated to Lineage B. Then one of the people with Lineage B went to the wet market and started a superspreader event. The authorities sampled the patients, found Lineage B, then started looking elsewhere. Later they detected some of the earlier Lineage A cases. The market is unlikely to be the origin of the pandemic, because the original Lineage A strain wasn’t found there. Peter: Although Lineage A is evolutionarily older, Lineage B started spreading in humans first. We know this because Lineage B is more common. Throughout the early pandemic, until the D614G variant drove all other strains extinct, a consistent 2/3 of the cases were B, compared to 1/3 A. Both strains spread at the same rate, so the best explanation is that B started earlier than A. Since COVID doubles every 3-4 days, probably Lineage B started 3-4 days earlier than Lineage A, which explains why it’s always been twice as many cases. But also, Lineage B also has more internal genetic diversity than Lineage A. In general, older viruses have more genetic diversity (the “molecular clock”). This is further evidence that B started spreading first. Pekar 2022 and Pipes 2021 do analyses with known parameters for spread rate and diversity, and find 90%+ odds that Lineage B was the first one in humans. Why did the older strain start spreading later? Probably the virus crossed from bats into raccoon-dogs on some raccoon-dog farm out in the country. It spread in the raccoon-dogs for a while, racking up mutations, including the (less mutated) Lineage A strain and the (slightly more mutated) Lineage B strain. Then several raccoon-dogs were taken to Wuhan for sale, including one with Lineage A and another with Lineage B. The one with Lineage B passed its virus to humans earlier. Then 3-4 days later, the Lineage A one passed its virus to humans. Lineage A was first found in a Wuhan neighborhood right next to the wet market (closer to the wet market than 97% of Wuhan’s population). Again, it would be a bizarre coincidence if a lab leak pandemic was first detected at a wet market. But it would be an even more bizarre coincidence if a lab leak pandemic separated into two strains, and both were first detected at a wet market! Although no known wet market cases were Lineage A, a positive Lineage A environmental sample was found at the wet market, and everyone agrees most cases went undetected. So maybe the Lineage B raccoon-dog spread its virus to a vendor, and that sub-strain mostly stayed in the market. But the Lineage A raccoon-dog spread its virus to a customer, who went back to his house nearby, and that strain spread in the neighborhoods next to the market. This is the only story that explains the evolutionary precedence of A, the greater spread and older molecular clock of B, and the fact that both strains were first found very close to the wet market. Yuri/Saar: Lineage B could be more common and diverse because it got the advantage of a super-spreader event in the wet market. There are a few scattered cases of intermediates between A and B, and a few other scattered cases of lineages that seem even more ancestral (ie closer to the bat virus) than either. This doesn’t make sense in a double spillover hypothesis. But it does make sense if the lineages separated in human transmission somewhere between the lab and the first super-spreader event at the wet market. Peter: Again, the wet market wasn’t a super-spreader event. COVID spread in the wet market at exactly its normal spread rate, doubling about once every 3.5 days. Stop calling the wet market a super-spreader event. The scattered cases of “intermediates” are sequencing errors. They were all found by the same computer software, which “autofills” unsequenced bases in a genome to the most plausible guess. Because Lineage B was already in the software, depending on which part of a Lineage A virus you sequenced, you might get one half or the other autofilled as Lineage B, which looked like an “intermediate”. We know this because all the supposed “intermediates” were partial cases sequenced by this particular software. We can confirm this by noting that there are too many intermediates! That is, where Lineage A is (T/C) and Lineage B is (C/T), the software found both (T/T) “intermediates” and (C/C) “intermediates”. But obviously there can only be one real intermediate form, and we have to dismiss one or the other. But in fact we can dismiss both, because they were both caused by the same software bug. The scattered “progenitor” cases - those closer to the ancestral bat virus than either A or B - are reversions, ie cases where a new mutation in the virus happened to hit an already-mutated base and shift it back towards the ancestral virus. We know this because all of these “progenitors” were scattered cases found months after the pandemic started, often in entirely different countries from Wuhan. If these were real progenitor viruses, they would have either fizzled out or exploded into a substantial portion of all cases, not be found one time in one guy in Malaysia. Given the number of mutations the virus developed over the course of the pandemic, it’s inevitable that some of them would be mutations that bring it closer to the original bat virus, and in fact we find the number of “progenitors” found very nicely matches the number of progenitor-appearing viruses we would expect by chance. And in many cases, we know the “progenitors” are newer than the original lineages, because they also have some of the later mutations that Lineage A or B picked up along the way, alongside their apparent ancestral-bat-virus-like mutations. Session 2: Viral Genetics Yuri: Two years before COVID, scientists at the Wuhan Institute of Virology, together with colleagues at the University of North Carolina, sent in a grant proposal for the DEFUSE program. This program, intended to locate and better understand potential future pandemic viruses, involved going into bat caves and collecting new coronaviruses. Once they had them, they would do gain-of-function: specifically, they would add a furin cleavage site to make them more infectious and see what happened. (quick interlude: COVID’s spike protein has two sections: one binds to human cells through the ACE2 receptor, the other helps fuse with the cell after binding. In order to avoid the immune system, it hides both of these into one spike. But when it reaches a cell, it needs to separate them again. It takes advantage of a human respiratory enzyme, furin, to do the separation - this also ensures that it only infects its primary target, human respiratory cells. The part of COVID that lets it get separated by furin is called the “furin cleavage site”. COVID’s bat-virus ancestors were gastrointestinal viruses; the addition of a furin cleavage site was what made them respiratory viruses.) We’ve found two close relatives of COVID: bat viruses called RATG-13 and BANAL-52. In particular, COVID looks more or less like BANAL-52 plus a furin cleavage site. There are 1500 sarbecoviruses, members of the family of viruses that includes SARS and SARS2/COVID. None of them except COVID have furin cleavage sites. BANAL-52, COVID’s closest ancestor, doesn’t even have anything resembling one that could mutate into a functional furin cleavage site like COVID’s. Instead, COVID - which mostly just resembles BANAL-52 with a few scattered single-point mutations - has twelve completely new nucleotides in a row - a fully formed furin cleavage site that came out of nowhere. There is nowhere else in the genome that COVID differs from BANAL-52 in such a profound way. It’s just BANAL-52 plus a little bit of random mutation plus a fully-formed furin cleavage site that came out of nowhere. Further, the furin cleavage site is weird. It uses the protein arginine twice. But instead of the nucleotides coding for arginine in the usual viral way, both times it uses the codons CGG - the way that higher animals code for arginine. This works fine - it’s just not how viruses do it. So the obvious conclusion is that WIV, which said in 2018 that it was going to find viruses and add furin cleavage sites to them, found a close relative of BANAL-52 and added a furin cleavage site. Since they were humans, and most familiar with the human way of encoding arginine, they added it as CGG both times. COVID seemed surprisingly optimized for infecting humans. Of fifty animals it was tested in, including the usual coronavirus intermediate hosts (pangolins, raccoon-dogs, etc), it was best at infecting human cells. Further, a virus that enters a new species will usually show a burst of mutations as it “figures out” the best way to adapt to that species’ unique biology. But COVID has had a pretty constant mutation rate in humans, from the beginning of the pandemic to the end. That suggests it was already adapted to humans. This could be because the lab screened for viruses with existing adaptations, because they passed it through humanized mice in the lab, or because it adapted in the hundreds of undetected cases that happened between the lab and detection in the wet market. Usually, research with potentially dangerous coronaviruses is done in BSL-3 or 4, ie high to very-high security. But WIV was irresponsibly doing it in BSL-2, ie medium security. The researchers weren’t even required to wear masks. In general, about 1/500 labs will leak any given pathogen they’re working on (?!). But because WIV was researching such an infectious virus in such an irresponsible way, the odds of a leak were much higher. The most likely explanation for all these facts is that WIV went ahead and did the gain-of-function research they said they were going to do (the particular DEFUSE grant proposal we know about got rejected, but it proves that Wuhan wanted to do this, and they could easily have gotten funding somewhere else, or done it out of their regular budget). They found a close relative of BANAL-52 and added a furin cleavage site as a simple twelve-nucleotide insertion, using the human method of encoding arginine that their genetic engineers were familiar with. Then it leaked, spread for a while in the general Wuhan population, and eventually made it to the wet market where it got detected. Peter: As mentioned earlier, the DEFUSE grant was rejected. Further, the grant said that the Wuhan Institute of Virology was responsible for finding the viruses, and the University of North Carolina would do all the gain-of-function research. This was a reasonable division of labor, since UNC was actually good at gain-of-function research, and WIV mostly wasn’t. They had done a few very simple gain-of-function projects before, but weren’t really set up for this particular proposal and were happy to leave it for their American colleagues. Even if WIV did try to create COVID, they couldn’t have. As Yuri said, COVID looks like BANAL-52 plus a furin cleavage site. But WIV didn’t have BANAL-52. It wasn’t discovered until after the COVID pandemic started, when scientists scoured the area for potential COVID relatives. WIV had a more distant COVID relative, RATG-13. But you can’t create COVID from RATG-13; they’re too different. You would need BANAL-52, or some as-yet-undiscovered extremely close relative. WIV had neither. Are we sure they had neither? Yes. Remember, WIV’s whole job was looking for new coronaviruses. They published lists of which ones they had found pretty regularly. They published their last list in mid-2019, just a few months before the pandemic. Although lab leak proponents claimed these lists showed weird discrepancies, this was just their inability to keep names consistent, and all the lists showed basically the same viruses (plus a few extra on the later ones, as they kept discovering more). The lists didn’t include BANAL-52 or any other suitable COVID relatives - only RATG-13, which isn’t close enough to work. Could they have been keeping their discovery of BANAL-52 secret? No. Pre-pandemic, there was nothing interesting about it; our understanding of virology wasn’t good enough to point this out as a potential pandemic candidate. WIV did its gain-of-function research openly and proudly (before the pandemic, gain-of-function wasn’t as unpopular as it is now) so it’s not like they wanted to keep it secret because they might gain-of-function it later. Their lists very clearly showed they had no virus they could create COVID from, and they had no reason to hide it if they did. COVID’s furin cleavage site is admittedly unusual. But it’s unusual in a way that looks natural rather than man-made. Labs don’t usually add furin cleavage sites through nucleotide insertions (they usually mutate what’s already there). On the other hand, viruses get weird insertions of 12+ nucleotides in nature. For example, HKU1 is another emergent Chinese coronavirus that caused a small outbreak of pneumonia in 2004. It had a 15 nucleotide insertion right next to its furin cleavage site. Later strains of COVID got further 12 - 15 nucleotide insertions. Plenty of flus have 12 to 15 nucleotide insertions compared to other earlier flu strains. Sometimes insertions happen because of a mistake in viral replication. Other times the virus gets confused between its own RNA and its host’s, and splices a bit of the host RNA into the virus. This would neatly explain why the insertion used the unusual coding CGG for arginine, which is common in animals but rare in viruses. On the other hand, it’s not that rare in viruses - COVID uses CGG for arginine about 3% of the time. And human engineers don’t necessarily use it any more than that - Peter was able to find one example of humans adding arginine to a virus, and 0 out of the 5 arginines added were CGG. COVID’s furin cleavage site is a mess. When humans are inserting furin cleavage sites into viruses for gain-of-function, the standard practice is RRKR, a very nice and simple furin cleavage site which works well. COVID uses PRRAR, a bizarre furin cleavage site which no human has ever used before, and which virologists expected to work poorly. They later found that an adjacent part of COVID’s genome twisted the protein in an unusual way that allowed PRRAR to be a viable furin cleavage site, but this discovery took a lot of computer power, and was only made after COVID became important. The Wuhan virologists supposedly doing gain-of-function research on COVID shouldn’t have known this would work. Why didn’t they just use the standard RRKR site, which would have worked better? Everyone thinks it works better! Even the virus eventually decided it worked better - sometime during the course of the pandemic, it mutated away from its weird PRRAR furin cleavage site towards a more normal form. Further, COVID’s furin cleavage site was inserted via what seems to be a frameshift mutation - it wasn’t a clean insertion of the amino acids that formed the site, it was an insertion of a sequence which changed the context of the surrounding nucleotides into the amino acids that formed the site. This is a pointless too-clever-by-half “flourish” that there would be no reason for a human engineer to do. But it’s exactly the kind of weird thing that happens in the random chance of evolution. COVID is hard to culture. If you culture it in most standard media or animals, it will quickly develop characteristic mutations. But the original Wuhan strains didn’t have these mutations. The only ways to culture it without mutations are in human airway cells, or (apparently) in live raccoon-dogs. Getting human airway cells requires a donor (ie someone who donates their body to science), and Wuhan had never done this before (it was one of the technologies only used at the superior North Carolina site). As for raccoon-dogs, it sure does seems suspicious that the virus is already suited to them. The claim that COVID is uniquely adapted to humans is false. The paper that claimed that defined how well COVID was adapted to different animals by those animals’ difference (on the relevant cell receptors) from humans. So in its methodology, humans came out #1 by default. If you don’t do that, COVID is better-adapted to many other animals. It’s not necessarily true that viruses see a burst of mutations when they enter a new host. COVID spread to deer and mink, and in neither case was there a burst of mutations. COVID has a pretty simple job of infecting respiratory cells and is already very good at it, regardless of species. In Yuri’s model, Wuhan Institute of Virology picked up a discarded grant and decided to do the gain-of-function half allotted to a different university, despite their relative inexperience. They skipped over all the SARS-like viruses they were supposed to work on, and all the standard gain-of-function model backbones, in favor of BANAL-52, a virus which would not be discovered for another two years, but which they somehow had samples of, which they had for some reason decided to keep secret despite its total lack of interestingness. Then they would have had to eschew all usual gain-of-function practices in favor of inserting a weird furin cleavage site that shouldn’t have worked according to the theory they had at the time, via a frameshift mutation. Then they would have had to culture it, a technique beyond their limited capabilities. Then it would have had to leak, and magically show up again in front of the raccoon-dog stall at a wet market. Yuri: WIV wouldn’t have needed to keep BANAL-52 “secret” in some kind of sinister way. Plenty of researchers have backlogs of work they haven’t published yet. Probably they a found BANAL relative in one of their normal sampling trips, did some preliminary studies on it, and planned to publish it later once they cleaned up their data. Everyone works like this. The part of DEFUSE saying that they would only work on viruses that were 95% similar to SARS is unclear and might mean something else. It looks more like they say they’ll start with those viruses, but also do some work on novel viruses. BANAL-52 could have been one of the novel viruses. The furin cleavage site is weird, but the researchers might have done that on purpose, to make the virus easier to keep track of, or to test different furin cleavage sites. Depending on the exact BANAL-52 relative they used, it might not even be a frameshift; there’s a particular way to spell serine that would make the insertion more natural. The claims that COVID can’t be cultured in normal media are based on speculative original research by Peter and might not hold up. Peter: WIV did most of its virus-gathering in a trip to a Yunnan cave between 2010 and 2015. All those viruses have long since been processed and added to the database. There’s no sign that they made more trips to Yunnan caves, and no reason for them to keep that secret. So the idea that they might just have some new viruses they didn’t publish doesn’t hold up. But suppose they did make more trips. Given the amount of time between the DEFUSE proposal and COVID, if they kept to their normal virus-collection rate, they would have gotten about thirty new viruses. What’s the chance that one of those was BANAL-52? There are thousands of bat viruses, and BANAL-52 is so rare that it wasn’t found until well after the pandemic started and people were looking for it very hard. So the chance that one of their 30 would be BANAL-52 is low. Also, they said in DEFUSE that they planned to go back to the same Yunnan cave. But BANAL-52 was found far away from that cave, so unless it ranged over a wide area, they probably couldn’t have found it even if they got very lucky. Session 3: Closing Arguments This third debate was supposed to be about “inference”, ie how much Bayesian evidence was provided by each of the facts given so far, and how to fit them into the Rootclaim probabilistic model. I’m going to relegate my summary of the more probabilistic half to the next section of this post, and just include the closing arguments here. Saar: Peter’s case hinges on the idea that it’s very improbable that a lab leak pandemic would first show up at a wet market. But this isn’t necessarily improbable. The Huanan Seafood Market had several factors that made it a likely location for a superspreader event. It was busy, with over 10,000 visitors a day. Many of the people there (eg the 1,000 vendors) came back daily, letting them reinfect each other. It had poor ventilation, especially in the high-positivity area near the raccoon-dog stall. It had cold wet surfaces on which the virus could survive for long periods. It was indoors, which prevented UV light from killing the virus. Given a small amount of sporadic COVID going around Wuhan, it’s not surprising for the first place it started spreading en masse to be a wet market. In fact, we have several examples of this. When China was COVID Zero, there would occasionally be small outbreaks that the authorities would have to contain. Most of these were at wet markets. For example, the big COVID outbreak in Beijing started at Xinfadi Market, their local seafood market. This couldn’t be an animal spillover, because there were no raccoon-dogs or other weird wildlife there. So it must be that wet markets are natural places for superspreader events. There are several other examples, which make up about half of the total outbreaks in Zero COVID era China, plus others in Singapore and Thailand. Since COVID clusters concentrate in wet markets even when there is no animal spillover, we should accept this as a property of the virus, and not attribute any significance to the fact that this happened in Wuhan too. Peter: About 1/10,000 citizens of Wuhan was a wet market vendor. So there’s a 1/10,000 chance that the first known COVID case should be a wet market vendor by chance alone. Weibo lists the most popular places for people to check in to their network on their phones, and the wet market was the 1600th most popular place in Wuhan, meaning that if you weight locations by busy-ness, there’s a less than 1/1600 chance that the first cases would be in the wet market. Yes, the wet market is indoors, has mediocre ventilation, has repeat visitors, etc. So do thousands of other places in Wuhan, like schools, hospitals, workplaces, places of worship. The wet market isn’t special in any way. And again, it wasn’t a superspreader event! COVID spread at the same rate in the wet market as it does everywhere else: doubling once per 3.5 days. It doesn’t matter what kinds of arguments you can come up with for why the wet market should have been the perfect superspreader event location, we can look at it and see that it wasn’t. It’s an environment that spreads COVID at exactly the normal rate. Zero COVID era Chinese outbreaks were concentrated in wet markets because they received infected animal products. We know why there was an outbreak in the Xinfadi Market in Beijing: it was because the seafood stall got frozen fish from some non-Zero-COVID country, the fish had COVID particles on it, and the vendor got infected and spread it to everyone else. Something like this is true for the other Chinese wet market based outbreaks we know about it. So this makes the opposite point you think it does: wet markets start outbreaks because there are infected goods being sold there. Then the virus spreads through the wet market at a completely normal rate. Saar: The Weibo list of 1600 places bigger than the wet market is likely inaccurate, because it's based on check-in data and people don't check in to seafood markets. Most of those 1600 places aren't amenable to superspread. The 70 markets supposedly bigger than Huanan are irrelevant, because they're supermarkets, open air markets, etc. Huanan is the largest seafood market in central China, and a more likely place for the first cluster of cases to be noticed. Markets weren't a common spillover location in SARS1, so the zoonosis hypothesis hasn't "called" this event in a way that should give them a high Bayes factor. And there’s still plenty of evidence for isolated (though not super-spreading) pre-market cases. A British expatriate in Wuhan, Connor Reed, says he got sick in November, three weeks before the first wet market case. Later the hospital tested his samples and said it was COVID. Another paper reports 90 cases before the first wet market one. Peter: Connor Reed was lying. The case wasn’t reported in any peer-reviewed paper. It was reported in the tabloid The Daily Mail, months after it supposedly happened. He also told the Mail that his cat died of coronavirus too, which is rare-to-impossible. Also, to get a positive hospital test, he would have had to go to the hospital, but he was 25 years old and almost no 25-year-olds go to the hospital for coronavirus. His only evidence that it was COVID was that two months later, the hospital supposedly “notified” him that it was. The hospital never informed anyone else of this extremely surprising fact which would be the biggest scientific story of the year if true. So probably he was lying. Incidentally, he died of a drug overdose shortly after giving the Mail that story; while not all drug addicts are liars, given all the other implausibilities in his story, this certainly doesn’t make him seem more credible. And in any case, he claimed he got his case at a market “like in the media” The other 90 cases are also fake. A lab leak guy found a paper that mentioned 90 more cases than other papers, and made up a conspiracy theory where the author was trying to secretly communicate that there had been 90 secret cases before any of the confirmed cases, even though there was nothing about this in the text of the paper. But actually that paper just counted cases differently than other papers, and they were referring to normal cases after the pandemic officially started. Again, I’ll come back to the discussion about inference later, but for now, here’s a table of both sides’ reasoning. This exact presentation comparing both analyses is mine3, but you can see Saar’s version here, and Peter’s starting at 45:33 of this video. Slightly made up; the two sides didn’t express their probabilities in the same way and I had to make editorial decisions to match them. Note that these aren't entirely comparable because Peter is being laxer about out-of-model probability than Saar. Although Saar's final odds here are 533-to-1, this just the central estimate. Rootclaim’s real final probability is 94% lab leak. You can see their analysis here. And The Winner Is . . . … … … … … Peter and the zoonosis hypothesis. This was a decisive victory. There were two judges, who each gave separate verdicts (or were allowed to declare a draw). Both judges decided in favor of Peter. You can see the judges’ own summary of their reasoning here (Will, Eric) Manifold agreed with the judges. There was a prediction market on who would win. It started out 70-30 in favor of lab leak. As the videos came out, zoonosis started doing better and better. I don’t want to take the exact final numbers too seriously, since I think some of the later price increases involved hints from the participants’ behavior. But it’s clear which way viewers thought the wind was blowing4. Around the same time, the Good Judgment Project - Philip Tetlock’s group studying superforecasters - put out a report on the lab leak hypothesis. After studying it in depth, his forecasters ended up 75-25 in favor of zoonosis. The Rootclaim debate was one of ten sources they said they found especially interesting. And also around the same time, and unrelated to any of this, the Global Catastrophic Risks Institute surveyed experts (“168 virologists, infectious disease epidemiologists, and other scientists from 47 countries”) and found the same thing (though see here for some potential problems with the survey): For what it’s worth, I was close to 50-50 before the debate, and now I’m 90-10 in favor of zoonosis. III. The Math And The Aftermath The third debate session was about “inference”, how to put evidence together. I put this part off until after disclosing the winner, because I wanted to talk about some of these issues at more length. The Math: Judges Both judges included a probabilistic analysis in their written decision. Here’s the same table as above, expanded to add the judges: I shoehorned the judges’ factors into the categories I already had; some of them were actually subtly different from Peter’s, Saar’s, and each other’s. The “priors” category is especially a mess here. We’ll go over these later, but I get the impression that they both thought of probabilistic analyses as an afterthought. For example, Judge Eric wrote 30,000 words about which considerations moved him, and only then includes the analysis, saying: I am not convinced that this Bayesian calculation is even an appropriate way to estimate the relative posterior probability of Z and LL; it just seemed fair that after criticizing Rootclaim’s calculations at length I should make an attempt at it myself. Judge Will’s decision ran to 10,000 words. He said he independently tried both reasoning it out intuitively, and running the Bayesian analysis, and was relieved when these two methods returned the same result. He said: I am skeptical that the Bayesian decision making/evaluation methods are any more "objective" than [intuitive reasoning]. I think they maximize legibility, not objectivity, and tend to hide the intuitive/heuristic portion in the data inclusion step and values, where it’s harder to see . . . I am not skilled in the Bayesian method, and I am sure I made significant mistakes. More time and practice would improve and refine my estimates. At the fundamental rules of the universe level, Bayesian analysis must be the best way to evaluate evidence. However, I am unsure that it’s a good strategy for a human given our cognitive limitations, and doubly unsure it’s truly being used (in the dispassionate sense) where the outcome is social desirability/fame/Twitter likes. I’m focusing on this because Saar’s opinion is that the debate went wrong (for his side) because he didn’t realize the judges were going to use Bayesian math, they did the math wrong (because Saar hadn’t done enough work explaining how to do it right), and so they got the wrong answer. I want to discuss the math errors he thinks the judges made, but this discussion would be incomplete without mentioning that the judges themselves say the numbers were only a supplement for their intuitive reasoning. That having been said, let’s look deeper into some of Saar’s concerns. The Math: Extreme Odds Saar complained that Peter’s odds were too extreme. For example, Peter said there was only a 1/10,000 chance that a lab leak pandemic would first show up at a wet market. Peter’s argument went something like: obviously a zoonotic pandemic would start at a site selling weird animals. But a lab leak pandemic - if it didn’t start at the lab - could show up anywhere. 1/10,000 Wuhan citizens work at the wet market. So if a lab leak was going to show up somewhere random, the wet market was a 1/10,000 chance. Saar had specific arguments against this, but he also had a more general argument: you should rarely see odds like 1/10,000 outside of well-understood domains. In his blog post, he gave this example: A prosecutor shows the court a statistical analysis of which DNA markers matched the defendant and their prevalence, arriving at a 1E-9 probability they would all match a random person, implying a Bayes factor near 1E9 for guilty. But if we try to estimate p(DNA|~guilty) by truly assuming innocence, it is immediately evident how ridiculous it is to claim only 1 out of a billion innocent suspects will have a DNA match to the crime scene. There are obviously far better explanations like a lab mistake, framing, an object of the suspect being brought by someone to the scene, etc. So the real p(wet market|lab leak) isn’t the 1/10,000 chance a pandemic arising in a random place hits the wet market, but the (higher?) probability that there’s something wrong with Peter’s argument. Then Saar tried to show specific things that might be wrong with Peter’s argument. I didn’t find his specific examples convincing. But maybe the question shouldn’t be whether I agreed with him. It should be whether I’m so confident he’s wrong that I would give it 10,000-to-1 odds. This makes total sense, it’s absolutely true, and I want to be really, really careful with it. If you take this kind of reasoning too far, you can convince yourself that the sun won’t rise tomorrow morning. All you have to do is propose 100 different reasons the sunrise might not happen. For example: The sun might go nova.
Inline links: NYT, https://substackcdn.com/image/fetch/$s_!Waxw!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F7276f6a2-1021-4fe7-952a-4492447cb26f_906x635.png, https://substackcdn.com/image/fetch/$s_!M2v4!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F148d0da1-bf00-47cb-bc25-9035280588e7_1280x960.jpeg, https://substackcdn.com/image/fetch/$s_!xaM6!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fa05681e0-d6d4-42e3-aa4c-a5c6620dd5e5_617x405.png, NPR, found that, Mr. Chen, failed to find a clear connection between COVID and raccoon-dogs, https://substackcdn.com/image/fetch/$s_!8Led!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F85e46d3a-c19e-407b-88b7-1d8ea1489df7_1190x503.png, Pekar 2022, Pipes 2021, says, 3, here, this video, https://substackcdn.com/image/fetch/$s_!8aU2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F815fe32d-d7ea-401b-b3a2-d8cd25b52ee8_490x780.png, https://substackcdn.com/image/fetch/$s_!0Tm_!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd0492f69-7b7e-4611-9d76-64ef8d7f59d5_511x511.png, Will, Eric, agreed, 4, put out a report on the lab leak hypothesis, https://substackcdn.com/image/fetch/$s_!g7k2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F37f1b493-b556-41ec-925e-03f9d8bc26cb_1456x849.webp, surveyed experts, see here, https://substackcdn.com/image/fetch/$s_!Zejl!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fe9c88e87-b6ca-4c6d-840e-24da726f50b7_975x365.png, https://substackcdn.com/image/fetch/$s_!T5rV!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F4983e2cd-4151-42de-9685-08037ef7a8e8_635x788.png
Awwww, come on, you can’t stay mad at this little guy. China said this rumor was false and refused to release any information. Scientists were finally able to confirm the existence of the raccoon-dog shop in the funniest possible way: a virologist had visited Wuhan in 2014, saw the awful conditions in the shop, and took a picture as an example of the kind of place that a future pandemic might start. Source: NPR. To be fair, we have only the scientist’s word that this is why he had the picture. But he definitely did have it. People say it would be a surprising coincidence if a zoonotic coronavirus pandemic just so happened to start in a city with a big coronavirus research lab, and this is true. But it would be an even more surprising coincidence if a lab-leak coronavirus pandemic just so happened to first get detected at a raccoon-dog stall in a wet market! Saar: It’s not clear that the first case was at the wet market; a certain Mr. Chen, with no connection to the market, seems to have fallen sick on December 8. An SCMP article suggested there were 92 previously-undetected cases suspicious for COVID as far back as November. And even if half of the first forty universally-agreed-upon cases had market connections that means another half didn’t. There was a bias towards detecting cases at the market: because authorities thought the market was the origin, and because everyone was thinking about zoonosis after SARS1, they only screened/diagnosed people with a market connection. One of the few non-market-connected COVID cases detected during this period was only detected because he was the relative of a hospital worker; the worker noticed the signs and insisted they go to the hospital despite the lack of a wet market connection. Although the map of positive samples and cases at the market was centered near the raccoon-dog stall, that could be because that area was sampled more; it’s also close to the mahjong room, where visitors and vendors at the market would go and unwind in a tight, poorly ventilated area. The next session will focus more on the WIV, but the short version is that they were doing lots of gain of function research. So one story compatible with the evidence is that a worker at WIV got infected with their modified coronavirus and passed it to his contacts. COVID started spreading quietly a few weeks to months before the first market-related case was detected. This accounts for the 92 earlier cases, Mr. Chen’s case, and the half of officially-detected cases with no wet market association. Then an infected person went to the market, causing a super-spreader event. Some of the infected market patrons went to the hospital, where doctors traced it back to the market and told other doctors to be on the lookout for wet market patrons coming in with weird viral pneumonias. They found some, declared victory, and the few anomalies - like the hospital worker’s relative - were forgotten, or assumed to have wet market connections that nobody could find. China quashed all evidence of the lab research (as was done in previous lab leak cases, eg the USSR) so all we have is the apparent wet market links that Peter found so convincing. Peter: The supposed pre-wet-market cases are confirmed fakes. Yes, the WHO did an investigation of whether there might have been COVID cases circulating before the wet market, and identified 92 unusual pneumonias that merited further review. But their final investigation, which included testing samples from these people after good tests became available, found that none of these people really had COVID. As for Mr. Chen, he said in an interview that he was hospitalized for dental issues on December 8, caught COVID in the hospital on December 16, and then was erroneously reported as “hospitalized for COVID on December 8”. The December 16 date is after the first wet market cases. Further, it seems epidemiologically impossible for COVID to have been circulating much before the first cases were officially detected December 11. The COVID pandemic doubles every 3.5 days. So if the first infection was much earlier - let’s say November 11 - we would expect 256x as much COVID as we actually saw. Even if the first couple of cases were missed because nobody was looking for them, the number of hospitalizations, deaths, etc, in January or whenever were all consistent with the number of people you’d expect if the pandemic started in early December - and not consistent with 256x that many people. So probably we should just accept that the first reported case - a wet market vendor, December 11 - was very early in the pandemic. She wasn’t literally the first case - that would most likely have been someone who worked at the raccoon-dog shop, whose case might (like 95% of COVID cases) have been mild enough not to come to medical attention. But she was certainly very early. Although authorities eventually decided COVID spread through a wet market and started deliberately looking for wet market connections, this only happened on December 30. So the earliest cases - including the 40 very earliest cases where half came from the wet market - weren’t biased (at least not through that particular route). So the claim that “the first case, and half of the first 40 cases, had wet market connections” stands as real and convincing evidence. Although the exact center of the map of positive COVID samples in the wet market was the mahjong room, the samples taken from the mahjong room were not, themselves, positive (cf: although a low-resolution population density map of New York might show Central Park in the exact center of the population density gradient, Central Park does not itself have population). There was no real “super-spreader event” at the wet market. There was a slow burn - one case the first day, a few more the next day, a few more the day after that. It’s hard to see how a single visit from an infected lab worker could do that. So the only way it could possibly be a lab leak is if the lab leaked sometime in late November, infected exactly one lab worker, that worker went straight to the wet market, infected a vendor, then went home, quarantined, recovered, and all other cases were downstream of that first infected wet market vendor. This is unparsimonious. Saar: The only source saying that Mr. Chen got sick early was an anonymous interview. And even if he was later than the first wet market cases, nobody was able to find any wet market connections. This means that whoever infected him was earlier than the index case and not linked to the wet market. Peter argued that COVID couldn’t have been more than a few weeks old when the first wet market cases were detected. But this was based on its known doubling rate. If pre-discovery COVID had a slower doubling time than known COVID, it could have been around longer. And post-lockdown serology suggested numbers that were larger than claimed at the time. So contra Peter’s claims, the infection could have been going on longer, which wouldn’t require the first lab worker to go straight to the market. It could have been weeks. Dr. Jesse Bloom’s investigation of the wet market samples, considered the final and most conclusive, failed to find a clear connection between COVID and raccoon-dogs or any other animals. Although the concentration of positive samples seemed highest near the raccoon dog stall, if you do a formal statistical analysis of which animals’ DNA was found near COVID samples most often, raccoon dogs are near the bottom. The top is wide-mouth bass, which can’t get COVID. This is obviously contamination, probably from infected humans touching wide-mouth bass tanks or something. Although the Chinese data included a negative sample from a mahjong table, it included a mention of poultry being sold nearby, which might mean this wasn’t the mahjong room itself, but some other mahjong table at a poultry shop elsewhere in the market, and (dry) mahjong tables might not hold the virus well anyway. Peter: Raccoon-dogs were sold in various cages at various stalls, separated by air gaps big enough to present a challenge for COVID transmission, and there’s no reason to think that one raccoon-dog would automatically pass it to all the others. The statistical analysis just proves there were many raccoon-dogs who didn’t have COVID. But you only need one. The raccoon dog shop and the drain leading out of the raccoon dog shop had some of the highest positive sample rates, which is more interesting than a statistical analysis which everyone agrees must be wrong (since it favors bass). It’s unclear why the negative mahjong sample says something about poultry, but based on the stated location, it’s definitely the one in the mahjong room. Session 1.5: Lineages This was technically part of Session 2, but formed enough of a discrete topic that I found it confusing to intermix it with all the other viral genetics points. I’m spinning it out into a separate summary, but the videos are all in the next session. Yuri: The coronavirus eventually mutated into many different strains. But the first big split, seen in some of the earliest samples, is between two different sub-strains called Lineage A and Lineage B, which differ by two mutations. In these two mutations, Lineage A is the same as BANAL-52, a bat virus which is the closest-known relative of COVID, but Lineage B is different. Since COVID probably evolved from something like BANAL-52, Lineage A must have come first, spread for a while, and then gotten two new mutations, turning it into Lineage B. All of the cases at the wet market, including the first detected case, were Lineage B. Lineage A wasn’t discovered until about a week later, and none of the Lineage A patients had been to the wet market. Lineage A (left) was used by the Minoan Cretans, but has never been deciphered. Lineage B (right) was used by the Mycaeneans for lists of palace goods. This matches Saar’s story above. The lab leaked to somewhere else in Wuhan, not the wet market. The virus spread undetected in the population for a while. During this time, it mutated to Lineage B. Then one of the people with Lineage B went to the wet market and started a superspreader event. The authorities sampled the patients, found Lineage B, then started looking elsewhere. Later they detected some of the earlier Lineage A cases. The market is unlikely to be the origin of the pandemic, because the original Lineage A strain wasn’t found there. Peter: Although Lineage A is evolutionarily older, Lineage B started spreading in humans first. We know this because Lineage B is more common. Throughout the early pandemic, until the D614G variant drove all other strains extinct, a consistent 2/3 of the cases were B, compared to 1/3 A. Both strains spread at the same rate, so the best explanation is that B started earlier than A. Since COVID doubles every 3-4 days, probably Lineage B started 3-4 days earlier than Lineage A, which explains why it’s always been twice as many cases. But also, Lineage B also has more internal genetic diversity than Lineage A. In general, older viruses have more genetic diversity (the “molecular clock”). This is further evidence that B started spreading first. Pekar 2022 and Pipes 2021 do analyses with known parameters for spread rate and diversity, and find 90%+ odds that Lineage B was the first one in humans. Why did the older strain start spreading later? Probably the virus crossed from bats into raccoon-dogs on some raccoon-dog farm out in the country. It spread in the raccoon-dogs for a while, racking up mutations, including the (less mutated) Lineage A strain and the (slightly more mutated) Lineage B strain. Then several raccoon-dogs were taken to Wuhan for sale, including one with Lineage A and another with Lineage B. The one with Lineage B passed its virus to humans earlier. Then 3-4 days later, the Lineage A one passed its virus to humans. Lineage A was first found in a Wuhan neighborhood right next to the wet market (closer to the wet market than 97% of Wuhan’s population). Again, it would be a bizarre coincidence if a lab leak pandemic was first detected at a wet market. But it would be an even more bizarre coincidence if a lab leak pandemic separated into two strains, and both were first detected at a wet market! Although no known wet market cases were Lineage A, a positive Lineage A environmental sample was found at the wet market, and everyone agrees most cases went undetected. So maybe the Lineage B raccoon-dog spread its virus to a vendor, and that sub-strain mostly stayed in the market. But the Lineage A raccoon-dog spread its virus to a customer, who went back to his house nearby, and that strain spread in the neighborhoods next to the market. This is the only story that explains the evolutionary precedence of A, the greater spread and older molecular clock of B, and the fact that both strains were first found very close to the wet market. Yuri/Saar: Lineage B could be more common and diverse because it got the advantage of a super-spreader event in the wet market. There are a few scattered cases of intermediates between A and B, and a few other scattered cases of lineages that seem even more ancestral (ie closer to the bat virus) than either. This doesn’t make sense in a double spillover hypothesis. But it does make sense if the lineages separated in human transmission somewhere between the lab and the first super-spreader event at the wet market. Peter: Again, the wet market wasn’t a super-spreader event. COVID spread in the wet market at exactly its normal spread rate, doubling about once every 3.5 days. Stop calling the wet market a super-spreader event. The scattered cases of “intermediates” are sequencing errors. They were all found by the same computer software, which “autofills” unsequenced bases in a genome to the most plausible guess. Because Lineage B was already in the software, depending on which part of a Lineage A virus you sequenced, you might get one half or the other autofilled as Lineage B, which looked like an “intermediate”. We know this because all the supposed “intermediates” were partial cases sequenced by this particular software. We can confirm this by noting that there are too many intermediates! That is, where Lineage A is (T/C) and Lineage B is (C/T), the software found both (T/T) “intermediates” and (C/C) “intermediates”. But obviously there can only be one real intermediate form, and we have to dismiss one or the other. But in fact we can dismiss both, because they were both caused by the same software bug. The scattered “progenitor” cases - those closer to the ancestral bat virus than either A or B - are reversions, ie cases where a new mutation in the virus happened to hit an already-mutated base and shift it back towards the ancestral virus. We know this because all of these “progenitors” were scattered cases found months after the pandemic started, often in entirely different countries from Wuhan. If these were real progenitor viruses, they would have either fizzled out or exploded into a substantial portion of all cases, not be found one time in one guy in Malaysia. Given the number of mutations the virus developed over the course of the pandemic, it’s inevitable that some of them would be mutations that bring it closer to the original bat virus, and in fact we find the number of “progenitors” found very nicely matches the number of progenitor-appearing viruses we would expect by chance. And in many cases, we know the “progenitors” are newer than the original lineages, because they also have some of the later mutations that Lineage A or B picked up along the way, alongside their apparent ancestral-bat-virus-like mutations. Session 2: Viral Genetics Yuri: Two years before COVID, scientists at the Wuhan Institute of Virology, together with colleagues at the University of North Carolina, sent in a grant proposal for the DEFUSE program. This program, intended to locate and better understand potential future pandemic viruses, involved going into bat caves and collecting new coronaviruses. Once they had them, they would do gain-of-function: specifically, they would add a furin cleavage site to make them more infectious and see what happened. (quick interlude: COVID’s spike protein has two sections: one binds to human cells through the ACE2 receptor, the other helps fuse with the cell after binding. In order to avoid the immune system, it hides both of these into one spike. But when it reaches a cell, it needs to separate them again. It takes advantage of a human respiratory enzyme, furin, to do the separation - this also ensures that it only infects its primary target, human respiratory cells. The part of COVID that lets it get separated by furin is called the “furin cleavage site”. COVID’s bat-virus ancestors were gastrointestinal viruses; the addition of a furin cleavage site was what made them respiratory viruses.) We’ve found two close relatives of COVID: bat viruses called RATG-13 and BANAL-52. In particular, COVID looks more or less like BANAL-52 plus a furin cleavage site. There are 1500 sarbecoviruses, members of the family of viruses that includes SARS and SARS2/COVID. None of them except COVID have furin cleavage sites. BANAL-52, COVID’s closest ancestor, doesn’t even have anything resembling one that could mutate into a functional furin cleavage site like COVID’s. Instead, COVID - which mostly just resembles BANAL-52 with a few scattered single-point mutations - has twelve completely new nucleotides in a row - a fully formed furin cleavage site that came out of nowhere. There is nowhere else in the genome that COVID differs from BANAL-52 in such a profound way. It’s just BANAL-52 plus a little bit of random mutation plus a fully-formed furin cleavage site that came out of nowhere. Further, the furin cleavage site is weird. It uses the protein arginine twice. But instead of the nucleotides coding for arginine in the usual viral way, both times it uses the codons CGG - the way that higher animals code for arginine. This works fine - it’s just not how viruses do it. So the obvious conclusion is that WIV, which said in 2018 that it was going to find viruses and add furin cleavage sites to them, found a close relative of BANAL-52 and added a furin cleavage site. Since they were humans, and most familiar with the human way of encoding arginine, they added it as CGG both times. COVID seemed surprisingly optimized for infecting humans. Of fifty animals it was tested in, including the usual coronavirus intermediate hosts (pangolins, raccoon-dogs, etc), it was best at infecting human cells. Further, a virus that enters a new species will usually show a burst of mutations as it “figures out” the best way to adapt to that species’ unique biology. But COVID has had a pretty constant mutation rate in humans, from the beginning of the pandemic to the end. That suggests it was already adapted to humans. This could be because the lab screened for viruses with existing adaptations, because they passed it through humanized mice in the lab, or because it adapted in the hundreds of undetected cases that happened between the lab and detection in the wet market. Usually, research with potentially dangerous coronaviruses is done in BSL-3 or 4, ie high to very-high security. But WIV was irresponsibly doing it in BSL-2, ie medium security. The researchers weren’t even required to wear masks. In general, about 1/500 labs will leak any given pathogen they’re working on (?!). But because WIV was researching such an infectious virus in such an irresponsible way, the odds of a leak were much higher. The most likely explanation for all these facts is that WIV went ahead and did the gain-of-function research they said they were going to do (the particular DEFUSE grant proposal we know about got rejected, but it proves that Wuhan wanted to do this, and they could easily have gotten funding somewhere else, or done it out of their regular budget). They found a close relative of BANAL-52 and added a furin cleavage site as a simple twelve-nucleotide insertion, using the human method of encoding arginine that their genetic engineers were familiar with. Then it leaked, spread for a while in the general Wuhan population, and eventually made it to the wet market where it got detected. Peter: As mentioned earlier, the DEFUSE grant was rejected. Further, the grant said that the Wuhan Institute of Virology was responsible for finding the viruses, and the University of North Carolina would do all the gain-of-function research. This was a reasonable division of labor, since UNC was actually good at gain-of-function research, and WIV mostly wasn’t. They had done a few very simple gain-of-function projects before, but weren’t really set up for this particular proposal and were happy to leave it for their American colleagues. Even if WIV did try to create COVID, they couldn’t have. As Yuri said, COVID looks like BANAL-52 plus a furin cleavage site. But WIV didn’t have BANAL-52. It wasn’t discovered until after the COVID pandemic started, when scientists scoured the area for potential COVID relatives. WIV had a more distant COVID relative, RATG-13. But you can’t create COVID from RATG-13; they’re too different. You would need BANAL-52, or some as-yet-undiscovered extremely close relative. WIV had neither. Are we sure they had neither? Yes. Remember, WIV’s whole job was looking for new coronaviruses. They published lists of which ones they had found pretty regularly. They published their last list in mid-2019, just a few months before the pandemic. Although lab leak proponents claimed these lists showed weird discrepancies, this was just their inability to keep names consistent, and all the lists showed basically the same viruses (plus a few extra on the later ones, as they kept discovering more). The lists didn’t include BANAL-52 or any other suitable COVID relatives - only RATG-13, which isn’t close enough to work. Could they have been keeping their discovery of BANAL-52 secret? No. Pre-pandemic, there was nothing interesting about it; our understanding of virology wasn’t good enough to point this out as a potential pandemic candidate. WIV did its gain-of-function research openly and proudly (before the pandemic, gain-of-function wasn’t as unpopular as it is now) so it’s not like they wanted to keep it secret because they might gain-of-function it later. Their lists very clearly showed they had no virus they could create COVID from, and they had no reason to hide it if they did. COVID’s furin cleavage site is admittedly unusual. But it’s unusual in a way that looks natural rather than man-made. Labs don’t usually add furin cleavage sites through nucleotide insertions (they usually mutate what’s already there). On the other hand, viruses get weird insertions of 12+ nucleotides in nature. For example, HKU1 is another emergent Chinese coronavirus that caused a small outbreak of pneumonia in 2004. It had a 15 nucleotide insertion right next to its furin cleavage site. Later strains of COVID got further 12 - 15 nucleotide insertions. Plenty of flus have 12 to 15 nucleotide insertions compared to other earlier flu strains. Sometimes insertions happen because of a mistake in viral replication. Other times the virus gets confused between its own RNA and its host’s, and splices a bit of the host RNA into the virus. This would neatly explain why the insertion used the unusual coding CGG for arginine, which is common in animals but rare in viruses. On the other hand, it’s not that rare in viruses - COVID uses CGG for arginine about 3% of the time. And human engineers don’t necessarily use it any more than that - Peter was able to find one example of humans adding arginine to a virus, and 0 out of the 5 arginines added were CGG. COVID’s furin cleavage site is a mess. When humans are inserting furin cleavage sites into viruses for gain-of-function, the standard practice is RRKR, a very nice and simple furin cleavage site which works well. COVID uses PRRAR, a bizarre furin cleavage site which no human has ever used before, and which virologists expected to work poorly. They later found that an adjacent part of COVID’s genome twisted the protein in an unusual way that allowed PRRAR to be a viable furin cleavage site, but this discovery took a lot of computer power, and was only made after COVID became important. The Wuhan virologists supposedly doing gain-of-function research on COVID shouldn’t have known this would work. Why didn’t they just use the standard RRKR site, which would have worked better? Everyone thinks it works better! Even the virus eventually decided it worked better - sometime during the course of the pandemic, it mutated away from its weird PRRAR furin cleavage site towards a more normal form. Further, COVID’s furin cleavage site was inserted via what seems to be a frameshift mutation - it wasn’t a clean insertion of the amino acids that formed the site, it was an insertion of a sequence which changed the context of the surrounding nucleotides into the amino acids that formed the site. This is a pointless too-clever-by-half “flourish” that there would be no reason for a human engineer to do. But it’s exactly the kind of weird thing that happens in the random chance of evolution. COVID is hard to culture. If you culture it in most standard media or animals, it will quickly develop characteristic mutations. But the original Wuhan strains didn’t have these mutations. The only ways to culture it without mutations are in human airway cells, or (apparently) in live raccoon-dogs. Getting human airway cells requires a donor (ie someone who donates their body to science), and Wuhan had never done this before (it was one of the technologies only used at the superior North Carolina site). As for raccoon-dogs, it sure does seems suspicious that the virus is already suited to them. The claim that COVID is uniquely adapted to humans is false. The paper that claimed that defined how well COVID was adapted to different animals by those animals’ difference (on the relevant cell receptors) from humans. So in its methodology, humans came out #1 by default. If you don’t do that, COVID is better-adapted to many other animals. It’s not necessarily true that viruses see a burst of mutations when they enter a new host. COVID spread to deer and mink, and in neither case was there a burst of mutations. COVID has a pretty simple job of infecting respiratory cells and is already very good at it, regardless of species. In Yuri’s model, Wuhan Institute of Virology picked up a discarded grant and decided to do the gain-of-function half allotted to a different university, despite their relative inexperience. They skipped over all the SARS-like viruses they were supposed to work on, and all the standard gain-of-function model backbones, in favor of BANAL-52, a virus which would not be discovered for another two years, but which they somehow had samples of, which they had for some reason decided to keep secret despite its total lack of interestingness. Then they would have had to eschew all usual gain-of-function practices in favor of inserting a weird furin cleavage site that shouldn’t have worked according to the theory they had at the time, via a frameshift mutation. Then they would have had to culture it, a technique beyond their limited capabilities. Then it would have had to leak, and magically show up again in front of the raccoon-dog stall at a wet market. Yuri: WIV wouldn’t have needed to keep BANAL-52 “secret” in some kind of sinister way. Plenty of researchers have backlogs of work they haven’t published yet. Probably they a found BANAL relative in one of their normal sampling trips, did some preliminary studies on it, and planned to publish it later once they cleaned up their data. Everyone works like this. The part of DEFUSE saying that they would only work on viruses that were 95% similar to SARS is unclear and might mean something else. It looks more like they say they’ll start with those viruses, but also do some work on novel viruses. BANAL-52 could have been one of the novel viruses. The furin cleavage site is weird, but the researchers might have done that on purpose, to make the virus easier to keep track of, or to test different furin cleavage sites. Depending on the exact BANAL-52 relative they used, it might not even be a frameshift; there’s a particular way to spell serine that would make the insertion more natural. The claims that COVID can’t be cultured in normal media are based on speculative original research by Peter and might not hold up. Peter: WIV did most of its virus-gathering in a trip to a Yunnan cave between 2010 and 2015. All those viruses have long since been processed and added to the database. There’s no sign that they made more trips to Yunnan caves, and no reason for them to keep that secret. So the idea that they might just have some new viruses they didn’t publish doesn’t hold up. But suppose they did make more trips. Given the amount of time between the DEFUSE proposal and COVID, if they kept to their normal virus-collection rate, they would have gotten about thirty new viruses. What’s the chance that one of those was BANAL-52? There are thousands of bat viruses, and BANAL-52 is so rare that it wasn’t found until well after the pandemic started and people were looking for it very hard. So the chance that one of their 30 would be BANAL-52 is low. Also, they said in DEFUSE that they planned to go back to the same Yunnan cave. But BANAL-52 was found far away from that cave, so unless it ranged over a wide area, they probably couldn’t have found it even if they got very lucky. Session 3: Closing Arguments This third debate was supposed to be about “inference”, ie how much Bayesian evidence was provided by each of the facts given so far, and how to fit them into the Rootclaim probabilistic model. I’m going to relegate my summary of the more probabilistic half to the next section of this post, and just include the closing arguments here. Saar: Peter’s case hinges on the idea that it’s very improbable that a lab leak pandemic would first show up at a wet market. But this isn’t necessarily improbable. The Huanan Seafood Market had several factors that made it a likely location for a superspreader event. It was busy, with over 10,000 visitors a day. Many of the people there (eg the 1,000 vendors) came back daily, letting them reinfect each other. It had poor ventilation, especially in the high-positivity area near the raccoon-dog stall. It had cold wet surfaces on which the virus could survive for long periods. It was indoors, which prevented UV light from killing the virus. Given a small amount of sporadic COVID going around Wuhan, it’s not surprising for the first place it started spreading en masse to be a wet market. In fact, we have several examples of this. When China was COVID Zero, there would occasionally be small outbreaks that the authorities would have to contain. Most of these were at wet markets. For example, the big COVID outbreak in Beijing started at Xinfadi Market, their local seafood market. This couldn’t be an animal spillover, because there were no raccoon-dogs or other weird wildlife there. So it must be that wet markets are natural places for superspreader events. There are several other examples, which make up about half of the total outbreaks in Zero COVID era China, plus others in Singapore and Thailand. Since COVID clusters concentrate in wet markets even when there is no animal spillover, we should accept this as a property of the virus, and not attribute any significance to the fact that this happened in Wuhan too. Peter: About 1/10,000 citizens of Wuhan was a wet market vendor. So there’s a 1/10,000 chance that the first known COVID case should be a wet market vendor by chance alone. Weibo lists the most popular places for people to check in to their network on their phones, and the wet market was the 1600th most popular place in Wuhan, meaning that if you weight locations by busy-ness, there’s a less than 1/1600 chance that the first cases would be in the wet market. Yes, the wet market is indoors, has mediocre ventilation, has repeat visitors, etc. So do thousands of other places in Wuhan, like schools, hospitals, workplaces, places of worship. The wet market isn’t special in any way. And again, it wasn’t a superspreader event! COVID spread at the same rate in the wet market as it does everywhere else: doubling once per 3.5 days. It doesn’t matter what kinds of arguments you can come up with for why the wet market should have been the perfect superspreader event location, we can look at it and see that it wasn’t. It’s an environment that spreads COVID at exactly the normal rate. Zero COVID era Chinese outbreaks were concentrated in wet markets because they received infected animal products. We know why there was an outbreak in the Xinfadi Market in Beijing: it was because the seafood stall got frozen fish from some non-Zero-COVID country, the fish had COVID particles on it, and the vendor got infected and spread it to everyone else. Something like this is true for the other Chinese wet market based outbreaks we know about it. So this makes the opposite point you think it does: wet markets start outbreaks because there are infected goods being sold there. Then the virus spreads through the wet market at a completely normal rate. Saar: The Weibo list of 1600 places bigger than the wet market is likely inaccurate, because it's based on check-in data and people don't check in to seafood markets. Most of those 1600 places aren't amenable to superspread. The 70 markets supposedly bigger than Huanan are irrelevant, because they're supermarkets, open air markets, etc. Huanan is the largest seafood market in central China, and a more likely place for the first cluster of cases to be noticed. Markets weren't a common spillover location in SARS1, so the zoonosis hypothesis hasn't "called" this event in a way that should give them a high Bayes factor. And there’s still plenty of evidence for isolated (though not super-spreading) pre-market cases. A British expatriate in Wuhan, Connor Reed, says he got sick in November, three weeks before the first wet market case. Later the hospital tested his samples and said it was COVID. Another paper reports 90 cases before the first wet market one. Peter: Connor Reed was lying. The case wasn’t reported in any peer-reviewed paper. It was reported in the tabloid The Daily Mail, months after it supposedly happened. He also told the Mail that his cat died of coronavirus too, which is rare-to-impossible. Also, to get a positive hospital test, he would have had to go to the hospital, but he was 25 years old and almost no 25-year-olds go to the hospital for coronavirus. His only evidence that it was COVID was that two months later, the hospital supposedly “notified” him that it was. The hospital never informed anyone else of this extremely surprising fact which would be the biggest scientific story of the year if true. So probably he was lying. Incidentally, he died of a drug overdose shortly after giving the Mail that story; while not all drug addicts are liars, given all the other implausibilities in his story, this certainly doesn’t make him seem more credible. And in any case, he claimed he got his case at a market “like in the media” The other 90 cases are also fake. A lab leak guy found a paper that mentioned 90 more cases than other papers, and made up a conspiracy theory where the author was trying to secretly communicate that there had been 90 secret cases before any of the confirmed cases, even though there was nothing about this in the text of the paper. But actually that paper just counted cases differently than other papers, and they were referring to normal cases after the pandemic officially started. Again, I’ll come back to the discussion about inference later, but for now, here’s a table of both sides’ reasoning. This exact presentation comparing both analyses is mine3, but you can see Saar’s version here, and Peter’s starting at 45:33 of this video. Slightly made up; the two sides didn’t express their probabilities in the same way and I had to make editorial decisions to match them. Note that these aren't entirely comparable because Peter is being laxer about out-of-model probability than Saar. Although Saar's final odds here are 533-to-1, this just the central estimate. Rootclaim’s real final probability is 94% lab leak. You can see their analysis here. And The Winner Is . . . … … … … … Peter and the zoonosis hypothesis. This was a decisive victory. There were two judges, who each gave separate verdicts (or were allowed to declare a draw). Both judges decided in favor of Peter. You can see the judges’ own summary of their reasoning here (Will, Eric) Manifold agreed with the judges. There was a prediction market on who would win. It started out 70-30 in favor of lab leak. As the videos came out, zoonosis started doing better and better. I don’t want to take the exact final numbers too seriously, since I think some of the later price increases involved hints from the participants’ behavior. But it’s clear which way viewers thought the wind was blowing4. Around the same time, the Good Judgment Project - Philip Tetlock’s group studying superforecasters - put out a report on the lab leak hypothesis. After studying it in depth, his forecasters ended up 75-25 in favor of zoonosis. The Rootclaim debate was one of ten sources they said they found especially interesting. And also around the same time, and unrelated to any of this, the Global Catastrophic Risks Institute surveyed experts (“168 virologists, infectious disease epidemiologists, and other scientists from 47 countries”) and found the same thing (though see here for some potential problems with the survey): For what it’s worth, I was close to 50-50 before the debate, and now I’m 90-10 in favor of zoonosis. III. The Math And The Aftermath The third debate session was about “inference”, how to put evidence together. I put this part off until after disclosing the winner, because I wanted to talk about some of these issues at more length. The Math: Judges Both judges included a probabilistic analysis in their written decision. Here’s the same table as above, expanded to add the judges: I shoehorned the judges’ factors into the categories I already had; some of them were actually subtly different from Peter’s, Saar’s, and each other’s. The “priors” category is especially a mess here. We’ll go over these later, but I get the impression that they both thought of probabilistic analyses as an afterthought. For example, Judge Eric wrote 30,000 words about which considerations moved him, and only then includes the analysis, saying: I am not convinced that this Bayesian calculation is even an appropriate way to estimate the relative posterior probability of Z and LL; it just seemed fair that after criticizing Rootclaim’s calculations at length I should make an attempt at it myself. Judge Will’s decision ran to 10,000 words. He said he independently tried both reasoning it out intuitively, and running the Bayesian analysis, and was relieved when these two methods returned the same result. He said: I am skeptical that the Bayesian decision making/evaluation methods are any more "objective" than [intuitive reasoning]. I think they maximize legibility, not objectivity, and tend to hide the intuitive/heuristic portion in the data inclusion step and values, where it’s harder to see . . . I am not skilled in the Bayesian method, and I am sure I made significant mistakes. More time and practice would improve and refine my estimates. At the fundamental rules of the universe level, Bayesian analysis must be the best way to evaluate evidence. However, I am unsure that it’s a good strategy for a human given our cognitive limitations, and doubly unsure it’s truly being used (in the dispassionate sense) where the outcome is social desirability/fame/Twitter likes. I’m focusing on this because Saar’s opinion is that the debate went wrong (for his side) because he didn’t realize the judges were going to use Bayesian math, they did the math wrong (because Saar hadn’t done enough work explaining how to do it right), and so they got the wrong answer. I want to discuss the math errors he thinks the judges made, but this discussion would be incomplete without mentioning that the judges themselves say the numbers were only a supplement for their intuitive reasoning. That having been said, let’s look deeper into some of Saar’s concerns. The Math: Extreme Odds Saar complained that Peter’s odds were too extreme. For example, Peter said there was only a 1/10,000 chance that a lab leak pandemic would first show up at a wet market. Peter’s argument went something like: obviously a zoonotic pandemic would start at a site selling weird animals. But a lab leak pandemic - if it didn’t start at the lab - could show up anywhere. 1/10,000 Wuhan citizens work at the wet market. So if a lab leak was going to show up somewhere random, the wet market was a 1/10,000 chance. Saar had specific arguments against this, but he also had a more general argument: you should rarely see odds like 1/10,000 outside of well-understood domains. In his blog post, he gave this example: A prosecutor shows the court a statistical analysis of which DNA markers matched the defendant and their prevalence, arriving at a 1E-9 probability they would all match a random person, implying a Bayes factor near 1E9 for guilty. But if we try to estimate p(DNA|~guilty) by truly assuming innocence, it is immediately evident how ridiculous it is to claim only 1 out of a billion innocent suspects will have a DNA match to the crime scene. There are obviously far better explanations like a lab mistake, framing, an object of the suspect being brought by someone to the scene, etc. So the real p(wet market|lab leak) isn’t the 1/10,000 chance a pandemic arising in a random place hits the wet market, but the (higher?) probability that there’s something wrong with Peter’s argument. Then Saar tried to show specific things that might be wrong with Peter’s argument. I didn’t find his specific examples convincing. But maybe the question shouldn’t be whether I agreed with him. It should be whether I’m so confident he’s wrong that I would give it 10,000-to-1 odds. This makes total sense, it’s absolutely true, and I want to be really, really careful with it. If you take this kind of reasoning too far, you can convince yourself that the sun won’t rise tomorrow morning. All you have to do is propose 100 different reasons the sunrise might not happen. For example: The sun might go nova.
Inline links: https://substackcdn.com/image/fetch/$s_!xaM6!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fa05681e0-d6d4-42e3-aa4c-a5c6620dd5e5_617x405.png, NPR, found that, Mr. Chen, failed to find a clear connection between COVID and raccoon-dogs, https://substackcdn.com/image/fetch/$s_!8Led!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F85e46d3a-c19e-407b-88b7-1d8ea1489df7_1190x503.png, Pekar 2022, Pipes 2021, says, 3, here, this video, https://substackcdn.com/image/fetch/$s_!8aU2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F815fe32d-d7ea-401b-b3a2-d8cd25b52ee8_490x780.png, https://substackcdn.com/image/fetch/$s_!0Tm_!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fd0492f69-7b7e-4611-9d76-64ef8d7f59d5_511x511.png, Will, Eric, agreed, 4, put out a report on the lab leak hypothesis, https://substackcdn.com/image/fetch/$s_!g7k2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F37f1b493-b556-41ec-925e-03f9d8bc26cb_1456x849.webp, surveyed experts, see here, https://substackcdn.com/image/fetch/$s_!Zejl!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fe9c88e87-b6ca-4c6d-840e-24da726f50b7_975x365.png, https://substackcdn.com/image/fetch/$s_!T5rV!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F4983e2cd-4151-42de-9685-08037ef7a8e8_635x788.png
Highlights From The Comments On The Lab Leak Debate
April 09, 2024 · Original source
2. Wang et al (2022) https://academic.oup.com/ve/article/8/1/veac046/6601809 also confirms that the raccoon dogs were wild caught in Hubei. What's more, Wang et al (2022) tested 15 wild raccoon dogs of suppliers of Wuhan markets, including the Huanan market, in January 2020 and found them to be negative for SARS-CoV-2. On average, 38 raccoon dogs were sold across the four markets in Wuhan from 2017 to 2019. So, the 15 raccoon dogs likely comprised nearly the whole inventory of raccoon dogs that would have been supplied to the Huanan market at the time […]
Inline links: https://academic.oup.com/ve/article/8/1/veac046/6601809
Xiao (2021) Table 1 only says that some raccoon dogs in Wuhan had wounds, suggesting they were wild-caught. It makes no claims that all raccoon-dogs were wild-caught. There are dozens of raccoon-dog farms in the same province as Wuhan.
IIUC Wang (2022) says that 38 raccoon dogs were sold in Wuhan per month, not 38 during the whole two-year study period, so the claim that the traders in Wang represent the whole supply fails. [EDIT: Possibly I misunderstood this, see here]
Inline links: here
24
May 13, 2024 · Original source
People changed their minds a little over time, but not in a very consistent way that mattered much in the end. What was the “client feedback”? The report says: Client feedback was provided to the Superforecasters on December 21. The client posed questions to the Superforecasters about their assessments up to that date and asked for their reactions to several studies and articles. In the days following the client engagement, the Superforecasters lowered their confidence in the natural zoonosis hypothesis from 73% to 67%, although zoonosis remained the most likely potential cause in their assessment. But following an active engagement with recent genomic studies and historical base rates of zoonotic spillovers, those numbers began to return to earlier levels. January also saw increased attention to the geopolitical context and transparency issues, particularly related to research activities in Wuhan Is this bad? I’m imagining a pro-lab-leak client saying “But what about [this list of pro-lab-leak arguments]?” and then the superforecasters read them and adjust. In one sense, it’s good that they got to see more arguments; on the other, it seems like a potential route by which clients could bias the results - probabilities never quite got back to where they were before the feedback, though they got pretty close. The last-minute spike for zoonosis might be the Rootclaim debate results, which were released on 2/18. So maybe the client feedback and the Rootclaim results both slightly affected the numbers, but mostly the superforecasters started out pro-zoonosis and stuck to their guns. Dan Schwarz and the FutureSearch team say that forecasting has a “rationale-shaped hole”. Despite the report making this sound like a pretty intense process, we don’t get much information about details: In their extensive discussions , Good Judgment’s Superforecasters assessed base rates and historical patterns, existing evidence and scientific analysis, geopolitical context and transparency concerns, trust in intelligence communities, and methodological constraints. 1. Base Rates and Historical Patterns: The Superforecasters frequently referenced base rates, i.e., the history of pandemics emerging from natural zoonosis versus the history of laboratory leaks, to anchor their probabilities. For the former, they discussed how the base rates are changing as the climate warms and as expanding human populations push farther into natural environments that previously saw little human presence. For the latter, they acknowledged that it has only been 12 years since the advent of CRISPR gene- editing tools, and the base rate of lab leaks in the short synthetic biology era is not yet well established. 2. New Evidence and Scientific Analysis: Throughout the period, the Superforecasters adapted their forecasts in light of new scientific evidence, including genomic analyses of SARS-CoV-2 and its relation to bat viruses, and the debate over potential laboratory manipulation. 3. Geopolitical Context and Transparency Concerns: The geopolitical implications of the virus’s origins, particularly in relation to China’s transparency and the involvement of international research institutions, played a significant role in the analysis. Concerns over data veracity, and over the political ramifications of determining that the pandemic’s origins were other than zoonosis, were extensively debated. 4. Trust in Intelligence: Commentary on trust in intelligence communities and discussions about the impact of geopolitical biases on the interpretation of evidence illustrated the complex interplay between science, politics, and human behavior in assessing the pandemic’s origins. 5. Methodological Critiques and the Evaluation of Evidence: The Superforecasters engaged in methodological critiques of the evidence base, including the scrutiny of laboratory practices and biocontainment levels [...] In the end, most Superforecasters were in rough agreement on issues like the base rates of zoonotic spillover. Where they most often disagreed was on the interpretation of actions by Chinese officials and whether their actions reflected how an authoritarian government would react in any crisis over which it did not have full control, or whether those actions were indicative of attempts to cover up a biomedical research-related accident that allowed the SARS-CoV-2 virus to enter circulation in China and, ultimately, the entire globe. Probably it would be too much to ask for to get a transcript of all their discussions - then they’d be nervous saying things that might make them look bad to an audience. What would be a good balance between getting more information and not imposing on their time? Forecasting is an unusually legible and easy-to-judge domain. One of the theories of change for forecasting was to use it to identify smart people with good reasoning, then turn them loose on less well-behaved problems. This is one of the first big attempts to do this at scale. How did it work? We can’t tell, because it’s inherently an illegible and hard-to-judge domain. Darn. I don’t know what I expected. Notes From A Local Optimum Austin’s concern - that forecasting has reached a local optimum - is widely shared. We have some good sites: Manifold, Metaculus, Polymarket, GJO, etc - all doing good work. We have good-ish probabilities for a few important questions. Every so often a news source cites them. Sometimes a decision-maker looks at them behind the scenes, maybe. Is this all there is? The FutureSearch team says the next step is to focus on “rationale”. We need to use forecasting not just to get a raw probability, but to explain what’s going on and why we think something. Then instead of just convincing policy-makers to trust forecasts, we can tell them why something is true, or inform their discussions even if they’re not willing to blindly trust a number. Is this a betrayal of the forecasting ethos? The original dream was that instead of a bunch of people giving arguments, we could just test who was right. Now we’re going back to the arguments? People have argued forever; what does forecasting add to that? Well, they add the knowledge that the arguments are from people who have been right a lot before and are incentivized to be right again. Still, it’s not a natural fit. Probably it’s relevant here that FutureSearch’s forecasting AI does a really good job of this by default, in a way humans can’t match. Nuno’s yearly forecasting roundup doesn’t have a single thesis, but the first part is a well-supported complaint that most forecasting sites aren’t good business. They either burn VC money, burn EA donations, or converge towards casinos to support themselves. He gives an honorable exception to Cultivate Labs, which sells prediction market software rather than the results themselves. Open Philanthropy (billionaire Dustin Moskovitz’s EA-aligned charitable foundation) has at least given forecasting a vote of confidence, recently choosing to promote it to one of their main donation areas. Still, they got a lot of pushback on the decision, for example SuperDuperForecasting here: This will be a total waste of time and money unless OpenPhil actually pushes the people it funds towards achieving real-world impact. The typical pattern in the past has been to launch yet another forecasting tournament to try to find better forecasts and forecasters. No one cares, we already know how to do this since at least 2012! The unsolved problem is translating the research into real-world impact. Does the Forecasting Research Institute have any actual commercial paying clients? What is Metaculus's revenue from actual clients rather than grants? Who are they working with and where is the evidence that they are helping high-stakes decision makers improve their thought processes? Incidentally, I note that forecasting is not actually successful even within EA at changing anything: superforecasters are generally far more relaxed about Xrisk than the median EA, but has this made any kind of difference to how EA spends its money? It seems very unlikely. And Marcus Abramovich here: I'm in the process of writing up my thoughts on forecasting in general and particularly EA's reverence for forecasting but I feel, similar to @Grayden that forecasting is a game that is nearly perfectly designed to distract EAs from useful things. It's a combination of winning, being right when others are wrong and seemingly useful, all wrapped into a fun game. I'd like to see tangible benefits to more broad funding of forecasting that seems to be done in t he millions and tens of millions of dollars. I would also be the type of person you would think would be a greater fan of forecasting. I'm the number one forecaster on Manifold and I've made tens of thousands of dollars on Polymarket. But I think we should start to think of forecasting as more of a game that EAs like to play, something like Magic the Gathering that is fun and has some relations to useful things but isn't really useful by itself. Eli Lifland has a long and hard-to-summarize comment here, response from Ozzie Gooen here, podcast between them on “Is Forecasting A Promising EA Cause Area?” here. I’m split on this. My previous hope was that the field would gradually grow, without any qualitative changes or discontinuities, until it became big enough that journalists and policy-makers were aware of it and took it seriously (compare eg the growth of the Internet as a scholarly resource). I think the strongest argument against this is Manifold’s relatively flat user numbers. Is there a new hope? I think if nothing else, forecasting might be useful as a testing ground: First, to create forecasting AIs (like FutureSearch) which can then get consulted on a variety of questions, eg by policy-makers. The biggest holdup has always been the need to gather 20 or 50 or however many hard-to-find superforecasters for whatever question you’re asking, and then trust their advice even though they’re fallible fleshbag humans. If you can use the 20 to 50 superforecasters to inspire an AI, and then test the AI and prove it’s good, people might be more interested. This is especially true if the AI can branch out beyond traditional forecasting questions. Once we have a few of these, we can start comparing the next generation of AIs to the previous generation, and skip the superforecasters.
Inline links: forecasting has a “rationale-shaped hole”, Nuno’s yearly forecasting roundup, Cultivate Labs, choosing to promote it to one of their main donation areas, here, superforecasters are generally far more relaxed about Xrisk, here, @Grayden, here, here, “Is Forecasting A Promising EA Cause Area?” here
Links For July 2025
July 01, 2025 · Original source
33: A few years ago, a group surveyed expert biologists on their beliefs about zoonotic vs. lab origins of COVID; most believed zoonosis. Lab leakers objected that many of the “experts” said they hadn’t heard of DEFUSE (the gain-of-function project Wuhan was involved in; knowing about this should be table stakes for this discussion), and many others said they had heard of a fake paper put into the poll to trap lazy/dishonest responses; this (they said) invalidated the survey. But I recently learned (X) that there’s crosstabs in the appendix, and neither of these matter - people who had heard of DEFUSE, or who honestly admitted not having heard of the fake paper, had the same answers as everyone else. In fact, opinion always divided about 77-23, regardless of whether participants had seen any particular piece of evidence, fell for the fake paper, or whatever. I guess this is good (the verdict wasn’t dependent on a few ignorant or dishonest people), but maybe also bad (shouldn’t being familiar with the best evidence for one side or the other make you believe that side more?)
Inline links: I recently learned (X)