Hi everyone! Thanks for reading my book review! I originally wrote the review in March, so I thought I’d leave a comment with an update on how things have progressed since then.
As far as I can tell, there is still no direct, conclusive evidence in favor of either hypothesis, and neither can be ruled out. However, earlier this week one of the pre-prints I linked to in Section 1 (Worobey et al.) was published in the journal Science, and has been getting a lot of coverage in the media and on Twitter. Although it does not offer any direct evidence, it might offer a piece of indirect, circumstantial evidence for the natural origins hypothesis — or it might not.
The main issue appears to be the same as I described in Section 1 — the paper is heavily based on a geospatial analysis of early COVID cases, and shows that they seem to cluster around the Huanan seafood market. Chan claims that this analysis is invalid due to ascertainment bias, since early investigators were specifically searching around the seafood market and surrounding hospitals for cases. Worobey et al. deny this claim of ascertainment bias.
There is also some debate over how well the Worobey et al. paper can distinguish between a scenario in which COVID first spread to humans from animals in the market, and a scenario in which a human brought COVID into the market and caused a super-spreader event.
I’m not knowledgeable enough on these technical issues to know who’s right. To me it looks like the Worobey et al. paper is either moderate, circumstantial (but still inconclusive) evidence in favor of natural origins, or neutral between the hypotheses.
By the way, I still stand by my strongest conclusion from the review — that the people who prematurely dismissed the lab leak hypothesis without sufficient evidence got it wrong, and the people who called for a full, open investigation of both hypotheses got it right. And this is still going to be the case even if the lab leak hypothesis turns out to be incorrect.
I am not an expert, but I know those who are, and we were all convinced by this latest paper. It ties together statistical arguments that:
- Early cases clustered close to the market
- Early cases clustered on the side of the market with mammals
- BOTH early strains are tied to the market
And crucially, it makes the claim that these early cases were not due to ascertainment bias, because they were detected before people even realized there was an outbreak around the market.
If the strain came from a lab worker, it seems unlikely that (a) they'd spread it at the market and not any other possible superspreader site, (b) it would just so happen to be on the side of the market with mammals, and (c) it would have happened TWICE, with TWO STRAINS. It's possible, but implausibly so.
While there have been many developments since we published, I don't expect the conclusion to change once we update (hopefully soon).
The nice thing about a good probabilistic model is that it takes into account that some of the evidence will turn out to be wrong, and new evidence will come up, so you rarely get large swings in the likelihoods over time.
Note that as with most of our controversial conclusions, we offer a $100,000 challenge to anyone who can convince unbiased judges that a natural origin is more likely than a lab leak.
None of the experts confidently claiming "case closed", "dispositive evidence", or "99.99% zoonosis" have even inquired about the challenge, despite readers repeatedly pointing them to it. I believe this gives a much stronger indication of their true confidence than their public statements.
From the review: "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."
Am I mis-reading this sentence? Because "uncommon" doesn't seem quite how I would describe over 200 accidental releases per year.
This is a GREAT review. FYI, I work in an organization directly affected by the pandemic-I spent hundreds of extra hours supporting people responding directly to the pandemic., so I have a bit of an inside view (with all the good and bad that brings). As best I can tell, this is an even-handed, insightful review. Thanks!
The closing of the cave - a sign that the Chinese themselves suspected the source of their Wuhan samples was the source of (ETA: some version of) the virus, and not the wet market?
The geospatial analysis studies are amazing in all the worst ways. Note one of the coauthors on the latest preprint is Kristian Andersen, who (imho) is intent on releasing one misleading study after another to disprove lab origin, but only after: 1) emailing Fauci in early 2020 to express serious concerns COVID-19 has hallmark signs of engineering; 2) joining a private Fauci-convened call with virologists, the majority of whom expressed private concerns this was probably engineered (notes from the call redacted); 3) days later doing a 180 and saying this absolutely couldn't have been engineered because a) pangolins and b) the spike protein isn't what his computer model would've optimized for; 4) receiving millions in subsequent NIH grants; 5) deleting all his tweet history after being called out for hypocrisy and inconsistencies, then claiming he didn't do it but rather Twitter autodeleted his tweets (as they do...?)
"I guess what I’m trying to say here is that scientific institutions, though probably not as bad as Wall Street"
I'm afraid you got it exactly backwards. A finance guy would have some incentive to lie to you (and themselves) about regulation, but the incentive is mildish as impact of stricter regulations would be distributed over whole industry and unlikely to be devastating to him personally. However scientific institutions have (or feel) their reputation and whole livelihood is tied to being correct on these single issues you are interested in and a good share of people in such situations will do whatever it takes to be "right".
A few notes on the points mentioned in the review that touch on the object-level origin debate:
I believe that it isn't that unlikely for diseases to spread long distances like COVID would have had to if it started in the countryside. Poor towns in rural China probably do not get many out of town visitors, so an epidemic could slowly work its way through the countryside until someone visits the city and starts a large outbreak. If I recall correctly, something like this did happen for at least one previous epidemic ( Ebola? Mers?). Relatedly, Wuhan may be far in absolute distance from those remote provinces, but there are not a lot of cities in Western China. Wuhan is not the closest big city, but it is one of the closer ones. And of course, as you note, 96.8% isn't that close--there are almost certainly many, many relatives of the viruses that have already been found in the wild hanging out in caves throughout the countryside, totally undiscovered.
The Furin Cleavage Site doesn't exist in all coronaviruses, but it exists in enough of them (and doesn't exist in others) that it probably evolved multiple times, so it wouldn't be surprising to appear on another one: https://www.sciencedirect.com/science/article/pii/S1873506120304165
It's now got to the point where I read this article, that seems to me like a perfectly reasoned and reasonable assessment of the evidence, and I'm clenching because the better it looks, the more I fear it's fooling me in some way. D'oh!
I guess the only conclusion is to strongly endorse the suggestion at the end: "What differentiates science from other ways of knowing is its self-correction mechanisms." If I had the technical expertise to pick my way painstakingly through each claim, I'd do it; I just have to be glad that someone somewhere is doing it, and is making their efforts public for the rest of us to verify.
Incidentally, this is the thing that Chinese politicians don't "get". They think their political censorship can be separated from the country's technical expertise, and that they can have a scientifically and technologically strong nation that loves the Party. They've make a lot of effort to realise this vision. But I can't help thinking that they're doomed to failure, because both scientific knowledge and the scientific process are messy and will spill over into politically sensitive areas. If they're chilled in those areas, the effects can be systemic.
I loved this book (and all books where Matt Ridley is involved) and after reading I am sure that the Wuhan virus was an accidental release of a gain-of-function strain created at the WIV. The furin cleavage site is just one of smoking gun pieces of evidence - what's the likelihood that a natural FCS would have the same two codons optimized for human expression rather than a random combination of codons? What's the likelihood that an FCS would naturally show up in a virus belonging to a group of viruses that normally do not have an FCS and that this virus would be found across the street from the one lab in the world that just a few years prior published articles on adding and removing FCSs to and from other viruses?
The wet market story was a part of the coverup right from the start. WIV of course knew about the leak as soon as one or more of their workers contracted a pneumonia very similar to the pneumonia seen in the Mojiang mine workers. The researchers were sloppy but not stupid and they went into emergency denial and coverup mode immediately, purging WIV databases and making up stories about the wet market that had nothing to do with their virus. Basically you cannot believe anything that came out of WIV or China in general regarding the Wuhan virus after October 2019, when the leak most likely occurred (or at least when the coverup started).
Read the whole book, it's very instructive, detailed and for me, absolutely convincing about the lab leak theory.
Another reason to separate the object-level issue from the institutional issue is our motivation to know / care about it. Whether the virus came from an animal or from a lab changes nothing in my life. I can’t make China change their laws and customs. But the question of whether our institutions are trustworthy impacts me very directly.
It’s pretty scary how low the level of trust is right now. You see those memes like, “What’s the difference between a conspiracy theory and the truth?” “About six months”.
There does seem to be one issue of either clarity or accuracy:
From review (section 5. Lack of institutional transparency and competence):
> Virus sample sequences from early COVID patients in China were originally uploaded to an online database, but later removed in an apparent attempt at obfuscation. However, evolutionary biologist Jesse Bloom came up with a clever way to recover this data. [link to June 23, 2021 article]
>Probably the most important – the WIV had previously maintained a database of at least fifteen thousand bat samples, including the dates and locations of samples as well as information about the viruses found in them. This database was taken offline and its contents have not been shared with independent researchers since.
> On July 5, more than a year after the researchers withdrew the sequences from the Sequence Read Archive and two weeks after Dr. Bloom’s report was published online, the sequences were quietly uploaded to a database maintained by China National Center for Bioinformation by Ben Hu, a researcher at Wuhan University and a co-author of the Small paper.
I think this is my favorite one so far. It's the first one that's been all of (a) well written, (b) on an interesting topic, and (c) (afaict) free of factual errors.
Its one real downside is that it didn't tell me much I didn't already know (though in fairness I've been at least somewhat tracking this debate already). I guess it could have managed that if it had explained some of the technical biology stuff, but maybe that's too hard to do for non-experts and would have gone over my head anyway.
Definitely refreshing to read the only review (so far) about a topic that's Completely Absolutely Undeniably No Refutability topical. Instead of the awkward shell game shoehorn of other reviews "well you see, this is *sort of* related to Russia and Ukraine, or climate change, or EA, or...etc". Those have been the weakest parts of several so far, the stretching-for-relevance, so it's nice not to have to do that here at all. Kind of like the South Park trope of every episode needing a Moral Lecture.
Of course, one trouble with reviewing recent history is not keeping up with the current Jones developments - but you did note those areas in the review, good on you. Otherwise there'd be a whole lot more critical comments already, heh. I like a review written thoughtfully enough to proactively address potential relevant criticism. Shows a good audience-model.
>This artificial scientific consensus was then picked up by tech companies, who used it to label discussion of the lab leak hypothesis as “misinformation”, as well as by media sources and fact checkers. I don’t want to get bogged down in all the convoluted details [...]
This was the most disappointing part for me; I'm unclear whether this was simply not a focus of the book at all, or just the author deciding not to review these (admittedly very-CW) aspects, but it's consistently been *the* most interesting part of the entire Covid Show. Watching - well, living through, in real time - the genesis of a Narrative, impervious to Facts or Criticism, the abuse of "fact checking", the new slur of "misinformation"...again and again and again, with so many people concluding This Is Fine, if they even noticed the smoke at all. Even a few in the Rationalist community, who did not call foul that this is not, in fact, the Way. Partisanship comes even to those who claim to be above it, when the stakes are high enough.
In other words, as some others have already commented, it doesn't really affect my life that much whether the origin is ultimately zoonotic or lab...but the erosion of institutional trust has had extreme ramifications, for me personally and also society-wide. I would still like an "autopsy" report on this failure-cascade at some point. The fact that it sure looks like these patterns are being repeated with monkeypox is not at all reassuring.
That being said, this was a well-written, comprehensive, factual, and even-handed (afaict, not an expert) summary of The Story So Far. Just a pretty solid book review, without any irritating writerly tics, needs-an-editor jankiness, or flat-falling humour. (It is entirely possible to write humour-lite without being academically dry!) I also very much appreciated the epistemic humility of admitting I Don't Know, especially coming from a somewhat-SME. There's a time and a place for big bold postulates and wild reaching claims, but I think still-ongoing history isn't one of those times and places. Can't remember who said it first, but if we know anything about covid, "covid is where intuition goes to die". When constant surprises are the name of the game, it's probably ideal-payoff to hedge on uncertainty, rather than double down or appeal to experts who might well turn out quite biased. Good job, I now have 3 potential finalists to decide my vote between.
If someone withholds, alters, or destroys evidence, as a matter of science, we don't know what the evidence is and can't update our theory. As you said, maybe the person destroys evidence for other reasons.
But in the law, which is the search for truth in messy human affairs, there is a rebuttable presumption that the evidence is bad for the person who did the destroying (it's called "spoliation" if you want to look it up).
If they also offer up an explanation that turns out to be false, the inference becomes stronger.
This approach could lead to the wrong answer, so it should be tempered, but shouldn't we shift the burden this way?
Good review and thoughts. On the question of rationality and how to update priors, it seems like there's a missing admission here that your priors were wrong at the start, and although you updated on this specific issue you haven't done any updating of the original bad priors that led you astray.
By the way: "When people talk about trusting the experts, I think they mean trusting people with technical expertise over people without technical expertise"
I don't think that's what it means. When people talk about trusting the "experts" they're using the word "expert" in a very limited way to mean, essentially, public sector bureaucrats or academics. Obviously the two groups overlap significantly. Any people with expertise that is NOT one of those groups, or which contradicts them, is simply dismissed immediately. Thus we can see that exhortions to "trust experts" or "follow the science" are in reality just leftist demands to obey public sector pseudo-intellectuals regardless of merit; the same problem that leftism has had since it first emerged.
An example of this is when it was discovered early on in the pandemic that epidemiological models were never validated against reality, and the Prof Ferguson Imperial College model was filled with programming errors. Or phrased another way, that academic epidemiology appeared to lack technical expertise. Many professional modellers and programmers from the private sector stated flat out that the Ferguson model was so incompetently made that it was broken to the point of worthlessness, creating entirely different predictions every time it was run even when all inputs were supposed to be controlled. But these claims were completely ignored by the media, government and academia, and in fact were swiftly suppressed via (yet another) academic conspiracy in which ICL found a corrupt academic at Cambridge to claim that he'd tested the model and found the outputs to be replicable. In fact the report he issued (never peer reviewed or submitted to any journal) stated that none of the results he got were the same but he considered the model replicable because the "trends" were the same.
Part of why this happens is that academics are able to talk to the press very freely. Every other kind of expertise works for organizations that long ago learned that the press aren't really trustworthy, and so are banned from talking to the media or can only do so after their statements are vetted very carefully. So journalists end up relying on academics to rapidly provide quotes on any topic, and lend their stories the aura of gravitas and specialist knowledge, but people see through it and so trust in the media falls. Falling trust in science itself is lagging behind but now catching up as people realize that when scientists claim the media is distorting what they said, it's not really true and the root source of the false claims is the academics themselves.
It seems to me that the old saying about being "unable to see the wood for the trees" might apply here. A novel virus arise thousands of kilometers from its natural home in Yunnan. But just a few hundred meters from an institute of virology that just happens to study that kind of virus, that has brought hundreds of samples back from Yunnan, one of which is a close genetic match, and which happens to conduct research into gain of function which includes changing the viral genome to make it better at infecting humans, including inserting an FCS, which is not found in other coronaviruses? And did so with grants from an American organisation that led the crusade in advance of any investigation to declare the lab-leak hypothesis a conspiracy theory?
Now we can wait for a full investigation (which we all know will never be permitted by the Chinese government), update our priors all day long, and avoid theorising in advance of the evidence. But if I asked you to bet on one side or the other here, would you really bet that it's all a string of the most unlikely co-incidences? I'm not a gambling man, but I can see the odds here. Those aren't just trees: they are together, a wood.
Excellent post. A lot of people in the comments are using the phrase "conspiracy theory" to describe the lab leak and cover-up, but to me that doesn't require much conspiracy. And if somebody came up to me wearing a tin-foil hat talking about an accidental lab leak, I would think "you need more interesting sources".
Actually I'm a bit surprised nobody here has posted a real tin-foil hat conspiracy theory about COVID origins, so in the interest of completeness here's one: what if COVID was deliberately released in Wuhan by a non-chinese malicious actor, with the intention of making it look like a lab leak from the WIV? If somebody did want to release a bio-weapon, Wuhan would be a natural choice because the WIV provides such a convenient scapegoat. With the virus origin unknown, the CCP would likely still engage in the same censorship we saw. And the more people argue about lab leak vs natural origin, the more our hypothetical bio-terrorist would be able to slink back into the shadows.
I don't know how likely that scenario is - more than 5% chance? - but it's worth keeping in mind that the space of possible origins is large, and P(lab leak) + P(natural origin) adds up to less than 1.
> Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published. [...] It’s completely unacceptable that BANAL-52 wasn’t mentioned.
Two months before publication mean almost certainly that the book was already off to the printer. Sorry, but production times in the book publication industry are usually quite long, and it's completely unreasonable to expect consideration of research that comes in so shortly before publication.
Methinks you characterize the claims of "conspiracy" too broadly.
Associated with the "lab leak" rhetoric was often the implication or outright claim that a leak was deliberate.. Which in turn was often associated with "big Pharma" conspiracy-mongering.
That was part of what the "conspiracy" claims were directed at, and often there was a spillover effect, where more limited arguments about an unintentional lab leak were mixed in with arguments about a deliberate one.
I think it's also important to discuss the importance of not alienating scientists in China, given the large risks associated with virus spread from China and in particular because of the huge animal trade there. Many assume that pushback against the lab leak argument was merely because of CYA, and I think for much of the scientific community much of the concern was legitimately because of the need to maintain strong institutional collaboration.
I saw a talk by a member of the WHO team in the early days of the pandemic. They were definitely extremely frustrated about not getting access to any actual evidence, and I think some of the premature definiteness of the WHO report can be put down to "we have to say something/absence of evidence is evidence of absence" type failures.
That said, one point that the WHO team member brought up that I haven't seen discussed much is that there are biomarkers for viruses raised in labs that aren't just biomarkers for genetic manipulation. In particular, viruses kept alive in cells in petri dishes often have adaptations specific to fitness in that environment, which can stick around when the virus makes the leap to humans, and COVID-19 lacks those biomarkers.
Since this was pretty early in the pandemic, I don't know whether this argument has since been refuted, though. Anyone know more?
> "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?"
I'm not an expert by any means, but this looks like a standard sample bias issue. Most Chinese caves are not sampled in search of coronaviruses. This was, but only because of the unlikely event of miners getting sick. If natural reservoirs for viruses are not uniformly sampled, closely related viruses aren't going to show up where they are, but where research has been conducted. Probably this have been said already.
Regarding the issue of the virus being first detected in Wuhan. I wouldn't be so sure as to consider this as unlikely or suspicious. IIRC, for unexposed populations SARS-CoV-2 had an R0 of about three, but most people won't transmit the virus to 3 new hosts, in fact the mode was closer to 0 and the mean was driven up by a small number of superspreader events. The virus may have self extinguished in the countryside and only picked up momentum in a large hub like Wuhan.
Even if there was an earlier or simultaneous rural outbreak outside Wuhan, the combination of asymptomatic/mild cases with the virus causing symptoms similar to other respiratory viruses would have made it more likely to be discovered, as a new illness, in Wuhan.
I’m not even close to having a working knowledge of virology let alone and expert, and please correct me if I’m wrong and I’ll delete the comment. But I believe you grossed over a pretty important aspect of the “cleavage site” in that there may have been a grant that was rejected for the WIV to research inserting the cleavage sites into bat corona viruses in the couple of years prior to the outbreak. The idea seems to be that they went ahead with this research anyway, otherwise it would be an insane coincidence.
I believe it was Alina Chan who said something along the lines of “imagine if there was a rejected proposal in a city to put horns on horses, then a year later unicorns started showing up right down the street from the lab that was supposed to be doing this research. It could be a strange coincidence, or more likely they did the research anyway.” Since the “cleavage site” had never been found in nature the comparison isn’t much of a stretch.
Did I hear this correctly or am I talking out of my ass? Lol I believe it was on Coleman Hughes podcast over a year ago and I haven’t revisited this topic.
"Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published"
This seems like a weak criticism. Publishing involves a lot of lead time devoted to copy editing, proofreading, typesetting, book design, manufacturing, etc. Even if the authors became aware of BANAL-52 before their book was officially released, it may have been too late to correct the first edition.
Re the location of the WIV, yes, it cannot help but be highly suspicious to any sensible person that the first outbreak occurred in the same city as a major research lab that is known to work with exactly this type of virus. However, I've seen statements in print that went too far with this idea, claiming that the lab is literally across the street from the Huanan Seafood Market. It isn't. They're nearly 8 miles apart, on opposite sides of the Yangtze River. Now, 8 miles, admittedly, is not that much, certainly not enough to discredit the lab leak hypothesis, but it's a lot more than "across the street".
This excellent review has caused me revisit how I formed my beliefs about COVID’s origins in the early days of the pandemic.
I now see that i dismissed the lab-origin theory not because I had evaluated the evidence, but because the Trumpers were pushing it.
Disbelieving something solely because you don’t like those who claim to believe it is an easy error to make. In this case, I assumed the Trumpers didn’t really believe the lab-origin theory either but claimed to because it gave Trump the opportunity to sound tough on China and to undercut the science community, both things he and his tribe were particularly keen to do. For that reason, I dismissed it.
I still think I’m right about why the Trump tribe embraced the lab-origin theory, but I was wrong to dismiss it solely for that reason.
>"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 ... "
The paperback edition's epilogue does discuss the preprints. (It isn't able to discuss the first two's final published versions, out this week, though.)
<quote>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.</quote>
My understanding of book publishing implies that this criticism is completely off base. The book publishing pipeline does not permit rewriting two months before publication. The period from the text being locked in until the time the book is in customers’ hands is more like four months.
Epistemic status: I’ve dealt with publishers for years, from the author side, and make tens of thousands per year from book royalties. But I’ve never tried a last-minute change to a book, so maybe I’m wrong. I welcome correction from an actual publisher.
> To say there was a lab origin, we would have to postulate that scientific institutions in China are lying and successfully engaged in a coverup, for which there have been no credible whistleblowers.
This is clearly not true; many people believe that the virus originated in a US lab.
I haven't read either document deeply but they seem to at least corroborate that dodgy gain of function research creating similar viruses happened in unsafe conditions at WIV around the same time. Whether this was the source of COVID I don't know, but it's alarming either way.
Hungarian scientists studying soil samples collected in Antarctica in 2018, belatedly noticed that they contained traces of a very early strain of SARS CoV-2. Which is ridiculously implausible. So what really happened is, they sent their samples to Sangon Biotech in Shanghai, China for phylogenetic analysis, and Sangon used a technique that is notoriously susceptible to cross-contamination from other samples being sequenced in parallel. Clearly someone else had sent actual SARS CoV-2 samples to Sangon for sequencing at the same time.
The Hungarians sent their samples to Sangon in December 2019. Exact date unknown, but almost certainly nobody was sending samples from the Wuhan outbreak to Sangon in 2019. The first report of genetic sequencing of suspected Covid-19 was 24 December 2019, but that was by Vision Medicals in Guangzhou, which did a lot of the (known) early Covid sequencing. Wikipedia also says that Beijing Boao Medical Laboratory sequenced Covid-19 on 27 December 2019. The Wuhan Central Hospital doesn't seem to have used Sangon in this area, and they didn't tell the Chinese equivalent of the CDC about any of this until 29 December 2019 so probably nobody else was doing so in 2019 either.
The Wuhan Institute of Virology had been using Sangon Biotech as a contract lab earlier in 2019. For research on bat coronaviruses, even, if I read this correctly: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6521148/
Furthermore, the contamination in the Hungarian samples included not just SARS CoV-2, but also Chinese hamster and Vero E6 green monkey genetic material. The Dazhak/EcoHealth proposal for gain-of-function research in bat coronaviruses at WIV, called for culturing in Chinese hamster and Vero E6 green monkey cell lines. I'm not sure, but I don't think that's standard procedure for "we found this gunk in the lungs of a mystery pneumonia patient; can you tell us what it is?"
This isn't quite smoking-gun proof that WIV was working with proto-Covid shortly before the known outbreak, but that's certainly a plausible interpretation.
Alternately, the Hungarians could be simply wrong. Their research by nature can't be replicated, but it could at least be closely scrutinized by outside experts, and I don't think really has been. I'd expect Sangon Biotech could confirm or deny that the sequences in the Hungarian paper are the ones they sent them, which should rule out "the Hungarians are just plain lying"
Also, it's barely possible to massage the timeline so that the Hungarian samples are delayed in processing to January 2020, and that the WIV or someone else in their early-2020 investigation of a natural COVID outbreak rushed their own samples to Shanghai and jumped the queue to get in the same run as the earlier Hungarian samples. But that would require that WIV do a hasty investigation that involves culturing their unknown virus in green monkey and hamster cells (why?), or that the Hungarian samples have been doubly contaminated, first with SARS CoV-2 and then from some completely different source that coincidentally used the same culture media as the proposed WIV GoF experiments.
Not smoking-gun proof, but a strong hint with some good pointers to where one might find smoking-gun proof. But as far as I know, there's been no real follow-up.
To repeat only a few parts relevant to the book review:
"Probably the most important – the WIV had previously maintained a database of at least fifteen thousand bat samples, including the dates and locations of samples as well as information about the viruses found in them. This database was taken offline and its contents have not been shared with independent researchers since."
That database was only really online for a few weeks in August-September 2019, and 5 days in December. This claim came from the DRASTIC group of freelance researchers, who cite the archives of a database monitoring system to claim that the database went offline on September 12, 2019. But the same monitoring system shows that the database only became reliably available on August 23, 2019. Before that, from June to August 23, the database was online sporadically. It would be online for 1–2 days, and then go down for several days or weeks at a time. Before June 2019, there’s no data, because v2 of the database was released in June 2019. After September 2019, it was online from December 12 to December 16, and access was sporadic through February 26, 2020. After that, the database was down for good.
"So it does seem like a pretty remarkable coincidence that the outbreak happened in Wuhan. "
Then it's a far, far more remarkable coincidence that most of the earliest cases have direct ties to the wet market. Out of all possible locations, SARS-CoV-2 just happened to pick a wet market as its first superspreader location--just like its cousin SARS-CoV-1--and it has nothing to do with the animals there? It didn't spread at a cinema, a school, a convention center, a train station, a mall, a grocery store, a university, an office building, a bar, or even the Wuhan Institute of Virology (which is 15 km from the wet market), but at a wet market, and this is totally a coincidence?
"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? "
If SARS-CoV-1 originated in an animal somewhere around the Yunnan area--which we now know it did--how did it make all the way to Guangdong without leaving a trail along the way? And of course, though it's plausible that SARS-CoV-2 originated near Yunnan, we don't know that.
In fact, it's very easy to imagine how both of these viruses could travel thousands of kilometers without leaving a trail along the way. In Yunnan, a villager traps a wild animal and gets infected with the virus. He takes a train to Wuhan--one of China's biggest cities and busiest transport hubs--and sells the animal to the wet market, infecting a few wet market retailers and customers along the way. The animal is sold, butchered, and eaten, but the infected humans start the pandemic.
"Some scientists claim that SARS-CoV-2 reached genetic stability early on, suggesting that it was already well-adapted to spread in humans at the start of the outbreak. Some have interpreted this as evidence that it was engineered for this purpose, or underwent serial passaging to encourage adaptation to human or humanized cells. "
The original virus was moderately transmissible, with R0 of 2-3. Since then, a series of variants have taken over the original strain, making the virus more and more transmissible. Delta had R0=5.1, Omicron BA.1 had R0=9.5, and BA.2, BA.4, and BA.5 are more infectious still. SARS-CoV-2 is now one of the most transmissible diseases in human history. So in what possible way has the virus "reached genetic stability early on", or "was already well-adapted to spread in humans at the start of the outbreak"?
This is Alina Chan, one of the co-authors of the book VIRAL: The Search for the Origin of Covid-19.
I really enjoyed reading this review and the comments, and I also appreciate the reviewer creating a list of other sources to check out. In particular, this line by the reviewer characterizes many of the moments I've experienced in the past 2 years spent looking into the origin of Covid-19 : “what the hell is going on if some random Twitter users are consistently correcting world-renowned virology institutes on various mistakes and omissions?”
Similar to the reviewer, Matt and I were also late to the lab origin hypothesis. The early information coming out of China seemed to point to a repeat of the 2003 SARS outbreak in China, and I didn't question this, assuming that they would very quickly find the infected animal source as they had in the case of SARS and MERS (I also recommend Laura Kahn's piece in the Bulletin Atomic). It was only later, as key evidence failed to turn up for a market origin, that I was motivated to consider the lab leak hypothesis. By May 2020, the Chinese CDC director had told the press that the market was most likely a victim, ie a sick person had ignited a superspreader event there (the CCDC director recently retired from his position, some speculate it was because he was a "loose cannon" and said too many things about the origin of Covid-19 and China's vaccines that didn't adhere to the official Party line).
One observation that raised my eyebrows (but is still consistent with both natural and lab origin hypotheses) was the highly transmissible nature of the virus right from the get go. And I wasn't the only scientist who was struck by the well adapted/pre-adapted nature of the virus. The natural origin proponents who wrote Proximal Origin in early 2020 were among the first to describe the virus as well adapted for human transmission. They later reversed their stance in their Critical Review in mid 2021. Now, in their latest preprints/papers in Science in mid 2022, they've gone back to saying that the virus needed no adaptive mutations to spread in humans. See my twitter thread for more details: https://twitter.com/Ayjchan/status/1513586519159697410
I also liked the coin flip analogy by the reviewer. I've been accurately described as a fence-sitter in the past 2 years. Back in 2020 it was very controversial for a scientist to be a fence-sitter on the origin of Covid-19 issue (and I took a lot of heat from other scientists, the press, and even friends for just saying we have to consider a lab origin, regardless of how likely or unlikely), but now it's become too boring for the media to report that a scientist thinks there's a 50:50 chance of a market vs lab origin. The media prefers to report on scientists who are willing to go so far as to say they can put one hypothesis to rest or that they have "dispositive" or "incontrovertible" evidence of one hypothesis. To do so is, imo, more like a form of gambling rather than science. Like the reviewer argues, whether it turns out that your bet was right or wrong, it doesn't justify shutting down one of the two plausible hypotheses.
I'd like to address the unavoidable weakness of the book raised by the reviewer: The manuscript is "locked" at least a couple of months before physical copies of it magically appear on bookshelves across the US and UK (not being sarcastic, as a first time book author, this truly was a magical experience for me!). This means that any developments in the 2 months before the book is released cannot possibly be written into the book unless the authors had insider information and knew that it was ok to write said information into the book (which then has to spend time being vetted by the publisher's lawyers). For this reason, the newly released paperbacks cover the new developments since mid-Sep 2021: the Laos BANAL viruses, the 3 preprints on the Wuhan market, and more FOIA'ed emails from prominent virologists who privately fretted about a lab origin of Covid-19. If the publishers will allow, I'd like to update the book 5 years from now because I'm optimistic about very compelling evidence being unearthed within this time range (at least this evidence should be more compelling than Western virologists misinterpreting data collected with ascertainment bias).
Thanks again for the review! I'm providing these links in case other readers are interested.
My peer reviewed analysis of the furin cleavage site:
And my critique of the recent Worobey et al. Science paper, which the reviewer has already linked to - I will update this in a few days to explain the "2 strain 2 spillover" hypothesis that the authors have come up with to support the market origin hypothesis:
The UPDATED paperback is out right now. - I seem to be just the 2nd to note - And Matt Ridley IS updating on his blog "rationaloptimist.com" . So to mimic the review: This review was published July 30, nearly three days after the updated paperback was published. Although I’m writing an overall-positive review here, I don’t want to go easy on the review where serious criticism is warranted. It’s completely unacceptable that the updated paperback wasn’t mentioned!* -
I am very glad to finally see the book mentioned in ACX. I found it irritating to be ignored by the "rationalist" blogosphere (except Scott Aaronson). It seems like a rather decent review too. So, many thanks, really appreciated! -
*(Original: "Why 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!")
It’s easy to dismiss the ridiculous claim that COVID began as a bioweapon, but other claims are more difficult to evaluate. <- I don't see where this claim was engaged with. Why is it easy to dismiss this claim?
"experts make decisions based on ass-covering, career advancement, and personal risk mitigation. autists make decisions based on spreadsheets. the natural result of this is that experts are useless and autists are good at things that are tractable with spreadsheets"
Great review but (and this is arguably the books fault not the review's fault) I'd have loved to have seen some base rate information. For instance, what fraction of unusual or novel viruses (just on some measure) are lab leaks given our best information.
If even the people warning us about lab leaks are going to say that it's 1000 to 1 or something that seems pretty conclusive.
> This is basically how I think we should handle this unwarranted overconfidence from respected institutions – it should decrease our trust in these institutions, but we need to be careful not to start favoring the lab leak hypothesis out of spite. In my opinion this loss of trust should not really affect our view of the object-level question of the virus’s origin at all (although it would be nice to see some of the data being hidden, like that WIV database that was taken offline).
I don't think this amount of compartmentalization is justifiable. Actions of public officials can be important pieces of evidence. They might reflect on the object-level question of whether there was a lab leak, or they might not; in order to find out whether they weigh on the question, we can apply everyone around here's favorite theorem.
P(A|B) = P(A)*P(B|A)/P(B), where A = "lab leak is true" and B = "database deleted". You have some prior credence in the lab leak P(A). Even if that's initially very low, you should ask yourself - if the lab leak were true, would that make it *more likely* or *less likely* that they would delete that database? I think the answer is pretty clearly that this would render the WIV *more likely* to delete the database, so P(B|A) / P(B) is above 1. That means that P(A|B) > P(A), which is to say, learning the database was actually deleted should cause you to update in favor of the lab leak theory.
It's not *conclusive* evidence. It might not even be the strongest evidence out there on the question. But it *is* evidence, and it means these two questions - the one about the origin of COVID, and the one about the trustworthiness of the authorities - can't be fully disentangled. Bayes' Theorem applies here too - why would it not?
Why is it a ridiculous, easy to dismiss claim that it started as a bioweapon?
I don't know anything about the subject, and it sounds ridiculous to me, but on the other hand without any justification it sounds like you're doing the same thing you rightly criticise others for doing with the lab leak hypothesis- dismissing it based on superficial weirdness. I'm sure you're not, but I think it'd be worth briefly explaining why it can be dismissed
Can we talk Dr. Limeng Yan for a minute and how she is like Edward Snowden, Rachel Carson, and Bob Woodward rolled into one. I would build a statue to this woman the size of Lady Liberty but I guess I would settle for like a Google Doodle one day?
To be clear, the author, reviewer, and everyone else are all in agreement of certain facts: Tedros, Daszak, and Fauci - and every other academic, public health official and NGO-er - all spouted "no evidence of community spread" (while knowing the opposite to be true), in February 2020 while Yan literally put her life on the line to blow the whistle that this thing was spreading like fire and is the reason all these officials were forced to recant the lie.
Let's see what the reviewer has to say about this doctor: "This is a sad story about a scientist who tried to do the right thing, but ended up intellectually corrupted by forces beyond her control." He writes this about Yan! Whose career is ruined, must be looking over her shoulder everyday for CCP payback, and she's the *one* scientist who had the courage to speak the truth to power make these sacrifices to save millions of lives. Meanwhile if the reviewer ever calls the aforementioned liars "corrupt" or anything half as disparaging, I didn't catch it.
Look, the review itself is excellent, and the reviewer does display a healthy amount of prior updating and introspection into personal bias that were blinding him. Dear reviewer, there is still work to done: If Bob Woodward gives an interview to a left wing source like Mother Jones or worked for left aligned institution like Berkley would you dismiss him as no longer credible? Of course not. The sad fact that shines through every dive into question is that the only journalists or politicians who were ever asking asking questions about this incident are going to be on the right. It was probably the same situation is the 60's for the left looking into ecological concerns of big agriculture or nuclear power. Hopefully one day this balances out, but for now we live in the world that exists.
I enjoyed this review, mostly because it roughly matches my pre-existing beliefs, so of course I'd be pre-disposed to like it. I didn't learn much, but hopefully it will Knock Some Sense Into Those Other EvilWrong People On The Internet.
The one section I found very frustrating was the vague attacks on "public health officials, scientists, journalists, and tech companies". Instead of looking into any specific claims by specific people and assessing how true they are and whether they were justified based on the evidence available at the time and what different actors motivations might have been, you just sort of gesture vaguely about "elites and institutions" and hope the audience will nod along in outrage (which they usually will, since this is a bias that is common in the rat-sphere.) Sometimes it seems like you go to more trouble to understand the action of the Chinese government than Americans!
I can't object to anything about this review's conclusions. I find the lab leak theory plausible but not probable. The only clear evidence pointing that direction is the pandemic's Wuhan origin, which would be roughly like a US origin happening to be in Atlanta-- a striking coincidence, but not so unlikely that it demands a causal explanation. And like the reviewer I'm more concerned about institutions not having trustworthy truth-finding processes than about the actual answer they get.
That being said, I'm not satisfied with the reviewer's truth-finding process either. The analysis is a bit too shallow and disorganized to actually reach confident conclusions or contribute to the discussion beyond summarizing the book's points. Nor does it present the evidence in a particularly striking or exciting way, as ACX reviews often do. Also, maybe this is just my filter bubble but I didn't find anything particularly surprising here-- "lab leak has to be acknowledged as a possibility given the evidence" has been well inside my Overton window since the theory was first proposed.
In short, I'm satisfied with this review as far as it goes but it left me wanting considerably more. As always, many thanks for contributing!
"... To say there was a lab origin, we would have to postulate that scientific institutions in China are lying and successfully engaged in a coverup, for which there have been no credible whistleblowers."
Stunning naivete. Countries that maintain gulags where ordinary people are sent to be slaves in the cotton fields, are not to be believed without the weightiest evidence.
Whether it was a lab leak or a virus that jumped out of the endangered species being butchered in a public market without any pretense of sanitation, it makes China look bad. As a propaganda move, the best possibility is to shut up all of the witnesses and destroy all of the evidence, and let the poisonous culture war politics of the US work its magic.
As far as the US is concerned, the real issue is not what happened in China. At best it is item number 10 on the bill of particulars against the regime. The real issue is what Saint Anthony Fauci did in funding research at WIV in apparent violation of Obama administration policy.
"Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published."
Book publishing has a much longer lead time than blogging. My wife just completed the final proof reading of her book in mid July. It will be published in October.
"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."
Hi everyone! Thanks for reading my book review! I originally wrote the review in March, so I thought I’d leave a comment with an update on how things have progressed since then.
As far as I can tell, there is still no direct, conclusive evidence in favor of either hypothesis, and neither can be ruled out. However, earlier this week one of the pre-prints I linked to in Section 1 (Worobey et al.) was published in the journal Science, and has been getting a lot of coverage in the media and on Twitter. Although it does not offer any direct evidence, it might offer a piece of indirect, circumstantial evidence for the natural origins hypothesis — or it might not.
Here’s the paper:
https://www.science.org/doi/10.1126/science.abp8715
And here’s a post by Alina Chan (author of Viral) with technical critiques of the paper:
https://ayjchan.medium.com/evidence-for-a-natural-origin-of-covid-19-no-longer-dispositive-after-scientific-peer-review-af95b52499e1
The main issue appears to be the same as I described in Section 1 — the paper is heavily based on a geospatial analysis of early COVID cases, and shows that they seem to cluster around the Huanan seafood market. Chan claims that this analysis is invalid due to ascertainment bias, since early investigators were specifically searching around the seafood market and surrounding hospitals for cases. Worobey et al. deny this claim of ascertainment bias.
There is also some debate over how well the Worobey et al. paper can distinguish between a scenario in which COVID first spread to humans from animals in the market, and a scenario in which a human brought COVID into the market and caused a super-spreader event.
I’m not knowledgeable enough on these technical issues to know who’s right. To me it looks like the Worobey et al. paper is either moderate, circumstantial (but still inconclusive) evidence in favor of natural origins, or neutral between the hypotheses.
By the way, I still stand by my strongest conclusion from the review — that the people who prematurely dismissed the lab leak hypothesis without sufficient evidence got it wrong, and the people who called for a full, open investigation of both hypotheses got it right. And this is still going to be the case even if the lab leak hypothesis turns out to be incorrect.
One unfortunate aspect of writing is that published words are frozen in time but our understanding of reality is not.
Just three days ago, work was published in Science that seems to strongly support the natural origins hypothesis: https://www.science.org/doi/10.1126/science.abp8715
I am not an expert, but I know those who are, and we were all convinced by this latest paper. It ties together statistical arguments that:
- Early cases clustered close to the market
- Early cases clustered on the side of the market with mammals
- BOTH early strains are tied to the market
And crucially, it makes the claim that these early cases were not due to ascertainment bias, because they were detected before people even realized there was an outbreak around the market.
If the strain came from a lab worker, it seems unlikely that (a) they'd spread it at the market and not any other possible superspreader site, (b) it would just so happen to be on the side of the market with mammals, and (c) it would have happened TWICE, with TWO STRAINS. It's possible, but implausibly so.
Twitter summary for those who don't want to read the article: https://twitter.com/angie_rasmussen/status/1551937826580824070?fbclid=IwAR3-tllY2CsR4UWE2812f1pGt-2Jr3x0jmVFyZhbO1x63Rh0L0Zy33-xvBA
Before this Science article I was open-minded to both possibilities. After it, I am convinced.
I'm still not sure why this isn't the only thing people are talking about.
Laura Kahn’s article is great, thank you for linking to it.
Saar from Rootclaim here. Thanks for the mention!
While there have been many developments since we published, I don't expect the conclusion to change once we update (hopefully soon).
The nice thing about a good probabilistic model is that it takes into account that some of the evidence will turn out to be wrong, and new evidence will come up, so you rarely get large swings in the likelihoods over time.
Note that as with most of our controversial conclusions, we offer a $100,000 challenge to anyone who can convince unbiased judges that a natural origin is more likely than a lab leak.
None of the experts confidently claiming "case closed", "dispositive evidence", or "99.99% zoonosis" have even inquired about the challenge, despite readers repeatedly pointing them to it. I believe this gives a much stronger indication of their true confidence than their public statements.
From the review: "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."
Am I mis-reading this sentence? Because "uncommon" doesn't seem quite how I would describe over 200 accidental releases per year.
Well done.
This is a GREAT review. FYI, I work in an organization directly affected by the pandemic-I spent hundreds of extra hours supporting people responding directly to the pandemic., so I have a bit of an inside view (with all the good and bad that brings). As best I can tell, this is an even-handed, insightful review. Thanks!
The closing of the cave - a sign that the Chinese themselves suspected the source of their Wuhan samples was the source of (ETA: some version of) the virus, and not the wet market?
The geospatial analysis studies are amazing in all the worst ways. Note one of the coauthors on the latest preprint is Kristian Andersen, who (imho) is intent on releasing one misleading study after another to disprove lab origin, but only after: 1) emailing Fauci in early 2020 to express serious concerns COVID-19 has hallmark signs of engineering; 2) joining a private Fauci-convened call with virologists, the majority of whom expressed private concerns this was probably engineered (notes from the call redacted); 3) days later doing a 180 and saying this absolutely couldn't have been engineered because a) pangolins and b) the spike protein isn't what his computer model would've optimized for; 4) receiving millions in subsequent NIH grants; 5) deleting all his tweet history after being called out for hypocrisy and inconsistencies, then claiming he didn't do it but rather Twitter autodeleted his tweets (as they do...?)
"I guess what I’m trying to say here is that scientific institutions, though probably not as bad as Wall Street"
I'm afraid you got it exactly backwards. A finance guy would have some incentive to lie to you (and themselves) about regulation, but the incentive is mildish as impact of stricter regulations would be distributed over whole industry and unlikely to be devastating to him personally. However scientific institutions have (or feel) their reputation and whole livelihood is tied to being correct on these single issues you are interested in and a good share of people in such situations will do whatever it takes to be "right".
A few notes on the points mentioned in the review that touch on the object-level origin debate:
I believe that it isn't that unlikely for diseases to spread long distances like COVID would have had to if it started in the countryside. Poor towns in rural China probably do not get many out of town visitors, so an epidemic could slowly work its way through the countryside until someone visits the city and starts a large outbreak. If I recall correctly, something like this did happen for at least one previous epidemic ( Ebola? Mers?). Relatedly, Wuhan may be far in absolute distance from those remote provinces, but there are not a lot of cities in Western China. Wuhan is not the closest big city, but it is one of the closer ones. And of course, as you note, 96.8% isn't that close--there are almost certainly many, many relatives of the viruses that have already been found in the wild hanging out in caves throughout the countryside, totally undiscovered.
Also related, finding the natural reservoir of a disease is quite hard. MERS and SARS-1 were much deadlier, which made tracking the disease easier. Apparently we still haven't found the origin for Ebola, despite decades of searching (https://www.reddit.com/r/science/comments/gehvui/why_do_viruses_often_come_from_bats_a_discussion/fpp3lzv/?context=999). Also, if the Chinese government isn't cooperating, would we expect a thorough search to ever really happen?
The Furin Cleavage Site doesn't exist in all coronaviruses, but it exists in enough of them (and doesn't exist in others) that it probably evolved multiple times, so it wouldn't be surprising to appear on another one: https://www.sciencedirect.com/science/article/pii/S1873506120304165
It's now got to the point where I read this article, that seems to me like a perfectly reasoned and reasonable assessment of the evidence, and I'm clenching because the better it looks, the more I fear it's fooling me in some way. D'oh!
I guess the only conclusion is to strongly endorse the suggestion at the end: "What differentiates science from other ways of knowing is its self-correction mechanisms." If I had the technical expertise to pick my way painstakingly through each claim, I'd do it; I just have to be glad that someone somewhere is doing it, and is making their efforts public for the rest of us to verify.
Incidentally, this is the thing that Chinese politicians don't "get". They think their political censorship can be separated from the country's technical expertise, and that they can have a scientifically and technologically strong nation that loves the Party. They've make a lot of effort to realise this vision. But I can't help thinking that they're doomed to failure, because both scientific knowledge and the scientific process are messy and will spill over into politically sensitive areas. If they're chilled in those areas, the effects can be systemic.
I loved this book (and all books where Matt Ridley is involved) and after reading I am sure that the Wuhan virus was an accidental release of a gain-of-function strain created at the WIV. The furin cleavage site is just one of smoking gun pieces of evidence - what's the likelihood that a natural FCS would have the same two codons optimized for human expression rather than a random combination of codons? What's the likelihood that an FCS would naturally show up in a virus belonging to a group of viruses that normally do not have an FCS and that this virus would be found across the street from the one lab in the world that just a few years prior published articles on adding and removing FCSs to and from other viruses?
The wet market story was a part of the coverup right from the start. WIV of course knew about the leak as soon as one or more of their workers contracted a pneumonia very similar to the pneumonia seen in the Mojiang mine workers. The researchers were sloppy but not stupid and they went into emergency denial and coverup mode immediately, purging WIV databases and making up stories about the wet market that had nothing to do with their virus. Basically you cannot believe anything that came out of WIV or China in general regarding the Wuhan virus after October 2019, when the leak most likely occurred (or at least when the coverup started).
Read the whole book, it's very instructive, detailed and for me, absolutely convincing about the lab leak theory.
Another reason to separate the object-level issue from the institutional issue is our motivation to know / care about it. Whether the virus came from an animal or from a lab changes nothing in my life. I can’t make China change their laws and customs. But the question of whether our institutions are trustworthy impacts me very directly.
It’s pretty scary how low the level of trust is right now. You see those memes like, “What’s the difference between a conspiracy theory and the truth?” “About six months”.
There does seem to be one issue of either clarity or accuracy:
From review (section 5. Lack of institutional transparency and competence):
> Virus sample sequences from early COVID patients in China were originally uploaded to an online database, but later removed in an apparent attempt at obfuscation. However, evolutionary biologist Jesse Bloom came up with a clever way to recover this data. [link to June 23, 2021 article]
>Probably the most important – the WIV had previously maintained a database of at least fifteen thousand bat samples, including the dates and locations of samples as well as information about the viruses found in them. This database was taken offline and its contents have not been shared with independent researchers since.
That seems to contradict the following:
(https://www.nytimes.com/2021/07/30/science/coronavirus-sequences-lab-leak.html):
> On July 5, more than a year after the researchers withdrew the sequences from the Sequence Read Archive and two weeks after Dr. Bloom’s report was published online, the sequences were quietly uploaded to a database maintained by China National Center for Bioinformation by Ben Hu, a researcher at Wuhan University and a co-author of the Small paper.
Supposedly the data is now found here: https://ngdc.cncb.ac.cn/search/?dbId=gsa&q=CRA004499
Have I misunderstood something?
I think this is my favorite one so far. It's the first one that's been all of (a) well written, (b) on an interesting topic, and (c) (afaict) free of factual errors.
Its one real downside is that it didn't tell me much I didn't already know (though in fairness I've been at least somewhat tracking this debate already). I guess it could have managed that if it had explained some of the technical biology stuff, but maybe that's too hard to do for non-experts and would have gone over my head anyway.
Obligatory link to forecasts on this topic: <https://metaforecast.org/?query=covid+laboratory>
I will also note that when asking forecaster friends, some tended to be be pretty high.
Definitely refreshing to read the only review (so far) about a topic that's Completely Absolutely Undeniably No Refutability topical. Instead of the awkward shell game shoehorn of other reviews "well you see, this is *sort of* related to Russia and Ukraine, or climate change, or EA, or...etc". Those have been the weakest parts of several so far, the stretching-for-relevance, so it's nice not to have to do that here at all. Kind of like the South Park trope of every episode needing a Moral Lecture.
Of course, one trouble with reviewing recent history is not keeping up with the current Jones developments - but you did note those areas in the review, good on you. Otherwise there'd be a whole lot more critical comments already, heh. I like a review written thoughtfully enough to proactively address potential relevant criticism. Shows a good audience-model.
>This artificial scientific consensus was then picked up by tech companies, who used it to label discussion of the lab leak hypothesis as “misinformation”, as well as by media sources and fact checkers. I don’t want to get bogged down in all the convoluted details [...]
This was the most disappointing part for me; I'm unclear whether this was simply not a focus of the book at all, or just the author deciding not to review these (admittedly very-CW) aspects, but it's consistently been *the* most interesting part of the entire Covid Show. Watching - well, living through, in real time - the genesis of a Narrative, impervious to Facts or Criticism, the abuse of "fact checking", the new slur of "misinformation"...again and again and again, with so many people concluding This Is Fine, if they even noticed the smoke at all. Even a few in the Rationalist community, who did not call foul that this is not, in fact, the Way. Partisanship comes even to those who claim to be above it, when the stakes are high enough.
In other words, as some others have already commented, it doesn't really affect my life that much whether the origin is ultimately zoonotic or lab...but the erosion of institutional trust has had extreme ramifications, for me personally and also society-wide. I would still like an "autopsy" report on this failure-cascade at some point. The fact that it sure looks like these patterns are being repeated with monkeypox is not at all reassuring.
That being said, this was a well-written, comprehensive, factual, and even-handed (afaict, not an expert) summary of The Story So Far. Just a pretty solid book review, without any irritating writerly tics, needs-an-editor jankiness, or flat-falling humour. (It is entirely possible to write humour-lite without being academically dry!) I also very much appreciated the epistemic humility of admitting I Don't Know, especially coming from a somewhat-SME. There's a time and a place for big bold postulates and wild reaching claims, but I think still-ongoing history isn't one of those times and places. Can't remember who said it first, but if we know anything about covid, "covid is where intuition goes to die". When constant surprises are the name of the game, it's probably ideal-payoff to hedge on uncertainty, rather than double down or appeal to experts who might well turn out quite biased. Good job, I now have 3 potential finalists to decide my vote between.
If someone withholds, alters, or destroys evidence, as a matter of science, we don't know what the evidence is and can't update our theory. As you said, maybe the person destroys evidence for other reasons.
But in the law, which is the search for truth in messy human affairs, there is a rebuttable presumption that the evidence is bad for the person who did the destroying (it's called "spoliation" if you want to look it up).
If they also offer up an explanation that turns out to be false, the inference becomes stronger.
This approach could lead to the wrong answer, so it should be tempered, but shouldn't we shift the burden this way?
Quick point. The lab-leak theory isn't a conspiracy theory. By definition, an accident isn't a conspiracy.
Deliberately covering it up would be a conspiracy.
Good review and thoughts. On the question of rationality and how to update priors, it seems like there's a missing admission here that your priors were wrong at the start, and although you updated on this specific issue you haven't done any updating of the original bad priors that led you astray.
By the way: "When people talk about trusting the experts, I think they mean trusting people with technical expertise over people without technical expertise"
I don't think that's what it means. When people talk about trusting the "experts" they're using the word "expert" in a very limited way to mean, essentially, public sector bureaucrats or academics. Obviously the two groups overlap significantly. Any people with expertise that is NOT one of those groups, or which contradicts them, is simply dismissed immediately. Thus we can see that exhortions to "trust experts" or "follow the science" are in reality just leftist demands to obey public sector pseudo-intellectuals regardless of merit; the same problem that leftism has had since it first emerged.
An example of this is when it was discovered early on in the pandemic that epidemiological models were never validated against reality, and the Prof Ferguson Imperial College model was filled with programming errors. Or phrased another way, that academic epidemiology appeared to lack technical expertise. Many professional modellers and programmers from the private sector stated flat out that the Ferguson model was so incompetently made that it was broken to the point of worthlessness, creating entirely different predictions every time it was run even when all inputs were supposed to be controlled. But these claims were completely ignored by the media, government and academia, and in fact were swiftly suppressed via (yet another) academic conspiracy in which ICL found a corrupt academic at Cambridge to claim that he'd tested the model and found the outputs to be replicable. In fact the report he issued (never peer reviewed or submitted to any journal) stated that none of the results he got were the same but he considered the model replicable because the "trends" were the same.
Part of why this happens is that academics are able to talk to the press very freely. Every other kind of expertise works for organizations that long ago learned that the press aren't really trustworthy, and so are banned from talking to the media or can only do so after their statements are vetted very carefully. So journalists end up relying on academics to rapidly provide quotes on any topic, and lend their stories the aura of gravitas and specialist knowledge, but people see through it and so trust in the media falls. Falling trust in science itself is lagging behind but now catching up as people realize that when scientists claim the media is distorting what they said, it's not really true and the root source of the false claims is the academics themselves.
It seems to me that the old saying about being "unable to see the wood for the trees" might apply here. A novel virus arise thousands of kilometers from its natural home in Yunnan. But just a few hundred meters from an institute of virology that just happens to study that kind of virus, that has brought hundreds of samples back from Yunnan, one of which is a close genetic match, and which happens to conduct research into gain of function which includes changing the viral genome to make it better at infecting humans, including inserting an FCS, which is not found in other coronaviruses? And did so with grants from an American organisation that led the crusade in advance of any investigation to declare the lab-leak hypothesis a conspiracy theory?
Now we can wait for a full investigation (which we all know will never be permitted by the Chinese government), update our priors all day long, and avoid theorising in advance of the evidence. But if I asked you to bet on one side or the other here, would you really bet that it's all a string of the most unlikely co-incidences? I'm not a gambling man, but I can see the odds here. Those aren't just trees: they are together, a wood.
Are you familiar with Dan Sirotkin's work? Here's his paper from August 2020 about the virus being possibly man-made:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7435492/
Excellent post. A lot of people in the comments are using the phrase "conspiracy theory" to describe the lab leak and cover-up, but to me that doesn't require much conspiracy. And if somebody came up to me wearing a tin-foil hat talking about an accidental lab leak, I would think "you need more interesting sources".
Actually I'm a bit surprised nobody here has posted a real tin-foil hat conspiracy theory about COVID origins, so in the interest of completeness here's one: what if COVID was deliberately released in Wuhan by a non-chinese malicious actor, with the intention of making it look like a lab leak from the WIV? If somebody did want to release a bio-weapon, Wuhan would be a natural choice because the WIV provides such a convenient scapegoat. With the virus origin unknown, the CCP would likely still engage in the same censorship we saw. And the more people argue about lab leak vs natural origin, the more our hypothetical bio-terrorist would be able to slink back into the shadows.
I don't know how likely that scenario is - more than 5% chance? - but it's worth keeping in mind that the space of possible origins is large, and P(lab leak) + P(natural origin) adds up to less than 1.
> Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published. [...] It’s completely unacceptable that BANAL-52 wasn’t mentioned.
Two months before publication mean almost certainly that the book was already off to the printer. Sorry, but production times in the book publication industry are usually quite long, and it's completely unreasonable to expect consideration of research that comes in so shortly before publication.
There was also this Covid lab leak in Taiwan in 2021
https://www.theguardian.com/world/2021/dec/10/mouse-bite-infected-taiwan-lab-woker-covid
Methinks you characterize the claims of "conspiracy" too broadly.
Associated with the "lab leak" rhetoric was often the implication or outright claim that a leak was deliberate.. Which in turn was often associated with "big Pharma" conspiracy-mongering.
That was part of what the "conspiracy" claims were directed at, and often there was a spillover effect, where more limited arguments about an unintentional lab leak were mixed in with arguments about a deliberate one.
I think it's also important to discuss the importance of not alienating scientists in China, given the large risks associated with virus spread from China and in particular because of the huge animal trade there. Many assume that pushback against the lab leak argument was merely because of CYA, and I think for much of the scientific community much of the concern was legitimately because of the need to maintain strong institutional collaboration.
I saw a talk by a member of the WHO team in the early days of the pandemic. They were definitely extremely frustrated about not getting access to any actual evidence, and I think some of the premature definiteness of the WHO report can be put down to "we have to say something/absence of evidence is evidence of absence" type failures.
That said, one point that the WHO team member brought up that I haven't seen discussed much is that there are biomarkers for viruses raised in labs that aren't just biomarkers for genetic manipulation. In particular, viruses kept alive in cells in petri dishes often have adaptations specific to fitness in that environment, which can stick around when the virus makes the leap to humans, and COVID-19 lacks those biomarkers.
Since this was pretty early in the pandemic, I don't know whether this argument has since been refuted, though. Anyone know more?
> "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?"
I'm not an expert by any means, but this looks like a standard sample bias issue. Most Chinese caves are not sampled in search of coronaviruses. This was, but only because of the unlikely event of miners getting sick. If natural reservoirs for viruses are not uniformly sampled, closely related viruses aren't going to show up where they are, but where research has been conducted. Probably this have been said already.
Regarding the issue of the virus being first detected in Wuhan. I wouldn't be so sure as to consider this as unlikely or suspicious. IIRC, for unexposed populations SARS-CoV-2 had an R0 of about three, but most people won't transmit the virus to 3 new hosts, in fact the mode was closer to 0 and the mean was driven up by a small number of superspreader events. The virus may have self extinguished in the countryside and only picked up momentum in a large hub like Wuhan.
Even if there was an earlier or simultaneous rural outbreak outside Wuhan, the combination of asymptomatic/mild cases with the virus causing symptoms similar to other respiratory viruses would have made it more likely to be discovered, as a new illness, in Wuhan.
P.S.: Great review, one of my favorites so far!
I’m not even close to having a working knowledge of virology let alone and expert, and please correct me if I’m wrong and I’ll delete the comment. But I believe you grossed over a pretty important aspect of the “cleavage site” in that there may have been a grant that was rejected for the WIV to research inserting the cleavage sites into bat corona viruses in the couple of years prior to the outbreak. The idea seems to be that they went ahead with this research anyway, otherwise it would be an insane coincidence.
I believe it was Alina Chan who said something along the lines of “imagine if there was a rejected proposal in a city to put horns on horses, then a year later unicorns started showing up right down the street from the lab that was supposed to be doing this research. It could be a strange coincidence, or more likely they did the research anyway.” Since the “cleavage site” had never been found in nature the comparison isn’t much of a stretch.
Did I hear this correctly or am I talking out of my ass? Lol I believe it was on Coleman Hughes podcast over a year ago and I haven’t revisited this topic.
"Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published"
This seems like a weak criticism. Publishing involves a lot of lead time devoted to copy editing, proofreading, typesetting, book design, manufacturing, etc. Even if the authors became aware of BANAL-52 before their book was officially released, it may have been too late to correct the first edition.
Re the location of the WIV, yes, it cannot help but be highly suspicious to any sensible person that the first outbreak occurred in the same city as a major research lab that is known to work with exactly this type of virus. However, I've seen statements in print that went too far with this idea, claiming that the lab is literally across the street from the Huanan Seafood Market. It isn't. They're nearly 8 miles apart, on opposite sides of the Yangtze River. Now, 8 miles, admittedly, is not that much, certainly not enough to discredit the lab leak hypothesis, but it's a lot more than "across the street".
This excellent review has caused me revisit how I formed my beliefs about COVID’s origins in the early days of the pandemic.
I now see that i dismissed the lab-origin theory not because I had evaluated the evidence, but because the Trumpers were pushing it.
Disbelieving something solely because you don’t like those who claim to believe it is an easy error to make. In this case, I assumed the Trumpers didn’t really believe the lab-origin theory either but claimed to because it gave Trump the opportunity to sound tough on China and to undercut the science community, both things he and his tribe were particularly keen to do. For that reason, I dismissed it.
I still think I’m right about why the Trump tribe embraced the lab-origin theory, but I was wrong to dismiss it solely for that reason.
>"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 ... "
The paperback edition's epilogue does discuss the preprints. (It isn't able to discuss the first two's final published versions, out this week, though.)
The review says:
<quote>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.</quote>
My understanding of book publishing implies that this criticism is completely off base. The book publishing pipeline does not permit rewriting two months before publication. The period from the text being locked in until the time the book is in customers’ hands is more like four months.
Epistemic status: I’ve dealt with publishers for years, from the author side, and make tens of thousands per year from book royalties. But I’ve never tried a last-minute change to a book, so maybe I’m wrong. I welcome correction from an actual publisher.
Pentagon collected the closest Laos Banal samples, in 2017
https://web.archive.org/web/20181126230256/https://www.pasteur.la/project-carried-on-in-the-lab12/assessment-of-the-potential-vector-threat-of-bat-borne-pathogens-and-the-host-associated-ectoparasites-in-the-provinces-of-vientiane-and-khammouane-of-the-lao-pdr/
site 1 is exact same GPS
https://www.researchsquare.com/article/rs-871965/v1
> To say there was a lab origin, we would have to postulate that scientific institutions in China are lying and successfully engaged in a coverup, for which there have been no credible whistleblowers.
This is clearly not true; many people believe that the virus originated in a US lab.
I was deep into Wuhan lab leak in ~March 2020 till June- July 2020.
What amazed me at the time was how quickly it became political.
Most of the D's thought it was nuts and most of the R's thought it was true.
(1. how can you all be so sure... it became a tribal signal/ flag...)
Anyway I mostly wanted to put this link here.
https://project-evidence.github.io/
(So I could find it again.)
When lab leak came around again in 2021, it was mostly a yawn for me.
The truth has been hidden in China, and who knows if we'll ever learn the truth.
I would totally bet on lab leak, but have no way to prove it.
And has anyone made a Mojiang mine, minecraft reference? Next update needs to infect the bats in the caves with a virus!
The EcoHealth / Daszak saga gets slightly more interesting: there was a DARPA grant proposal by his org to add a cleavage site to a coronavirus sampled by WIV (see https://theintercept.com/2021/09/23/coronavirus-research-grant-darpa/) which wasn't funded.
But a WIV PhD thesis seems to have confirmed the research went ahead anyway and in BSL-2 conditions no less (https://www.minervanett.no/china-drastic-sars-cov-2/chinese-researchers-created-new-corona-viruses-under-unsafe-conditions/381476), which would make a lab leak fairly likely. This PhD thesis was by Lei Ping Zeng and was also found by Seeker on Twitter.
I haven't read either document deeply but they seem to at least corroborate that dodgy gain of function research creating similar viruses happened in unsafe conditions at WIV around the same time. Whether this was the source of COVID I don't know, but it's alarming either way.
On the subject of things published too late to make it into this book,
https://assets.researchsquare.com/files/rs-1330800/v1_covered.pdf?c=1644271052
Hungarian scientists studying soil samples collected in Antarctica in 2018, belatedly noticed that they contained traces of a very early strain of SARS CoV-2. Which is ridiculously implausible. So what really happened is, they sent their samples to Sangon Biotech in Shanghai, China for phylogenetic analysis, and Sangon used a technique that is notoriously susceptible to cross-contamination from other samples being sequenced in parallel. Clearly someone else had sent actual SARS CoV-2 samples to Sangon for sequencing at the same time.
The Hungarians sent their samples to Sangon in December 2019. Exact date unknown, but almost certainly nobody was sending samples from the Wuhan outbreak to Sangon in 2019. The first report of genetic sequencing of suspected Covid-19 was 24 December 2019, but that was by Vision Medicals in Guangzhou, which did a lot of the (known) early Covid sequencing. Wikipedia also says that Beijing Boao Medical Laboratory sequenced Covid-19 on 27 December 2019. The Wuhan Central Hospital doesn't seem to have used Sangon in this area, and they didn't tell the Chinese equivalent of the CDC about any of this until 29 December 2019 so probably nobody else was doing so in 2019 either.
The Wuhan Institute of Virology had been using Sangon Biotech as a contract lab earlier in 2019. For research on bat coronaviruses, even, if I read this correctly: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6521148/
Furthermore, the contamination in the Hungarian samples included not just SARS CoV-2, but also Chinese hamster and Vero E6 green monkey genetic material. The Dazhak/EcoHealth proposal for gain-of-function research in bat coronaviruses at WIV, called for culturing in Chinese hamster and Vero E6 green monkey cell lines. I'm not sure, but I don't think that's standard procedure for "we found this gunk in the lungs of a mystery pneumonia patient; can you tell us what it is?"
This isn't quite smoking-gun proof that WIV was working with proto-Covid shortly before the known outbreak, but that's certainly a plausible interpretation.
Alternately, the Hungarians could be simply wrong. Their research by nature can't be replicated, but it could at least be closely scrutinized by outside experts, and I don't think really has been. I'd expect Sangon Biotech could confirm or deny that the sequences in the Hungarian paper are the ones they sent them, which should rule out "the Hungarians are just plain lying"
Also, it's barely possible to massage the timeline so that the Hungarian samples are delayed in processing to January 2020, and that the WIV or someone else in their early-2020 investigation of a natural COVID outbreak rushed their own samples to Shanghai and jumped the queue to get in the same run as the earlier Hungarian samples. But that would require that WIV do a hasty investigation that involves culturing their unknown virus in green monkey and hamster cells (why?), or that the Hungarian samples have been doubly contaminated, first with SARS CoV-2 and then from some completely different source that coincidentally used the same culture media as the proposed WIV GoF experiments.
Not smoking-gun proof, but a strong hint with some good pointers to where one might find smoking-gun proof. But as far as I know, there's been no real follow-up.
I did a rudimentary investigation a year ago, finding that most of the purported evidence for a lab leak falls apart: https://medium.com/@pseudodionysus/did-covid-19-come-from-a-lab-a-critical-examination-208f0eff7c3
To repeat only a few parts relevant to the book review:
"Probably the most important – the WIV had previously maintained a database of at least fifteen thousand bat samples, including the dates and locations of samples as well as information about the viruses found in them. This database was taken offline and its contents have not been shared with independent researchers since."
That database was only really online for a few weeks in August-September 2019, and 5 days in December. This claim came from the DRASTIC group of freelance researchers, who cite the archives of a database monitoring system to claim that the database went offline on September 12, 2019. But the same monitoring system shows that the database only became reliably available on August 23, 2019. Before that, from June to August 23, the database was online sporadically. It would be online for 1–2 days, and then go down for several days or weeks at a time. Before June 2019, there’s no data, because v2 of the database was released in June 2019. After September 2019, it was online from December 12 to December 16, and access was sporadic through February 26, 2020. After that, the database was down for good.
"So it does seem like a pretty remarkable coincidence that the outbreak happened in Wuhan. "
Then it's a far, far more remarkable coincidence that most of the earliest cases have direct ties to the wet market. Out of all possible locations, SARS-CoV-2 just happened to pick a wet market as its first superspreader location--just like its cousin SARS-CoV-1--and it has nothing to do with the animals there? It didn't spread at a cinema, a school, a convention center, a train station, a mall, a grocery store, a university, an office building, a bar, or even the Wuhan Institute of Virology (which is 15 km from the wet market), but at a wet market, and this is totally a coincidence?
"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? "
If SARS-CoV-1 originated in an animal somewhere around the Yunnan area--which we now know it did--how did it make all the way to Guangdong without leaving a trail along the way? And of course, though it's plausible that SARS-CoV-2 originated near Yunnan, we don't know that.
In fact, it's very easy to imagine how both of these viruses could travel thousands of kilometers without leaving a trail along the way. In Yunnan, a villager traps a wild animal and gets infected with the virus. He takes a train to Wuhan--one of China's biggest cities and busiest transport hubs--and sells the animal to the wet market, infecting a few wet market retailers and customers along the way. The animal is sold, butchered, and eaten, but the infected humans start the pandemic.
"Some scientists claim that SARS-CoV-2 reached genetic stability early on, suggesting that it was already well-adapted to spread in humans at the start of the outbreak. Some have interpreted this as evidence that it was engineered for this purpose, or underwent serial passaging to encourage adaptation to human or humanized cells. "
The original virus was moderately transmissible, with R0 of 2-3. Since then, a series of variants have taken over the original strain, making the virus more and more transmissible. Delta had R0=5.1, Omicron BA.1 had R0=9.5, and BA.2, BA.4, and BA.5 are more infectious still. SARS-CoV-2 is now one of the most transmissible diseases in human history. So in what possible way has the virus "reached genetic stability early on", or "was already well-adapted to spread in humans at the start of the outbreak"?
Why have we not discovered checkouts or snapshots from the missing WIV database(s) on someone's lab server or laptop?
This is Alina Chan, one of the co-authors of the book VIRAL: The Search for the Origin of Covid-19.
I really enjoyed reading this review and the comments, and I also appreciate the reviewer creating a list of other sources to check out. In particular, this line by the reviewer characterizes many of the moments I've experienced in the past 2 years spent looking into the origin of Covid-19 : “what the hell is going on if some random Twitter users are consistently correcting world-renowned virology institutes on various mistakes and omissions?”
Similar to the reviewer, Matt and I were also late to the lab origin hypothesis. The early information coming out of China seemed to point to a repeat of the 2003 SARS outbreak in China, and I didn't question this, assuming that they would very quickly find the infected animal source as they had in the case of SARS and MERS (I also recommend Laura Kahn's piece in the Bulletin Atomic). It was only later, as key evidence failed to turn up for a market origin, that I was motivated to consider the lab leak hypothesis. By May 2020, the Chinese CDC director had told the press that the market was most likely a victim, ie a sick person had ignited a superspreader event there (the CCDC director recently retired from his position, some speculate it was because he was a "loose cannon" and said too many things about the origin of Covid-19 and China's vaccines that didn't adhere to the official Party line).
One observation that raised my eyebrows (but is still consistent with both natural and lab origin hypotheses) was the highly transmissible nature of the virus right from the get go. And I wasn't the only scientist who was struck by the well adapted/pre-adapted nature of the virus. The natural origin proponents who wrote Proximal Origin in early 2020 were among the first to describe the virus as well adapted for human transmission. They later reversed their stance in their Critical Review in mid 2021. Now, in their latest preprints/papers in Science in mid 2022, they've gone back to saying that the virus needed no adaptive mutations to spread in humans. See my twitter thread for more details: https://twitter.com/Ayjchan/status/1513586519159697410
I also liked the coin flip analogy by the reviewer. I've been accurately described as a fence-sitter in the past 2 years. Back in 2020 it was very controversial for a scientist to be a fence-sitter on the origin of Covid-19 issue (and I took a lot of heat from other scientists, the press, and even friends for just saying we have to consider a lab origin, regardless of how likely or unlikely), but now it's become too boring for the media to report that a scientist thinks there's a 50:50 chance of a market vs lab origin. The media prefers to report on scientists who are willing to go so far as to say they can put one hypothesis to rest or that they have "dispositive" or "incontrovertible" evidence of one hypothesis. To do so is, imo, more like a form of gambling rather than science. Like the reviewer argues, whether it turns out that your bet was right or wrong, it doesn't justify shutting down one of the two plausible hypotheses.
I'd like to address the unavoidable weakness of the book raised by the reviewer: The manuscript is "locked" at least a couple of months before physical copies of it magically appear on bookshelves across the US and UK (not being sarcastic, as a first time book author, this truly was a magical experience for me!). This means that any developments in the 2 months before the book is released cannot possibly be written into the book unless the authors had insider information and knew that it was ok to write said information into the book (which then has to spend time being vetted by the publisher's lawyers). For this reason, the newly released paperbacks cover the new developments since mid-Sep 2021: the Laos BANAL viruses, the 3 preprints on the Wuhan market, and more FOIA'ed emails from prominent virologists who privately fretted about a lab origin of Covid-19. If the publishers will allow, I'd like to update the book 5 years from now because I'm optimistic about very compelling evidence being unearthed within this time range (at least this evidence should be more compelling than Western virologists misinterpreting data collected with ascertainment bias).
Thanks again for the review! I'm providing these links in case other readers are interested.
My peer reviewed analysis of the furin cleavage site:
https://academic.oup.com/mbe/article/39/1/msab327/6426085
My latest summary of the evidence surrounding the origin of Covid-19:
https://ayjchan.medium.com/the-evidence-for-a-natural-vs-lab-origin-of-covid-19-6946407f3310
And my critique of the recent Worobey et al. Science paper, which the reviewer has already linked to - I will update this in a few days to explain the "2 strain 2 spillover" hypothesis that the authors have come up with to support the market origin hypothesis:
https://ayjchan.medium.com/evidence-for-a-natural-origin-of-covid-19-no-longer-dispositive-after-scientific-peer-review-af95b52499e1
The UPDATED paperback is out right now. - I seem to be just the 2nd to note - And Matt Ridley IS updating on his blog "rationaloptimist.com" . So to mimic the review: This review was published July 30, nearly three days after the updated paperback was published. Although I’m writing an overall-positive review here, I don’t want to go easy on the review where serious criticism is warranted. It’s completely unacceptable that the updated paperback wasn’t mentioned!* -
I am very glad to finally see the book mentioned in ACX. I found it irritating to be ignored by the "rationalist" blogosphere (except Scott Aaronson). It seems like a rather decent review too. So, many thanks, really appreciated! -
*(Original: "Why 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!")
It’s easy to dismiss the ridiculous claim that COVID began as a bioweapon, but other claims are more difficult to evaluate. <- I don't see where this claim was engaged with. Why is it easy to dismiss this claim?
>the easy option is to just become an insane person
Love this!
2018 covered up denied DARPA grant seems pretty incriminating, is that not talked about in the book?
"experts make decisions based on ass-covering, career advancement, and personal risk mitigation. autists make decisions based on spreadsheets. the natural result of this is that experts are useless and autists are good at things that are tractable with spreadsheets"
-- https://rationalists-out-of-context.tumblr.com/post/663451258254999552/
Great review but (and this is arguably the books fault not the review's fault) I'd have loved to have seen some base rate information. For instance, what fraction of unusual or novel viruses (just on some measure) are lab leaks given our best information.
If even the people warning us about lab leaks are going to say that it's 1000 to 1 or something that seems pretty conclusive.
> This is basically how I think we should handle this unwarranted overconfidence from respected institutions – it should decrease our trust in these institutions, but we need to be careful not to start favoring the lab leak hypothesis out of spite. In my opinion this loss of trust should not really affect our view of the object-level question of the virus’s origin at all (although it would be nice to see some of the data being hidden, like that WIV database that was taken offline).
I don't think this amount of compartmentalization is justifiable. Actions of public officials can be important pieces of evidence. They might reflect on the object-level question of whether there was a lab leak, or they might not; in order to find out whether they weigh on the question, we can apply everyone around here's favorite theorem.
P(A|B) = P(A)*P(B|A)/P(B), where A = "lab leak is true" and B = "database deleted". You have some prior credence in the lab leak P(A). Even if that's initially very low, you should ask yourself - if the lab leak were true, would that make it *more likely* or *less likely* that they would delete that database? I think the answer is pretty clearly that this would render the WIV *more likely* to delete the database, so P(B|A) / P(B) is above 1. That means that P(A|B) > P(A), which is to say, learning the database was actually deleted should cause you to update in favor of the lab leak theory.
It's not *conclusive* evidence. It might not even be the strongest evidence out there on the question. But it *is* evidence, and it means these two questions - the one about the origin of COVID, and the one about the trustworthiness of the authorities - can't be fully disentangled. Bayes' Theorem applies here too - why would it not?
Why is it a ridiculous, easy to dismiss claim that it started as a bioweapon?
I don't know anything about the subject, and it sounds ridiculous to me, but on the other hand without any justification it sounds like you're doing the same thing you rightly criticise others for doing with the lab leak hypothesis- dismissing it based on superficial weirdness. I'm sure you're not, but I think it'd be worth briefly explaining why it can be dismissed
Possibly relevant articles:
When Conspiracy Theories are Real: https://questioner.substack.com/p/when-conspiracy-theories-are-real
TrUsT tHe ExPeRts: https://questioner.substack.com/p/trust-the-experts
Can we talk Dr. Limeng Yan for a minute and how she is like Edward Snowden, Rachel Carson, and Bob Woodward rolled into one. I would build a statue to this woman the size of Lady Liberty but I guess I would settle for like a Google Doodle one day?
To be clear, the author, reviewer, and everyone else are all in agreement of certain facts: Tedros, Daszak, and Fauci - and every other academic, public health official and NGO-er - all spouted "no evidence of community spread" (while knowing the opposite to be true), in February 2020 while Yan literally put her life on the line to blow the whistle that this thing was spreading like fire and is the reason all these officials were forced to recant the lie.
Let's see what the reviewer has to say about this doctor: "This is a sad story about a scientist who tried to do the right thing, but ended up intellectually corrupted by forces beyond her control." He writes this about Yan! Whose career is ruined, must be looking over her shoulder everyday for CCP payback, and she's the *one* scientist who had the courage to speak the truth to power make these sacrifices to save millions of lives. Meanwhile if the reviewer ever calls the aforementioned liars "corrupt" or anything half as disparaging, I didn't catch it.
Look, the review itself is excellent, and the reviewer does display a healthy amount of prior updating and introspection into personal bias that were blinding him. Dear reviewer, there is still work to done: If Bob Woodward gives an interview to a left wing source like Mother Jones or worked for left aligned institution like Berkley would you dismiss him as no longer credible? Of course not. The sad fact that shines through every dive into question is that the only journalists or politicians who were ever asking asking questions about this incident are going to be on the right. It was probably the same situation is the 60's for the left looking into ecological concerns of big agriculture or nuclear power. Hopefully one day this balances out, but for now we live in the world that exists.
I enjoyed this review, mostly because it roughly matches my pre-existing beliefs, so of course I'd be pre-disposed to like it. I didn't learn much, but hopefully it will Knock Some Sense Into Those Other EvilWrong People On The Internet.
The one section I found very frustrating was the vague attacks on "public health officials, scientists, journalists, and tech companies". Instead of looking into any specific claims by specific people and assessing how true they are and whether they were justified based on the evidence available at the time and what different actors motivations might have been, you just sort of gesture vaguely about "elites and institutions" and hope the audience will nod along in outrage (which they usually will, since this is a bias that is common in the rat-sphere.) Sometimes it seems like you go to more trouble to understand the action of the Chinese government than Americans!
Review-of-the-review: 8/10
I can't object to anything about this review's conclusions. I find the lab leak theory plausible but not probable. The only clear evidence pointing that direction is the pandemic's Wuhan origin, which would be roughly like a US origin happening to be in Atlanta-- a striking coincidence, but not so unlikely that it demands a causal explanation. And like the reviewer I'm more concerned about institutions not having trustworthy truth-finding processes than about the actual answer they get.
That being said, I'm not satisfied with the reviewer's truth-finding process either. The analysis is a bit too shallow and disorganized to actually reach confident conclusions or contribute to the discussion beyond summarizing the book's points. Nor does it present the evidence in a particularly striking or exciting way, as ACX reviews often do. Also, maybe this is just my filter bubble but I didn't find anything particularly surprising here-- "lab leak has to be acknowledged as a possibility given the evidence" has been well inside my Overton window since the theory was first proposed.
In short, I'm satisfied with this review as far as it goes but it left me wanting considerably more. As always, many thanks for contributing!
"... To say there was a lab origin, we would have to postulate that scientific institutions in China are lying and successfully engaged in a coverup, for which there have been no credible whistleblowers."
Stunning naivete. Countries that maintain gulags where ordinary people are sent to be slaves in the cotton fields, are not to be believed without the weightiest evidence.
Whether it was a lab leak or a virus that jumped out of the endangered species being butchered in a public market without any pretense of sanitation, it makes China look bad. As a propaganda move, the best possibility is to shut up all of the witnesses and destroy all of the evidence, and let the poisonous culture war politics of the US work its magic.
As far as the US is concerned, the real issue is not what happened in China. At best it is item number 10 on the bill of particulars against the regime. The real issue is what Saint Anthony Fauci did in funding research at WIV in apparent violation of Obama administration policy.
"Viral was published November 16, 2021, nearly two months after the discovery of BANAL-52 was published."
Book publishing has a much longer lead time than blogging. My wife just completed the final proof reading of her book in mid July. It will be published in October.
"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."
By way of comparison, Chimpanzees are 98.8% similar to humans and the last common ancestor lived millions of years ago. Dogs are 94% similar. https://razib.substack.com/p/i-wanna-be-like-you
96.8% for members of the same genus doesn't sound very close at all.