Will I agree with Nate Silver's claim about a dishonest COVID origins paper in Nature in Mar2020?
resolved Aug 20



I'll read it later and then research more

Here’s the scandal. In March 2020, a group of scientists ... published a paper in Nature Medicine that seemingly contradicted their true beliefs about COVID’s origins and which they knew to be misleading. The paper, “The proximal origin of SARS-CoV-2”, has been cited more than 5,900 times and was enormously influential in shaping the debate about the origins of COVID-19.

In the Slack and email messages, the authors worked to manipulate the media narrative about COVID-19’s origins and to ensure that their private uncertainty wasn’t conveyed in conversations with reporters. They also thought they were going to get away with it. “The truth is never going to come out ”, wrote Rambaut in one message

Specifically, this resolves YES if I think it's more than 50% likely and NO if less than 50% likely by usual bayes stuff

I would have guessed such a thing wouldn't have happened a year ago! But there seems to be a lot of evidence, which I will look into.

Get Ṁ600 play money

🏅 Top traders

#NameTotal profit
Sort by:
predicted YES

Very kind but why did you send us money?

predicted NO

because I need those 5 star ratings don't like that I kept peoples' mana locked up for two weeks after the market closed

predicted NO

I'm leaning towards YES because the paper was misleading, even though 90% of the specific ways the articles nate links claims it is misleading are not true, and the only contradiction between the authors' private beliefs and the paper they published hinged on a single word that (as far as I can tell) the reviewers pressured them to add and wasn't in the preprint. But it still does leave out something very important, and is quite misleading as a reuslt. Does that sound reasonable?

I don't think the articles Nate cites, and his by extension, are particularly well-done or accurate, and if I were a liberal partisan I'd say the right-wing media outlets wre coordinating to tell the same lies about noble scientists to defame them. But they weren't, they just converged on misunderstandings that served their interests. Without any particular malice or planning ... hm ... just like the scientists did?

predicted YES

Resolving YES would be in my favor. I do not take issue with that. It does surprise me that you think so many of Nate's specific claims are wrong, and seem to think it's much less of a big deal than he implies. I don't think I care enough to want to carefully check your reasoning though.

predicted NO

I don't think I care enough to want to carefully check your reasoning though.

I barely cared enough, which migt've been why it took so long. (I'll also manalink everyone some interest payments to compensate for that)

It does surprise me that you think so many of Nate's specific claims are wrong

These aren't claims Nate (who I find to generally be accurate) made directly, they're claims he read from articles that had a bunch of out-of-context screenshots of internal communications, and the way in which they were out-of-context was quite complicated.

predicted NO

In journalism, you'll often get reports of things sources said, things sources heard other people said and inferred from them, and so on. These are a big part of how poltiical journalism works. There aren't any source documents to check for consistency there! Doing this kinda makes me wonder how often that kind of reporting is materially misleading. Probably often! I remember a politician who said that the way the news reported on his internal affairs was only somewhat related to how it actually happened.

predicted YES

@jacksonpolack Your assessment is way off. They definitely misled Don McNeil.

Your intention to seek truth is a positive but the way you try to do that using markets is not the right way to do it. Since you have a perverse incentive to come across as "right", it tends to become an exercise in motivated reasoning.

predicted NO

One of the more convincing claims below was that the authors posted comments that appeared neutral between lab-escape and natural origin after the preprint was publicly posted and submitted to nature.

but the preprint doesn't claim a laboratory origin is impossible! it claims a 'laboratory construct' is impossible, which is an entirely different claim that they never contradicted. the preprint is much more limited in scope.

I think the quality of discourse around this paper is very low.

I do think the authors made a mistake by muddling together 'lab escape with selection or engineering' and 'lab escape', and this is, in my opinion, negligence if they didn't know it and dishonest if they did know. And they should know!

But everyone writing articles calling out the covid lies should ... also know that, and they consistently don't.

His testimony is ... mostly reasonable, except for his claim that proximal origins didn't consider an accidental, unmodified lab release. It sort of makes sense in context - his methods can't differentiate lab escape from zoonosis - but it's still weird to omit that.

Another article I read - lots of interesting information, but - if you read it while willing to separate out 'direct engineering (i.e. gene editing)', 'selection through passage', 'acceidental escape of wild virus', and 'zoonosis', and the differences between the original preprint and the final paper - I don't see any explicit contradictions in any of the author's statements. Most of the claimed lies and contradictions conflate the preprint (february) and the final paper (march), or the different kinds of possible escapes (direct editing vs passage vs escape).

this tweet correctly claims the first version of the preprint (which many of these articles claim to contradict the authors' private claims) https://twitter.com/BallouxFrancois/status/1684672499953332224 is quite reasonable and balanced.

Slowly, they accumulated the evidence for their paper, assessing whether or not the virus’s features could be explained through natural evolution. They wrote it, submitted it to the journal Nature. And it was rejected.

The reviewers, they believed, felt they needed to be less equivocal. In the next version, the final version, they were.

“None of this disproves accidental lab infection,” said one virologist, commenting on the key February 24 evidence. What it showed, he said, was merely “that all the steps can occur in nature”. Who that virologist was? Andersen — lead author of Proximal Origin.

But, still, they clearly misrepresented the possibility of accidental lab escsape without passage in the final paper.

predicted NO


predicted NO

just found 35 more pdfs. gonna call it a night i think

predicted NO

im not sure how im gonna resolve this yet, but it's giving me a look into how the 'sausage is made' in regards to news controversies, and making me very nervous about ever trusting a random substack or journalistic again, because it kinda looks like everyone involved is wrong

predicted NO

good article - https://theintercept.com/2023/01/19/covid-origin-nih-emails/

I think it's clear the proximal origins paper unnecessarily dismissed the possibility of a non-cultured lab escape (which would be difficult to distinguish from a natural origin the way they're trying to). Although ... from the emails below, the leaks supposing to show his internal state, Kristian seems to genuinely believe such a non-cultured lab escape is very implausible (imo on confused but not malicious grounds). I'm just not sure it's a misstatement to the level these people are claiming, or that it had either malicious intent or was shaped by a desire to not find a lab leak. but i still have a ton of stuff to go so

predicted NO

in general ill give 500 mana if i evaluate one of these points incorrectly and you convince me of that

predicted NO

The fourth claim by public is, I think, also nonsense?

The scientists attempted to deliberately misdirect a New York Times veteran science journalist, Donald McNeil

They didn't attempt to misdirect him, they attempted to get him to go away. "I'll also be able to deflect requests for a call - and also gives me a get out of jail free card for ignoring ap otential request" - they just don't want to talk to a journalist. That's fine! You cao do that. Their email makes to factual claims: "the data is consistent with a natural scenario" and "inconsistent with a scenario involving deliberate genetic engineering, including a bioweapon". These are both fine claims on their own, and most of the concern otherwise was about uncultured lab-escapes or evolution in culture, not deliberate engineering. This is also consistent with his messages - "bioweapon and engineering totally off the table", is point 1 of his screenshot in the 3rd point below. I think the correct interpretation is that 'deliberate genetic engineering, including a bioweapon' corresponds to 'bioweapon and engineering totally off the table' in point 1 , and 'engineering (basic research)' (which they'd argue contradicts his claim here) is not what he was talking about with 'delibrate genetic engineering'.

Point 5 recapitulates point 3's concern about culture. Point 3 still seems somewhat strong but idk so far.

Point 6 is about mentioning 'higher-ups'? Bosses exist, people with higher status than you exist, whatever.

predicted YES

@jacksonpolack I think you are too biased by your priors to see this clearly. They explicitly mention "can't give him the whole scientific story".

The reason they don't want to talk to him is because they want to avoid facing someone who will ask them tough questions. Do you honestly think going to that lengths to coordinate a response is normal behavior if they are so convinced by their analysis?

There's a reason why only the hardcore partisans on the left are defending those fraudsters.

I have a relevant comment on the origins market - https://manifold.markets/IsaacKing/did-covid19-come-from-a-laboratory#oRzENtYRfthBGMAEwR3V

Did COVID-19 come from a laboratory?
76% chance. This market resolves once we have a definitive answer to this question. (i.e. "I've looked at all notable evidence presented by both sides and have upwards of 98% confidence that a certain conclusion is correct, and it doesn't seem likely that any further relevant evidence will be forthcoming any time soon.") This will likely not occur until many years after Covid is no longer a subject of active political contention, and motivations for various actors to distort or hide inconvenient evidence have died down. I will be conferring with the community extensivly before resolving this market, to ensure I haven't missed anything and aren't being overconfident in one direction or another. As some additional assurance, see @/IsaacKing/will-my-resolution-of-the-covid19-l (For comparison, the level of evidence in favor of anthropogenic climate change would be sufficient, despite the existance of a few doubts here and there.) If we never reach a point where I can safely be that confident either way, it'll remain open indefinitely. (And Manifold lends you your mana back after a few months, so this doesn't negativly impact you.) "Come from a laboratory" includes both an accidental lab leak and an intentional release. It also counts if COVID was found in the wild, taken to a lab for study, and then escaped from that lab without any modification. It just needs to have actually been "in the lab" in a meaningful way. A lab worker who was out collecting samples and got contaminated in the wild doesn't count, but it does count if they got contaminated later from a sample that was supposed to be safely contained. In the event of multiple progenitors, this market resolves YES only if the lab leak was plausibly responsible for the worldwide pandemic. It won't count if the pandemic primarily came from natural sources and then there was also a lab leak that only infected a few people. I won't bet in this market.

I would be happy to answer if you have some particular questions but I think you are too biased and don't seem to have the entire context.

predicted NO

I don't think my priors are the issue here. I don't like scientific reporting in the mainstream media. I don't really like most individual scientific articles, either. And I do think the proximal origins paper was clearly incorrect in dismissing lab escapes.

In general, even if you aren't biased, you'll just come to wrong conclusions a lot, because a lot of things are difficult. It's not ideal to, when you see someone wrong in a way that disagrees with you, conclude they are wrong because they are biased. Fundamental attribution error, or something.

The reason they don't want to talk to him is because they want to avoid facing someone who will ask them tough questions.

It is not misconduct to not want to talk to a journalist. That is fine. You do not have to talk to journalists about a paper before it is published. It's often a bad idea to do that. The message you're referring to was posted on Feb 19, pg 54. Before publication. In messages a few pages before that, he says that they should not talk to the media because they are working on the science and having their work peer reviewed. He refers to the pre-drafted response he used in the past for a Zika paper, which says the exact same thing. I think this is reasonable.

Do you honestly think going to that lengths to coordinate a response is normal behavior if they are so convinced by their analysis?

YES it is normal and good to coordinate responses when you talk to the media. Any media is dangerous, it's like talking to cops. How have they interacted with rationalists? Again, this is independent of whether they did something wrong or not. If they did something wrong, maybe they're hiding it from the NYT, but if they'd act the same way in both situations we can't conclude anything from it.

There's a reason why only the hardcore partisans on the left are defending those fraudsters.

I am not a hardcore left partisan. There are many non-left people who defend them

Also i'm not even halfway through all the stuff yet.

predicted YES

@jacksonpolack I mean they explicitly mention "can't give him the whole scientific story".

And on Feb 20, Andrew Rambaut says "I literally swivel day by day thinking it is a lab escape or natural".

It is reasonable to craft responses, but their responses were not what they believed in.


predicted NO

I am pretty sure that "can't give him the whole scientific story" means something like "i don't want to tell him what's probably going to be in our paper because he'll ask follow-up questions we don't want to answer yet because it's still in progress", because the context is him saying 5 other things about making him go away.

It is possible I am wrong, it is difficult to interpret things like this. I do not think that your '1+1=2' comment is quite right, this stuff is complicated, it is notoriously difficult to understand precisely why people did complicated things in contested situations, especially at a first look when you don't have much context.

predicted YES


i don't want to tell him what's probably going to be in our paper because he'll ask follow-up questions"

Incorrect. Their preprint was already out by then on Feb 17. They were too hesitant to discuss their own arguments that were already out in the public. Understandable given they were swiveling between lab and natural on Feb 20 AFTER their preprint was out.

predicted YES


Point 6 is about mentioning 'higher-ups'? Bosses exist, people with higher status than you exist, whatever

That's not what they have said in public all these years though. They've tried to portray the analysis as something that they have done independently and have explicitly denied any political pressure. Again, you clearly lack the context to analyze this. What you are doing is jumping on points at a superficial level to justify your existing priors.

predicted NO

addressed this above, the preprint doesn't make the same claims the final paper made, it doesn't contradict any of that

predicted NO

ak, in regards to your review - I don't know what to tell you? I'm not both a virologist and an investigative journalist. There's a very good chance I'm wrong in some significant way here, but it's the conclusion I've come to, and I don't think I have any better options here.

The market is about "will I agree with nate silver's claim". I'm just, like, a random person who spent four hours reading about this. There's no counterfactual here where I have a virology degree and spend fifty hours on this, that's not happening.

Fundamentally, reasonable people can disagree on things without there being some bias behind it. Reasonable people (i.e. me) can also be catastrophically wrong just by making mistakes, sometimes.

Market creator lacks the context to evaluate the evidence and doesn't go deeper into the context despite being told what to look for. Says he will award a bounty for showing he is incorrect but then doesn't engage meaningfully when shown he is incorrect. He also seems to overestimate his ability to understand the context in a short amount of time despite being shown multiple times that he is missing context.

I substantively responded to many of your points, which I didn't find particularly convincing on their own. I stopped substantively responding because I only have four hours to spend on this, not fifty. I use manifold for fun, and while I enjoyed that four hours of reading, I wouldn't fifty.

If I had resolved this to 'YES' with no comment besides a single two-sentence summary, you'd probably have no issue with it.

And because of having limited context, he is under the wrong impression that his assessment is more accurate than that of other journalists/commentators who have spent much more time on this.

Look, if I was writing a journal paper, I'd agree that my assessment in this market was not strong enough. But this is a manifold market for twenty people, and it's a market about my view. Nothing in the description indicated I'd do anything more than read for 20 minutes. I think I've identified significant problems with the way various substacks have interpreted this. Nevertheless, those problems don't prevent the original paper authors from being misleading, which they clearly were.

Since the creator often creates similar markets, I would be wary of betting against him. These markets are mostly an exercise in motivated reasoning to justify the creator's first impressions.

I don't know what to say here. My first impression was that I was probably going to resolve this no, and you accused me of being biased towards no. I then ... changed my mind as I looked at the evidence, and resolved it yes. You want me to defer to "journalists/commentators who have spent much more time on this" ... so ... the experts? Wait, no, not those experts, just the ones you agree with. Seriously, there are a number of commentators who have spent time on this who think no misconduct happened, how does this even work?

predicted NO

Scientists thought lab leak was possible months after saying otherwise 

The messages reveal that Andersen still suspected that a lab leak was possible in mid-April, a full month after Nature Medicine officially published “Proximal Origin,” and two months after the authors published a preprint. “I’m still not fully convinced that no culture was involved,” Andersen wrote to his co-authors on April 16. “We also can't fully rule out engineering (for basic research).” As we noted on Tuesday, if Andersen wasn’t convinced that no culturing was possible, why did he rule out “any type of laboratory-based scenario” in his paper?

This is worse!

Full context from the paper:

The genomic features described here may explain in part the infectiousness and transmissibility of SARS-CoV-2 in humans. Although the evidence shows that SARS-CoV-2 is not a purposefully manipulated virus, it is currently impossible to prove or disprove the other theories of its origin described here. However, since we observed all notable SARS-CoV-2 features, including the optimized RBD and polybasic cleavage site, in related coronaviruses in nature, we do not believe that any type of laboratory-based scenario is plausible.

More scientific data could swing the balance of evidence to favor one hypothesis over another. Obtaining related viral sequences from animal sources would be the most definitive way of revealing viral origins

Andersen's full message




Although the evidence shows that SARS-CoV-2 is not a purposefully manipulated virus, it is currently impossible to prove or disprove the other theories of its origin described here ... we do not believe that any type of laboratory-based scenario is plausible.

really entirely incompatible with

But here's the issue - I'm still not fully convinced no culture was involved.

I think they are different enough that if you believed latter posting former is very misleading. But this is weaker than I thought it'd be. This is not "They also thought they were going to get away with it." tier stuff. So far, the only bad thing I've sen's badness hinges on a single word (plausible). If that word was instead 'probable', I think they'd be entirely in the clear. (again, so far, who knows what comes next)

predicted NO

public's second:

Scientists thought lab leak was “Friggin’ likely”

The scientists were far more suspicious of a lab origin than was previously known. The clearest example of this was when Andersen said on February 1, 2020, “I think the main thing still in my mind is that the lab escape version of this is so friggin' likely to have happened because they were already doing this type of work and the molecular data is fully consistent with that scenario.” In fact, the original name of the channel was “project-wuhan_engineering” until February 6, when Andersen changed it to “project-wuhan_pangolin.”

It looks like this claims he thought lab escape was likely at the very start of the project. you are allowed to change your mind when you see new evidence. I'd say this is incredibly weak evidence for yes, but just incredibly weak.

predicted YES

@jacksonpolack Yes, not sure even if it is evidence at all. If there’s similar view before and after the paper, different from the paper, then that would be pretty damning.

predicted YES

@jacksonpolack You need to go a bit deeper than that. They dismiss lab engineering scenario using arguments they don't buy. The reasons they mention for dismissing the engineering scenario are the ones put forward by Ron Fouchier in their Feb 1 teleconference. After the teleconference, they explicitly mention in their Slack channel that Ron Fouchier is too biased to see this clearly (since he is one of the most prominent gain of function researcher).

They are also on record explaining why Fouchier's arguments are weak. Yet they try to mislead NAS on Feb 4 with the exact same arguments.

It's totally reasonable to expect scientists to change their minds but when they claim to do so using arguments they don't buy, that's a big red flag.

predicted NO

can you be specific please, like, links to relevant passages in articles? lots of moving parts here

predicted NO

specifically the reasons they dismiss in slack / on record then try to mislead nas with them, can you link to the excerpts

predicted YES


You can read more here -

Feb 1 teleconference call with Andersen, Garry, Rambaut, Holmes (leaning lab leak) with Drosten, Koopmans, Fouchier (heavily leaning natural). The link below has notes by Ron Fouchier that includes Fouchier's arguments against engineering.

Their discussion about those arguments and why they are not strong are in the Slack - Feb 1 to Feb 3.

Feb 3 is a call with NAS in which Andersen is a participant.

Feb 4 - Andersen tries to mislead NAS (email in the same link below) by suggesting they need to be more "firm" on the engineering scenario. He says data conclusively shows there is no engineering and cites Fouchier's arguments that he himself finds unconvincing days before. The same arguments also appear in Proximal Origins.


predicted NO

There was some new evidence, and engineering and passage are different. I don't think Andersen's denouncing of engineering was dishonest, there was new and reasonable evidence that pointed against engineering. I think it was dishonest that he dismissed lab escape w/o passage by simply ignoring it and saying 'lab scenario not possible'.

predicted YES

@jacksonpolack What new evidence are you referring to? Andersen explicitly mentions the reasons for claiming "data conclusively shows no engineering". None of that is new evidence. Those are just rehashed versions of what Ron Fouchier sent them and he explicitly says in Slack chats that Fouchier is conflicted and his arguments are weak.