Table of Content:
Part1:
1. CONTEXT: The fight for the sequence publication (Zhang, Holmes, Farrar, WHO)
2. CONTEXT: Questions about human transmission (Holmes, Farrar, WHO)
3. Concerns about the sequence (unspecified US scientists, MI5, Farrar, Holmes)
4. The Farrar call
5. Farrar and the ‘core-group’ compare notes and estimates
6. Farrar shares the passaging hypothesis with the ‘Bethesda boys’
7. In Parallel: Trying to get the WHO to pick up the issue
8. CONTEXT: US parallel effort: Track II via NASEM
9. Writing a confidential report while waiting for the WHO
10. Drafts are shared with Fauci and Collins (4 Feb)
11. WHO agrees to get involved (5–7 Feb)
12. ‘Pangomania’ and a fateful decision (7 Feb)
13. Sharing the decision with the Tony Call group and some pushback (8 Feb)
Part 2:
14. CONTEXT: Pressure mounts as NASEM, US Gov and WHO all urgently need access to China
15. Farrar has to keep pushing for publication (8–11 Feb)
16. Working towards publication (10–16 Feb)
17. CONTEXT: Main-street media turn ugly just as the full WHO mission arrives in China (16–17 Feb)
18. Publication of manuscript on virological.org and submission to Nature (15–17 Feb)
19. In parallel: Daszak’s pre-emptive strike (6–18 Feb)
20. The pangolin claims implode (18 Feb–20 Feb)
21. Rejection by Nature (20 Feb)
22. RmYN02 and its consequences (21–25 Feb)
23. End-Game: Publication in Nature Medicine (26 Feb-17 Mar)
24. CONTEXT: White House vs. China Wolf-Warriors (6–17 Mar)
25. The Nature Medicine version
26. Post publication doubts
27. Egomaniac strikes back (Jul 2020)
28. CONTEXT: A long feud and some funding woes
footnotes
Appendix A: The evolution of a scenario
Appendix B: The evolution of a name
Appendix C: Plausible research-related accident scenarios
Appendix D: Virus Genesis Scenarios
Appendix E: Evolutionary vs. Outbreak Origins in P.O.
14. CONTEXT: Pressure mounts as NASEM, US Gov and WHO all urgently need access to China
14.1 Farrar contact Dzau (NASEM) to share notes (7 Feb, 3:06 pm UK)
On 7 Feb, Farrar emailed Victor Dzau (the president of the US National Academy of Medicine), with Fauci and Vallance (who keeps popping up in the key emails to the US side) in CC. Farrar suggested a call so that he may share his analysis of the virus origins. From the email, we can see that it was the first time Farrar mentioned to Dzau the effort he was leading.
14.2 Farrar reiterates the need for sequences from Wuhan (8 Feb, 11:04 am UK)
We do not know when Dzau replied. What we know is that Farrar next sent a tweet, on 8 Feb at 11:04 am UK, in which he insisted on the need for Chinese authorities to release recent SARS-CoV-2 sequences from Wuhan, so that one might better understand how the virus mutated and spread under evolutionary pressure in that hotspot.
He added correctly that no new sequence from Wuhan samples had been published since 4 Jan.
14.3 Tedros sets the tone for the WHO (8 Feb, before 4 pm UK)
During a daily briefing on the novel coronavirus, on 8 Feb 2020, Tedros made some rather surprising statement, with some mention of ‘battling the trolls and conspiracy theorists’. The choice of words was rather out of character for someone who is usually much more careful, but may be related to the fact that the WHO was, at that very time, trying hard to get access to data and to get an advance team admitted in China.
At WHO, we’re not just battling the virus; we’re also battling the trolls and conspiracy theorists that push misinformation and undermine the outbreak response. [..]
We are also engaging with search, social and digital companies such as Facebook, Google, Tencent, Baidu, Twitter, TikTok, Weibo, Pinterest and others. We are asking them to filter out false information and promote accurate information from credible sources like WHO, CDC and others. And we thank them for their efforts so far.
source: Tedros’ remarks at the media briefing on 2019 novel coronavirus on 8 February 2020 (before 4 pm UK)
Insight: The road to hell is paved with good intentions
Dr. Tedros’ strong words reflects the fact that the drafting of P.O., under the direction of Farrar, was being instrumentalised to serve the WHO agenda and Dr. Tedros’ negotiating position with China.
There is absolutely no way Dr. Tedros was going to be able to secure access to China for the WHO, if Farrar was to publish a paper questioning the natural zoonosis hypothesis, in anything but the most ambiguous and anodyne terms. And reciprocally, there may be a way to secure access if Farrar published a piece that denied any possibility of a lab-product, while Dr. Tedros did beat his chest and fustigated some conspiracy theorists, and — incidentally — encouraged a very Chinese form of social media control in the Western world, in the process.
As it is, by 8 Feb 2020, Farrar had committed to deliver the report Dr. Tedros needed for smooth sailing.
14.4 Call between Farrar, Holmes and Dzau (8 Feb, 9.27 pm UK)
The call that Farrar suggested on 7 Feb (see 14.1) took place the following day, 8 Feb at 9:27 pm UK, with Farrar and Holmes on one side. It is not clear who was on the NASEM side with Dzau.
NASEM had already expressed the need for more data in its answer to the OSTP (White House) on 6 Feb (see 8.3.g), and Andersen has also made that point clear, both in his NASEM call on 3 Feb and in the co-authors Slack conversation. We may thus suspect that this point was again discussed during the call with Dzau.
Less than two hours later, Garry and Andersen had a quick chat about it on Slack, with Andersen not exactly bothered by the NASEM initiative, which he considered unlikely to come to any proper fruition, possibly due to his own experience when he took part to the NASEM call of experts on 3 Feb (see 8.3.c).
We must note that at the time of his call with Dzau, Holmes most likely did not yet know that the pango <99%> sequence was badly hyped, since the first time he mentioned that disappointment was about two hours later, at 11:42 pm UK (see 13.14).
14.5 Chinese ambassador pushes back against Tom Cotton (9 Feb, ~4 pm UK)
Sen. Tom Cotton, a Republican who joined the US army after the 9/11 attacks (combat tours in Iraq and Afghanistan), and a member of the Senate Intelligence Committee and Armed Services Committee, had already attracted some criticism for a letter he sent to the US administration on 28 Jan 2020, in which he called for a ban on travellers from China. In the same letter, he:
- asked for China to allow US and WHO experts in, and
- mentioned the possibility of a lab-escape, alongside a natural zoonosis:
Regarding this last point, I ask that the administration urge President Xi to allow American and international scientists and medical experts from the World Health Organization into Wuhan immediately to help treat infected persons and research the origins of this coronavirus.
If the virus didn’t originate in the seafood market, it’s critical to determine where it did — especially, I must add, since Wuhan is also home to a biosafety-level-4 super-laboratory that engages in the study of coronavirus, among other deadly pathogens.
source: letter from Sen. Tom Cotton., 28 Jan 2020
On 3 Feb, during a session of the Senate Armed Service Committee, Sen. Cotton detailed his doubts about the market zoonosis theory, pointing instead at some farm zoonosis, some food processing chain, or a Wuhan laboratory, all as valid theories, and asked again for the US to ban travel from China.
The Lancet published a study last weekend demonstrating that of the original 40 cases, 14 of them had no contact with the seafood market. As one epidemiologist said, ‘that virus went into the seafood market before it came out of it’.
We still don’t know where it originated. Could have been another seafood market, could have been a farm, could have been a food processing company. I would note that Wuhan has China’s only biosafety level 4 super-laboratory that works with the world’s most dangerous pathogens, to include — yes — coronavirus. [..]
As a defensive measure, I just say, again, it is essential that we shut down all commercial air travel, immediately, between the United States and China,
source: Sen. Cotton, session of the Senate Armed Service Committee, 3 Feb 2020.
This was followed by a rebuke from the Chinese ambassador to the US, Cui Tiankai, on CBS ‘Face the Nation’ (Sunday 9 Feb, ~4 pm UK), who denounced these accusations as ‘absolutely crazy’. During that interview, the ambassador was asked repetitively if CDC experts would be allowed in China, and gave a rather elusive answer.
14.6 The WHO advance team leaves for China (9 Feb, ~9 pm UK)
The second week of February 2020 was crucial for the WHO, as it was working hard to get access into China and restore some credibility. The record of the WHO so far was at best mixed: it had mishandled the clear signs of human-to-human transmission for most of January, and had been left looking rather silly after the sudden imposition of a lockdown in Wuhan on 23 Jan 2020, on the same day it decided not to declare a Public Health Emergency of International Concern, before declaring one a week later (30 Jan).
As a result, not only to better understand the epidemiological characteristics of the virus and the best treatment options, but also to try to demonstrate that it could not just be brushed aside any longer by the Chinese authorities (see sections 1 & 2), the WHO urgently needed to send a new mission to China. Barely anybody had taken notice of the first mission on 20–21 Jan (see 2.2), which anyway had contributed to the bungled decision not to declare an PHIEC on 23 Jan. The WHO credibility and direct ability to shape an international response fully depended on a new, larger and longer mission.
Dr. Tedros had flown to Beijing to meet Xi Jinping on 28 Jan 2020. His visit allowed him to extract a vague promise to let WHO experts enter China (see 2.6). After 10 days of negotiations, on the evening of 9 Feb (UK time), a WHO advance team of three people (Maria Van Kerkhove, Dr Jun Xing, and Dr. Bruce Aylward as leader [footnote 38]), left for Beijing. Their immediate aim was to negotiate some Terms of Reference for a joint China-WHO mission in the following days, including objectives, workstreams, method of work and baseline information/data requests. [footnote 96]
On the same day (9 Feb), Tedros tweeted about the departure of the advance team. His tweet was the object of a Reuters news item, which highlighted the quiet diplomacy behind that mission, and the protracted negotiations over the past week, for a group of 15 international experts to be allowed to travel to China.
14.7 The US is getting concerned about access for the coming WHO mission (9 Feb, ~11 pm UK)
In that Reuters news item of 9 Feb, Tedros was quoted as saying that “he hoped that the team would include experts from the U.S. CDC.”, where ‘the team meant’ the full team of around 15 that was to follow the advance team in Beijing. The US CDC had already offered to send people to China in January and had been rebuked — Tedros’ mention of mere ‘hopes’ sounded as if an irresolute WHO may not be successful either. Tedros’ quote thus raised alarms for Grigsby (Director for the Office of Global Affairs at HHS) who forwarded the Reuters piece to Schwartländer (Dr. Tedros’ right arm).
Schwartländer’s particularly evasive answer only increased the concerns of Grigsby, Stew Simonson (Assistant Director-General of the World Health Organization responsible for the WHO Office at the United Nations and the WHO-US Liaison Office, see 4.10 and 7.4) and Fauci himself, who together had been pressing hard to get two US members on the WHO team. Without any fioritura, Schwartländer effectively told the US that it was just another country among others — despite being by far the main monetary contributor to the WHO. Incidentally, that argument would be echoed the very next day (10 Feb) in Chunli Bai’s response to U.S. requests to share isolates (see 8.8).
Schwartländer to Grigsby, 9 Feb:
As you are much aware, the US has given us a number of names who will be able and willing to join such a mission. We have received similar proposals from other countries and will now match the “long list” of experts with the required specific expertise. [..] The overall number will be kept at a level to make sure that the team is fully operational.source: HHS_Garrett-Grigsby_12.30.21_production.pdf, p. 38
That same evasiveness was repeated in a timely interview of Schwartländer’s by James Chau (WHO Goodwill Ambassador, president of the China-US Foundation), published on the same day (9 Feb) in his usual media outlet, most likely as a signal to China (see footnote 81 for the controversial background of James Chau).
It all sounded as if the US were being left in the dark by the WHO, with no guarantee at all that US CDC experts would make it to the list. In the end, on 13 Feb, so only two days before their due departure, Weigong Zhou of the US CDC and Clifford Lane of the NIH (Bethesda) were selected.
Dr F,
I hope that’s not the case. Stew emailed me a few hours ago assuring that Bernard and Tedros are working hard on quiet diplomacy to ensure our folks get in. Who knows what the final number will be, but he Sec [note: securely?] spoke to Tedros the other day and was very firm that Tedros must make sure our folks are let in.source: email from Garrett Grigsby to Fauci, 9 Feb 2020, 18:34 EST
15. Farrar has to keep pushing for publication (8–11 Feb)
15.1 Disappointing environmental samples create doubts about the market story (8–11 Feb, UK)
a. Holmes gets hold of the sequence, Rambaut does a quick analysis (8–9 Feb)
On 8 Feb (11:33 pm UK), Holmes had announced on Slack that the Chinese CDC had 3 environmental samples from the market, and hoped to access them first, maybe before any uploading on GISAID.
Soon after, Holmes must have got hold of these sequences through his Chinese CDC back-channel, since on the following evening (9 Feb, 9:09 pm UK), Rambaut disclosed that he had checked the not-yet public environmental sequences obtained via Holmes.
b. Disappointment all around (9 Feb)
The sequences in these samples appeared identical to the ones of the early Wuhan patients, so the suspicion was that they may have been deposited by humans, not by animals (hence the quotation marks around ‘environmental’ in the Slack message).
Garry wondered whether this was ‘more evidence that the market was not the point source from which the outbreak sprang?’ (9 Feb, 11:18 pm UK). Holmes then jumped in with some rather harsh words: ‘the environmental seqs are spectacularly uninformative. Pretty shocking if this is the best they have.’ (11:37 pm UK), and offered to relay that disappointment (based on Rambaut’s analysis) back to the Chinese CDC, in the hope of convincing them to release more environmental sequences.
Meanwhile, the co-authors’ best hopes laid with the SCAU pangolin sequence, for which they nevertheless harboured modest expectations after the revelation of the <up to> 99%.
Garry (9 Feb, 11:46 pm UK):
Waiting on pango up to 99. I was hoping the environmental samples would help, but the results made me uncomfortable. Afraid Pango99 might not be any more informative either.Holmes (9 Feb, midnight UK):
Andrew, can I pass this info back to China CDC? Hopefully might loosen them to send more data.Rambaut (10 Feb, 0:55 am UK):
Of course!
Andersen only joined the discussion on 10 Feb, after doing his own quick analysis. His conclusion was identical: “the ‘environmental’ samples were entirely uninformative”, ‘Kinda bummed that the ‘environmental’ samples didn’t show anything at all’.
c. Feedback from Drosten (11 Feb)
On 11 Feb, Rambaut met Drosten at the WHO Blueprint meeting in Geneva. [footnote 40], which was also attended by Farrar and Daszak (‘Dastwat’ below). Drosten shared with him his opinion that the virus had been circulating for a while, and thus did not jump at the market, which would be consistent with the environmental sample being virus shed by humans.
Rambaut (11 Feb 2020, 7:05 am UK):
Heading over to WHO now. [..] Hope to have a few minutes to chat with Jeremy too.Holmes (9:37 am UK):
Have fun at WHO. Ask Dastwat about the Guinea Ebola seq[uence]. [..]Rambaut (9:52 am UK):
Had a quick chat with Christian Drosten. He is strongly of the opinion that the virus has adapted in humans. He thinks it has been circulating in some part of China for a while.
This triggered some discussions on Slack about what the molecular clock may reveal. Rambaut, looking again at phylogenetic trees and their rooting, was starting to consider a long pre-detection circulation in Wuhan, which would mean that the market jump hypothesis they were considering would be less likely, against the alternative hypotheses (including passaging).
d. Updating Farrar (11 Feb)
Rambaut was expecting to talk to Farrar the next day at that WHO BluePrint event, and to discuss these latest data-driven findings.
Rambaut (11 Feb 2020, 7:26 pm UK):
Going to chat with Jeremy tomorrow morning.
I am beginning to be more convinced about the mid-point root. I think that means a long pre-detection period in Wuhan (possibly outside). Basically once you lose the market as the origin, all bets are off.Andersen (7:34 pm UK):
Yeah, I think that’s an interesting possibility too Andrew — and the root is definitely challenging. Thing is, given what we’re seeing on the cruise ships, in the hospitals and communities, clearly this thing spreads extremely easily between humans — so as you say, it’s highly plausible that while the market was were it was detected (and potentially amplified) it’s not because of an animal reservoir there, it’s because of extended human-to-human transmission. If you look at the environmental samples they also look like patient samples — which would be consistent in such a scenario.Rambaut (7:46 pm UK):
That is my thought. I suspect the surveillance system picked it up because it was a market — this is essentially an avian influenza surveillance system. But it may have just been spread within the market.
15.2 Koopmans argues that one should not publish anything specific about a lab-escape (9 Feb, 8.07 pm UK)
After she had tweeted on 7 Feb that one should be waiting to see the SCAU pangolin sequence before reaching any conclusion (see 12.1.b), Koopmans did not check her emails on Saturday 8 Feb. Hence, she did not contribute to the unsuccessful pushback on Farrar’s decision to go for publication that took place on that day (see 13).
On Sunday 9 Feb, Koopmans got back into the email discussion. There is no sign that she was by then aware of the revised <up to> 99% homology. That information was held by the co-drafters, and there is no record of them sharing it with the Farrar’s call participants.
In the email she sent that day, her point was not whether one should go for publication or not; she argued that one should not mention specific lab-escape scenarios in the main body of the paper, when going to publication, since such publicity would likely attract detrimental attention, given that the specifics (especially the passaging scenario) were right now not being discussed by the public. Instead, she would rather have any mention of a lab-escape scenario relegated to the discussion section of the paper (the contextualisation part of the paper, or more speculative theories can be discussed).
In effect, she was trying to reconcile the point made by Drosten earlier on in the thread (‘Are we working on debunking our own [passaging] conspiracy theory?’, see 13.7), with the pushback from Holmes and Farrar, who wanted to go for publication and address some specifics.
15.3 Farrar cuts the opposition short and reaffirms his decision (10 Feb, 9:26 am UK):
In his reply to Koopmans, Farrar basically told Fouchier, Koopmans and Drosten that his decision was final, with a polite but firm sentence: ‘Appreciate not everyone will agree on the next plans but the discussion has been very constructive, thank you’.
He then told them that ‘the people who have led the analysis’ (i.e.: the co-authors: Holmes, Andersen, Garry and Rambaut) would take it from there. Fouchier, Koopmans and Drosten had to accept that decision, which they seemingly did.
Essential Insight: Access is all that matters for the WHO and Farrar
Farrar later argued, in his book, that he had to change course and go for publication because the WHO was too busy to pick up the question about the ‘evolutionary origins’ of Covid-19.
That explanation makes unfortunately no sense at all, since Farrar decided to go for publication on 8 Feb, just after the WHO had just given its green light (5 Feb) to discuss how to onboard the question during the WHO Blueprint meeting to be held on 11–12 Feb. On 6 Feb, Farrar had even started feeding names to the WHO, as candidates for a future WHO origin group. So why did Farrar decide to go for publication just 3 days after receiving the WHO green light and only 3 days before that key 11–12 Feb meeting?
The real reason is simple.
The NASEM answer to OSTP sent on 6 Feb had diplomatically not mentioned any specific origin scenario, while not explicitly denying a potential non-natural origin either. It instead put the focus on the importance of getting more data. Then suddenly, on 6 Feb, Fauci was quoted as saying that the US did have the means to determine if the virus was a lab product (which would also include passaging), and that it would be looking into it.
Right at the same time, the WHO was struggling to get its advance team over to China, with intense negotiations still underway, and a clear difference of understanding of what that visit was supposed to be about, between the WHO and Chinese officials (see 2.6.d). So, to help get WHO access to China, Farrar urgently needed a statement that would explicitly address both the passaging and gene editing scenarios, and then reject these, or at least consider them as extremely unlikely. The deck had to be cleared first.
- That was not what Drosten, Koopmans or Fouchier wanted to hear, as such publication would indirectly show the potential danger of the experiments they specialise in, especially passaging in humanised animals / cells.
- That was not either what Andersen wanted to hear, as he rightly considered that the pangolin data was less than convincing, which should not allow closing the passaging possibility (if writing honestly).
But that is what Farrar and Holmes (very involved in the pangolin studies and happy to support Farrar), urgently wanted to hear. Priorities had changed and getting access to the WHO, via a very watered done paper, was all that mattered in the short term for Farrar, whether Andersen and his co-authors fully understood it by that time or not.
From that point on, nothing could change Farrar’s decision to go for publication. Not even Holmes learning of the likely rather low homology of the pangolin sequences (<up to> 99%). In fact, that would only make it even more urgent to close the discussions, as speculations would likely redouble otherwise when the disappointing news of the low homology would be made public.
The idea of writing a non-public report about the possibility of a lab-related accident, sending it over to the WHO for a review with the Chinese side, was de facto dead by 9 Feb.
Additionally, one could argue that any ongoing doubt about the origin would be better dealt with behind the scene (as UK intel had done when asking Farrar to get a quiet evaluation by subject-matter experts), rather than publicly and by risking an open confrontation:
- First, because publicly expressing some doubts about the origin could not force China to cooperate, quite the contrary, and thus would not save lives, while getting WHO access to China may well.
- Secondly, once the deliberate engineering option has been discarded, without that access to basic epidemiological data and early sequences from China, there was — in any case — likely no way to reach any conclusion about the origin. Passaging may not leave an obvious smoking gun (see all the discussions about the glycan shields that went nowhere).
Hence, in the absence of any willingness to confront China based on principles (a willingness always in short supply after SARS-1), some form of cooperation was the only game in town. By leveraging the access concerns, China had won without having to make a concession of its own.
For instance, in an interview with James Chau, on his platform chinacurrent.com, Schwartländer (Tedros’ right arm), never mentioned anything about any enquiry into the origins when describing the coming mission on 9 Feb. His description of the purpose of the mission was purely for the WHO to learn about the outbreak and the way China was fighting it, with Wuhan not being necessarily the best place to do so. This was exactly in line with what Xi Jin-Ping had in mind for the mission when he agreed in principle to that mission during Tedros’ visit to Beijing on 28 Jan (see 2.6). The closest Schwartländer may have gone to the origin subject was when describing the cooperation between the WHO and social media companies about fighting misinformation; he never mentioned the unmentionable.
The timeliness of that interview with Schwartländer strongly suggests that it was intended to deliver a reassuring message to the Chinese government. In fact, James Chau, the host of chinacurrent.com, who has softball-interviewed Tedros and Schwartländer a few times, also happens to be a rather controversial WHO Goodwill Ambassador for Sustainable Development Goals and Health [footnote 81], and one of the founders of the China-United States Exchange Foundation.
Eventually, with the eventual upload of P.O. manuscript on virological.org on 17 Feb (the day after the full WHO team arrived in China), the origin question was for all purposes settled in the public space, whatever doubt intelligence services may have behind the scene — until Australia (and not the US) spoilt that cosy arrangement, by resuscitating the non-natural origin question at the WHO level in May 2020. [footnote 53]
Key Insight: A misunderstanding
Once Farrar confirmed his decision to go for publication to the call group on 10 Feb, the ‘bounce-off’ group of Fouchier, Koopmans and Drosten was dropped from the editing process. There is no more communication by email with them in the FOIs. In contrast, it is worth remembering that Koopmans was the very one asking Farrar to consider going for publication on 4 Feb (see 11.2), when she had not yet read the draft (which was not circulated to the full call group untill 8 Feb, see 13.1).
It is therefore very likely that the Fouchier-Koopmans axis was supportive of going for publication based on their understanding that they had successfully killed the deliberate engineering option during the Farrar call, while not expecting the laboratory hypothesis to be revived through the passaging option.
Once Farrar had decided to go for publication to address the WHO needs for access, the sudden — but predicted — opposition of the Fouchier-Koopmans axis (the ‘krewe’) became a pure inconvenience to him and Andersen. Fouchier, Koopmans, and Drosten had been looped in as a group against which to ‘bounce off’ ideas during the Farrar call, at the suggestion of Fauci. They were not part of Farrar’s trusted core of Holmes, Andersen, Gary, and Rambaut. Once the krewe’s objections had been politely acknowledged by Farrar, their job was done.
With the Fouchier-Koopmans axis (the ’krewe’) apparently out of the way, Andersen was able to keep the passaging option alive in the version uploaded on virological.org, even if de-emphasised to accommodate some of Farrar’s concerns, and likely also to edge his own risk of being proven wrong soon after (potentially a career destroying move) [see notes on 15 Feb (v11) and 17 Feb (v12) versions in Appendix A].
The exclusion of the ‘krewe’ from that point on, down to their non-acknowledgement in the final version of P.O., and the persistence of the passaging option in the virological.org version, against the krewe’s expectations, would be the very motivation for some malicious rear-guard actions. This is key to understanding the later part of the P.O. story.
15.4 Meeting between Baric, Fauci and Auchincloss (11 Feb)
On 11 Feb, 2:20 pm EST [footnote 77], Fauci and Baric had a meeting in the presence of Fauci’s right arm Auchincloss (see 4.3) and of Alan Embry (NIAID/DMID, Head of the Respiratory Diseases Branch), plus other representatives of both the extramural and intramural programs of NIH.
A description of the meeting was given by Baric in his House interview. Fauci was present for only 5–10 minutes, and the focus was on the regulatory compliance for the work done in the Baric-Shi Zhengli chimera paper of 2015, and more generally for that kind of research. That paper, ‘A SARS-like cluster of circulating bat coronaviruses shows potential for human emergence’, was the one that Fauci had asked Auchincloss to review quickly before the Farrar call (see 4.2). It was also the paper that Andersen had noticed, and the first one mentioned in the presentation that he prepared for the call (see 4.6).
A: I met with him [Fauci] in his office with several staff, high level staff, both including himself and other representatives from both the extramural and intramural program for NIH on, I think, February 12, 2020. [..] Auchincloss. Alan Embry [..]
Q: [..] I just want to ask, Dr. Fauci was there at the meeting?
A: He was there for a short period of time. I already mentioned some of the names that were there. So he was there for somewhere between five and ten minutes, at most. And he got — a secretary came in and said that he had a call in the SCIF [note: Sensitive Compartmented Information Facility] that he apparently had to go to, so he apologized. So he wasn’t there for the whole time.
Q: Do you recall, specifically while he was there, what you discussed?
A: [..] Ultimately, the goal of the meeting, to my recollection, was primarily focused on the 2015 paper that we published in Nature Medicine that basically, in my opinion, warned the world that there were viruses that existed in nature that could threaten human health.
And so the first thing they wanted to do was talk about that paper, and then they wanted to talk about the regulatory — the P3C0 regulatory compliance that was associated with that. So we talked about that. And then we talked about the specific experiments that were done, the first of which we
compared the growth of this isolate to the parental virus
that we introduced the spike gene into. And it replicated the same. So from our perspective, in terms of P3CO, that’s not called gain of function, that’s called retention of function, right?[..] So that’s kind of the whole context, that’s kind of — and Fauci left in the early stages of the discussion, right, because that took about 25, 30 minutes. I don’t know how long it took, probably damn long probably.
source: Baric-TI-Transcript.pdf, extracts, p.15. then p.38–39
As Baric alluded to in his interview, that meeting felt ‘damn long’. His gloomy feeling was also confirmed in a Slack message from Matthew Frieman to Vineet Menachery. Frieman (now at the University of Maryland) is a friend of Baric and a former post-doc from his lab, he was also the lead-author one of the 6 papers that Andersen included for review on the Farrar call on 1 Feb (Frieman et al., 2011), while Menachery was the lead-author of another one (Menachery et al., 2016) (see 4.6).
I talked to Ralph for a long time last night. He sounds beat. [..] He said that he sat in Fauci’s office talking about the outbreak and chimeras. Clearly he is in other kind of meetings than what we are invited to! I joked about his link to the WIV, he wasn’t amused. [..]
source: Slack message from Matthew Frieman to Vineet Menachery, 18 Feb 2020
16. Working towards publication (10–16 Feb)
16.1 Lipkin joins the authoring group (10–12 Feb):
a. Lipkin gets in touch with Holmes (10 Feb 7:11 pm UK)
On 9 Feb (5:47 pm UK), Ian Lipkin contacted Holmes, asking him to call him back. After some delays due to a violent storm in Sydney, on 10 Feb (~7:11 pm UK), Lipkin managed to talk to Holmes and explained that he was concerned about the Furin Cleavage Site and the restriction site [footnote 42] in the recently-released sequence, as were also ‘higher-ups’ and intelligence people.
His intervention was timely. For context, Lipkin has been quite open about having regularly provided information to the US intel community over his years spent chasing epidemics abroad, which is actually not to be unexpected. He has also been open about having regular links with the intelligence community after the outbreak started, for example by meeting two FBI agents on a quarterly basis, or being in regular touch with the CIA.
Lipkin explained to Holmes that his value may be in his ability to understand how intelligence services work, and the type of evidence they consider, which should really be part of any assessment of the origins.
My concern is that as one of the only — to the best of my knowledge, I’m the only person who works in public health and infectious diseases who has received this honor from China — two different honors from China. This gives me access to information and insights that nobody else can transmit, to NIH and CDC and other policymakers, including, I would say, people in the Department of Defense and the intelligence community with whom I share these kinds of data over a period of almost 25 years.
And my concern about this making public is that not only will this impact my ability to get information from China, but other people will say, whether it’s India or, you know, or Argentina — doesn’t make any difference — I will not be trusted anymore.
source: 2023.04.06-Lipkin-Transcript.pdf, p.47
Q: This is a more general question, not linked to any particular document. Is it important, from your point of view, that inquiry into the origins of the coronavirus or SARS-CoV-2 be guided by scientific principles of inquiry?
A: That’s an interesting question. You asked earlier about, you know, whether or not I had been trained in looking at certain types of evidence, and I believe that I am limited in the types of evidence that I’m able to consider. And I think that the intelligence community has different ways of looking at evidence that are also valid. So I think this really should be a partnership between subject matter experts.
source: 2023.04.06-Lipkin-Transcript.pdf, p.88
b. Holmes and Lipkin discuss the latest draft (10–11 Feb EST)
Less than 3 hours later after having a chat with Lipkin (10 Feb, 9:33 pm UK), Holmes shared the latest draft of the report with him (v6c), and explained that he was swayed towards a natural origin by the latest pangolin sequences which had 6/6 key mutations (the same argument as Farrar had previously highlighted in his email telling the group to go for publication).
Considering the unresolved question of the FCS, Lipkin quickly replied that the draft ‘does not eliminate the possibility of inadvertent release following adaptation through selection in culture at the institute in Wuhan’ and concluded his email by noting that ‘we have a nightmare of circumstantial evidence to assess’ (10 Feb, 10:01 pm UK).
The following day (11 Feb, 2:40 pm UK), Lipkin followed with some good remarks and further questions.
c. Lipkin is vetted by the other co-authors (10–12 Feb EST)
On 10 Feb, the co-authors discussed the possible addition of Lipkin as co-author, both on Slack and by emails, and quickly agreed to it.
A few elements of their discussions are worth noting:
- Rambaut confirmed that Lipkin’s addition should also help balance the Fouchier / Koopmans / Drosten opposition.
“We should get him on the group. Will make it more entertaining and balance the German/Dutch a bit”.
(Proximal_Origin_Emails.pdf, p. 66, 10 Feb 7:52 pm UK) - Garry clearly stated that the co-authors were already aware of the concerns of ‘higher-ups and intel’:
“But if Lipkin says higher ups are concerned and intel involved, it’s consistent with all we know too.”
(Proximal_Origin_Emails.pdf, p. 65, 10 Feb, 8:51 pm UK) - Garry showed that he was well ware that Daszak and possibly Fouchier would push hard against the passaging option in the draft, while discussing Daszak’ drafting of his Statement of Support:
“Not surprised Ego krewe (maybe Fouchier too) writing some sort of counter to the white paper with the allusion to scenario 3 “passage”. Preemptive strike?
(Proximal_Origin_Emails.pdf, p. 65, 10 Feb, 8:51 pm UK) - Holmes concurred, noting that, since Lipkin’s was also involved in GOF, having him as co-author of a text discussing the possibility of a passaged virus would partially offset the opposition of the GoF axis of Fouchier, Koopmans and Drosten:
“Think we should add him as an author. Safety in numbers. In his own mind he brings a lot of gravitas…plus because he is involved in the GOF I think it add weights. Happy to be over-ruled though.”
(Proximal_Origin_Emails.pdf, p. 76, 10 Feb, before 0:15 am UK)
For the sake of clarity, Lipkin was not admitted to the inner sanctum of the Slack group. He was involved in the email discussions from that point on. He was also active in editing the draft, after Holmes gave him editing rights on 12 Feb (~8:24 am UK, as per Slack conversation), and thus became the latest and last addition as a co-author of Proximal Origin.
d. Something more on Lipkin
Farrar may not have known it, and it likely did not play any role in the story in any case, but Lipkin was also a good friend of Baric, who testified that:
A: I have known Lipkin for a long time. [..] Any time I go to New York, I visit him and talk to him, sometimes stay at his home. We talk about science off and on all the time, potential collaborative research that we want to do, interesting results. He’s a friend and a colleague’.
source: Baric-TI-Transcript.pdf, p.18
Lipkin’s position on the origin question did indeed mirror Baric’s; both did try to highlight the possibility of a lab-escape of a passaged virus, within the limitations of their positions and intelligence involvements.
16.2 Nature is contacted to discuss publication (12–13 Feb)
On 11–12 Feb, Farrar was at the WHO BluePrint meeting in Geneva, which gathered 100s of scientists. Rambaut and Drosten were there too, as well as Daszak, Koopmans, and Embarek. The last three took part in an animal origin working group, one year precisely before they joined force again for the 2021 WHO mission to Wuhan [footnote 40].
On 12 Feb (11:09 pm UK, 3.:09 pm CA), Andersen emailed Clare Thomas, Senior Editor at Nature, to discuss a possible publication of P.O., which was in the final stages of drafting (without sending the draft itself).
Prompted by Jeremy Farrah, Tony Fauci, and Francis Collins, Eddie Holmes, Andrew Rambaut, Bob Garry, Ian Lipkin, and myself have been working through much of the (primarily) genetic data to provide agnostic and scientifically informed hypotheses around the origins of the virus.
source: Proximal_Origin_Emails.pdf, p.81, 12 Feb 2020, 11:09 pm UK
The next morning (13 Feb, 9:34 am UK), Clare answered that she was happy ‘to take a look to consider whether it might be suitable for Nature’. She also suggested that it might fit a Perspective format. On the same day (13 Feb, before 8:13 pm UK), Farrar connected with Magdalena Skipper (‘Magda’), Editor-in-Chief at Nature and thus above Clare, to discuss publication too. This was soon followed by a phone call between Farrar and Magda (13 Feb, around 11:40 pm UK), during which Magda seemed positive (‘She gets it’).
Based on the feedback shared by Farrar, Holmes then emailed Magda to start a direct communication channel with Nature. Holmes shared Magda’s answer on Slack on 14 Feb, 9:48 am UK, probably shortly after Magda replied to him that morning (UK time).
In other words, within 36 hours, Andersen had contacted Clare to discuss publication in Nature, Farrar had followed with a call to Magda, Clare’s boss, to push the case. Holmes then wrapped it up with an email to Magda (cc. Farrar) who confirmed her interest and put Clare to work on it.
16.3 Some new figures are considered for publication (12 Feb)
On 12 Feb, the co-authors discussed including some extra figures in the perspective they intended to publish in Nature. Three new figures were sketched that day:
❖ One was a schematic summary of the three scenarios that the report discussed. That figure never made it to the draft. It included at first a pangolin as putative intermediate host, before switching to a generic rodent, in line with the reservation in the draft about the possible role of pangolins. The figure, which was rather good, was the occasion of some joke about the possible role of Daszak in the 3rd scenario (‘C. Selection in cell culture’), with his profile used in the illustration of that scenario, the day after Holmes had called him ‘Dastwat’ (11 Feb, 9:37 am UK, see 15.1.c).
❖ Another figure showed the features of the spike protein in human SARS-CoV-2 and related coronaviruses. It appeared for the first time in the draft report on 15 Feb (v11 of Appendix A).
❖ The last figure showed some previous outbreaks in a schematic way (a. Severe Acute Respiratory Syndrome virus, b. Middle East Respiratory Syndrome, c. Ebola Virus, d. ?). That figure never made it to the draft.
16.4 Holmes temporarily takes the lead (13–16 Feb)
Once the decision to go for publication was taken, the draft needed to be reworked to make references to the pangolin sequences, and thus to keep it up to date with the pangomania, even if the latest pangolin sequences were objectively not that informative at all.
Andersen did not have much time to spend on the document later that week, being off on a break from Thursday 13 Feb to Sunday 16 Feb, with just some time to work on it on Wednesday 12 Feb. Holmes was by far the most experienced with these pangolin sequences, and the one in full tune with Farrar. So Holmes took the editing lead from Andersen.
16.5 The Passaging option is weakened for a possible publication (13–14 Feb)
Just after Nature expressed some interest, and as Holmes took the editing lead over from Andersen, the passaging option was weakened in the manuscript.
👉🏻 In the 12 Feb version (v8), passaging was still very much a valid hypothesis. The ‘Selection during passage’ section stated only a weak argument against it in the form of the O-linked glycans (an argument eventually proven wrong):
Basic research involving passage of bat SARS-like coronaviruses in cell culture and/or animal models have been ongoing in BSL-2 for many years across the world, including in Wuhan. In theory it is possible that SARS-CoV-2 acquired the RBD mutations and furin cleavage site during adaptation to passage in cell culture, as has been observed in studies with SARS-CoV. However, it is less clear how the O-linked glycans — if functional — would have been acquired in such a manner, as these typically suggest the involvement of an immune system, that is not present in vitro.
👉🏻 On 13 Feb (v9, 8:10pm EST), the text of ‘Selection during passage’ section became very defensive: that scenario was downgraded to ‘extremely unlikely’, weak arguments were added against the scenario, and the mention of Wuhan as a place where cell culture/animal model work took place at BSL-2 (which happened to be an essential argument for Andersen and Lipkin) was removed:
Basic research involving passage of bat SARS-like coronaviruses in cell culture and/or animal models have been ongoing in BSL-2 for many years in multiple laboratories across the world. Because we have taken an unbiased approach to attempt to discern the origin of SARS-CoV-2, we considered the possibility that SARS-CoV-2 originated in a laboratory and inadvertently infected a laboratory worker who transmitted the infection. We consider this scenario to be extremely unlikely.
However, after the emergence of SARS-CoV-1 several instances of laboratory acquisition of this virus by laboratory personnel working under BSL-2 containment have been documented and thus we cannot eliminate this possibility beyond doubt (ref). In theory, it is possible that SARS-CoV-2 acquired the observed RBD mutations site during adaptation to passage in cell culture, as has been observed in studies with SARS-CoV as well as MERS-CoV (ref). However, acquisition of the polybasic cleavage site or O-linked glycans — if functional — argues strongly against this scenario. [etc]
👉🏻 Then on 14 Feb, ‘extremely unlikely’ was replaced by ‘highly unlikely’, as it must have been judged that the weakening had gone too far. At the same time, most of the work in that version was on the polishing of the two other scenarios (acquisition of the FCS in some animal host or in humans).
There is nothing in the Slack or the emails that explains that overall weakening (unless that was discussed in the blanked-out sections). It is not clear if Andersen saw that change before leaving on his three days break (and thus before the latest draft was sent by Holmes to Nature on 17 Feb). The last Slack message of Andersen is at 3:27pm EST on 13 Feb, while the version with the downgrade to ‘extremely unlikely’ has a stamp-time of 8:10 pm EST.
Nevertheless, that weakening was perfectly in line with the multiple discussions by the co-authors of the risk of making public a text that would too clearly point at the possibility of a lab-product, both in terms of geopolitics and in terms of risk to personal careers (see 11.4 and 13.11). From a private report to a public text, some adjustments had to be made.
17. CONTEXT: Main-street media turn ugly just as the full WHO mission arrives in China (16–17 Feb)
17.1 The full WHO mission arrives in China (16 Feb)
The WHO advance team had left for China on 9 Feb evening (Geneva time), the day after Farrar had decided to go for publication and Dr. Tedros had promised to fight ‘trolls and conspiracy theorists’. On 15 Feb, the full WHO team departed for China, and had its first day of meetings in Beijing on 16 Feb. At the time, the full visit agenda was not yet agreed.
The apparent success or failure of the WHO mission would a defining moment for Tedros and the top ranks at the WHO; if the WHO mission did not deliver some results, the WHO risked losing support from the international community, already rattled by its earlier mishaps largely caused by the WHO inability to oppose the obstruction efforts of China (see sections 1 and 2).
But China had still not agreed much on sharing hard data or on the visit. Nor was there any agreement for a WHO visit to Wuhan; for China’s government, that visit had already happened in January, and this new WHO visit was about China letting the WHO learn from its successes, for which the WHO did not need to go to Wuhan, a point that nobody else but Dr. Tedros‘ right arm, Schwartländer, had repeated in an interview on 9 Feb (see Insight under 15.3).
Hence, from the moment questions started being raised about non-natural origins, the WHO needed to downplay them, (i) first to get its team in China, without yet a clear agenda and access negotiated, then (ii) to quickly finalise the agenda and access negotiations once the full WHO team was there, and last, (iii) to have a chance of sending a team to Wuhan (a possibility that the WHO was not raising any expectation for, as per Schwartländer’s interview with James Chau on 9 Feb, despite it being attested on the US HHS/CDC wish-list as early as 13 Feb) [footnote 96].
17.2 Washington Post vs. Tom Cotton
As we saw in 14.4, Sen. Tom Cotton had already been the favourite target of commentators that considered any enquiry into a possible origin as purely politically motivated. Not the person to being deterred by such hasty reactions, Sen. Tom Cotton gave an interview on Fox News on 16 Feb (morning EST), during which he mentioned the possibility of a lab escape. In that interview, he emphasized that there was no evidence either way for the origins, but that there were nevertheless valid questions to ask.
The interview was violently decried in the Washington Post, in an article by Paulina Firozi published online in the afternoon (16 Feb, ~5 pm EST), in the Politics section, and not in the Health one. The article was entitled ‘Tom Cotton keeps repeating a coronavirus conspiracy theory that was already debunked’.
The intensely political article mixed up fringe theories with perfectly valid ones — not helped by Cotton himself not making an obvious distinction between these — and ended up calling any lab-product scenario a ‘conspiracy theory that has been repeatedly debunked by experts’. In doing so, the article deformed the statement given by an expert, Richard Ebright [footnote 82].
“We don’t know where it originated, and we have to get to the bottom of that,” Cotton said. “We also know that just a few miles away from that food market is China’s only biosafety level 4 super laboratory that researches human infectious diseases.”
Yet Cotton acknowledged there is no evidence that the disease originated at the lab. Instead, he suggested it’s necessary to ask Chinese authorities about the possibility, fanning the embers of a conspiracy theory that has been repeatedly debunked by experts.
source: Washington Post, 16 Feb 2020, ~5 pm EST
17.3 Political radicalisation
Within one hour of the Washington Post article appearing on Twitter (16 Feb, ~6 pm EST), Sen. Cotton penned a detailed reply, but the US media would mostly adopt the Washington Post position in the following days, with the public taking predictably very political sides.
18. Publication of manuscript on virological.org and submission to Nature (15–17 Feb)
18.1 Getting the manuscript ready (15–16 Feb UK)
On Saturday 15 Feb and Sunday 16 Feb, the co-authors discussed finishing the edits, selecting a cover image, selecting the reviewers and who should be listed as author and in which order, etc. This was done without Andersen, who was off on vacation until Monday.
In that mundane process, a few elements are worth noting:
- Holmes, who was firmly in the lead editing position in the absence of Andersen, explained that he would include some new comments from Farrar, and some comments from someone else at Wellcome (possibly Mike Ferguson, see 13.3).
(15 Feb 7:58 pm UK) - Garry discussed acknowledging Farrar and the members of bounce-off group (the ‘email squad’: i.e. Fouchier, Koopmans, Drosten, Farzan, etc).
(15 Feb 11:13 pm UK) - Holmes clearly stated that their paper does not present any new data: ‘The only unpublished data we cite is a reference to Andrew’s dating analysis from Virological. We don’t actually present anything specific. Practically, this meant that the P.O. manuscript was very much a perspective — with some synthesis and opinion expressed — and not a piece of research.
(15 Feb 11:59 pm UK) - Garry and Holmes seemed quite confident that the peer review would be fast and easy.
(16 Feb 8:36 am UK) - Lipkin discussed asking Nature to actively push the paper, with at the minimum a short press release.
(16 Feb 11:46 am UK) - Holmes reminded everybody that the manuscript would also have to be sent to the WHO.
(16 Feb 10:04 pm UK)
Key Insight: A surprising omission
As discussed in 15.3, the ‘bounce-off group’ of Fouchier, Koopmans and Drosten was not part of any communication with the co-authors group since Farrar’s final confirmation of going for publication on 10 Feb (9:26 am UK). They were properly and definitely cut off.
It is clear that the co-authors resented the ‘krewe’ of Fouchier, Koopmans and Drosten, and especially Fouchier, whom they criticised at length in the Slack conversation. Still Garry suggested acknowledging the contribution of the ‘email squad’, which makes sense since Fouchier, Koopmans, and Drosten played an essential role in the discussions, when, rightly or wrongly, they convinced the co-authors that SARS-CoV-2 was not the result of some gene editing.
Garry’s suggestion was not followed up or even answered. No member of the ‘email squad’ was eventually acknowledged on virological.org, while the Wellcome Trust was acknowledged ‘for support’, in an indirect reference to Farrar and Ferguson. The version published in Nature Medicine a month later would only add Farzan to the acknowledgements.
18.2 Sense of urgency after the Washington Post piece on Tom Cotton (16 Feb, 10:52 pm UK)
On 16 Feb (10:52 pm UK), Garry shared the Washington Post article published a few hours earlier, about Tom Cotton allegedly pushing bioweapon conspiracy theories, on both Slack and by email. He added that this was a good reason to get their manuscript out.
18.3 Approval from Farrar and Collins (16 Feb, 11:06 pm UK)
On 16 Feb, in the afternoon EST (morning next day in Sydney), Collins received the latest version of the P.O. draft from Farrar (v11a). Collins answered a few hours later, at around 6 pm EST that day, and urged to go for publication ‘ASAP’. Holmes was one of the recipients of that email, and he immediately forwarded Collins’ answer to the other co-authors, including Lipkin (11:06 pm UK).
Lipkin then sent a quick email to Collins to say that he was ‘pleased that you liked it’. Another email from Holmes to the co-authors group, sent a bit later (17 Feb, 1:59 am UK), confirmed that both ‘Jeremy Farrar and Francis Collins are very happy’.
At this stage, the first objective the co-authors had was to get agreement from Nature that they could first publish the manuscript as a preprint, while Nature considered it for publication. This way, they could address the urgency that Farrar, Collins, and others had been highlighting.
18.4 Urgent submission to Nature (17 Feb, ~2:30 am UK)
On Monday 17 Feb, around ~2:30 am UK, within 4 hours of receiving the approval from Farrar and Collins, Holmes submitted the manuscript to Nature [footnote 52].
Initially, Andersen was supposed to send the manuscript himself on Monday 17 Feb (US time), once back from his 4-day break. It would indeed have seemed rather logical to wait for Andersen to have had at least a chance to quickly review the latest draft that Holmes had managed in his three days absence.
Instead, ‘pressure from on high’ meant that Holmes was asked (most likely by Farrar, with the WHO behind) to send the piece out a few hours earlier, in time for Nature to have received it by the start of the day in London on 17 Feb (which would be the middle of the night in California). It did not make much sense, though, as it was then about 2:30 am in London. [footnote 12]
Holmes to Andersen, 17 Feb, 2:48 am UK:
Anyway, it’s done. Sorry the last bit had to be done without you…pressure from on high.
18.5 Nature agrees for the manuscript to be released while it is being reviewed (17 Feb, 8:07 am UK)
After submitting the manuscript through Nature’s portal, Holmes emailed Clare Thomas (Senior Editor at Nature, based in London) to let her know of it. On Monday 17 Feb, at 8:07 am UK, answered Holmes’s email. A few minutes later (8:50 am UK), Edward Holmes (in Sydney) forwarded Clare Thomas’ answer to the co-authors group with some quick instructions.
In her email, Clare explained that she would send the manuscript (that she can see in her system) the same day for expedited peer review, and added that if the referees came back positive, the authors may still need to update their copy to take into account the latest publications. Most importantly, she also gave them the green light to release the manuscript while waiting for the review.
In his email to the co-authors group, Holmes instructed Rambaut that he could “put this on virological and do some precision tweeting”. Virological.org is a virological discussion board that Rambaut manages (so not strictly speaking a preprint site). At the same time, Holmes was still wondering if bioRxiv (a preprint site) would accept it.
18.6 Tidying up of manuscript on Virological.org (17 Feb, 10:41 am UK)
Soon after being given the green light by Holmes, Rambaut put together a final version with all the latest changes, ‘Andersen.Nature.Perspective.Final_v2.docx’ (10:51 am UK). [footnote 83]
Rambaut next used that text and started formatting it on his website, virological.org. Even as Rambaut was uploading and formatting the manuscript on virological.org, a few minor corrections and changes were done by Garry, Holmes and Rambaut himself [footnote 45].
The first publication to refer to it was the Daszak’s Statement of Support in Lancet (published there at the instigation of Farrar, see 19.5), which accessed it on 17 Feb, most likely after Farrar let Daszak know of it (Daszak was aware that the manuscript was in peer-review with Nature on 18 Feb, only one day after it had been sent there, see Insight under 8.3).
18.7 Farrar requests the final version for the WHO (17 Feb, 3:28 pm UK)
On 16 Feb, at 10:04 pm UK, Holmes had reminded his co-authors that the manuscript would also have to be sent to the WHO (see 18.1). Accordingly, as the uploading on virological.org was being finalised, on 17 Feb at 3:28 pm UK, Farrar asked to be quickly sent that latest version to make sure that the WHO may see it as soon as possible.
Farrar, 17 Feb, 3:28 pm UK:
When you have been able to update with the extra sentence and data can you forward on to me — ke[y] that WHO see [this] ASAP.
Farrar’s request is also mentioned in the Slack conversation three minutes later (3:31 pm UK), with Garry stating :
Garry, 17 Feb, 3:31 pm UK:
‘The version on virological is pretty good — Jeremy is asking for it — makes a much stronger case against bioengineering.’
Painful Insight: Unforeseen Consequences
There is a superb irony in the fact that the very partisan anti-Cotton hysteria of 16–17 Feb, following the deformation of his origins questioning into a bioweapon rant by the Washington Post, actually made the situation worse for the WHO, at the very time when the WHO was just starting a highly publicised mission in Wuhan. A mission that had been painfully negotiated over two weeks, but without yet any agreement for a visit of Wuhan and for access to detailed epidemiological data.
Farrar thus tried to limit the potential damages and maximise the chances of proper access, by organising the publication of Daszak’s Statement in Lancet and choreographing the release of a toned down P.O. on virological.org, both on 17 Feb, the second day of the WHO mission in China.
The parallel with Farrar’s decision to go for publication on 7 Feb is striking; as we have noted (see ‘Insight’ under section 15.3), that decision had been triggered by the open-ended NASEM call of 6 Feb followed by Fauci, declaring on 7 Feb that the US had the means to determine if the virus was a lab product, and would be looking into it — all precisely at the time when the WHO was trying hard to get its advanced team in China.
Farrar’s actions, on 7 Feb and again on 17 Feb, unfortunately resulted in a scientific and health-diplomacy fiasco of epic proportions, in the form of an aggressive pushback against any level-headed questioning of the origins, with — entirely predictably — no proper access to China at all to show for it in the end. The road to WHO hell is paved with good intentions.
18.8 Confirmation of the importance of the Cotton backlash (17 Feb, 5:41 pm UK)
On 17 Feb, 3:53 pm UK, Andersen noted ‘I gotta say — the guy isn’t totally wrong’, while discussing Tom Cotton’s interview on Fox News the previous day (see 17.2). By this, Andersen was referring not to the bioweapons accusations but to the real possibility of a lab escape, especially of a passaged virus.
Shortly after, at 4:09 pm, Rambaut added that people were starting to notice that the manuscript on virological.org did not rule out passaging, ‘which we don’t because it is still plausible’. In the same message, Rambaut included a tweet discussing the possibility of leak from a lab of a natural virus studied there, which is yet another possible lab-escape scenario, but this time simply not discussed at all in the manuscript. One may thus argue, like that tweet author, that, for clarity and exhaustiveness, that possibility should still have been mentioned, even if it was indeed indistinguishable from a zoonosis at the sequence level.
At 4:42 pm UK, Farrar confirmed in an email answer to Lipkin, that the recent reporting of Sen Cotton was in part responsible for his sense of urgency about publication.
Lipkin, 4:41 pm UK:
Jeremy,
Thanks for shepherding this paper. Rumors of bioweaponeering are now circulating in China.
IanFarrar, 4:42 am UK:
Yes I know and in US — why so keen to get out ASAP.
I will push Nature
Answering Rambaut on Slack, Andersen mentioned that the conclusions were too open-ended in the previous versions of the draft report, and that his main concern was to be effectively treated like Tom Cotton, by only picking up on the possibility of a lab origin of the FCS, and not the other plausible origins presented in PO.
Andersen, 17 Feb 2020, 5:41 pm UK:
There is no question that this’ll be picked up with “top scientists consider this could have come from the lab”. This was my main concern with this ‘backfiring’ based on our previous versions where the conclusions we too open ended — I feel [that] in the current version we do everything possible to property discuss everything, but yes, at this stage we unfortunately just can’t rule out a potential accidental infection from the lab.
In line with Andersen’s concerns, we can see how the manuscript was modified on 15 and 17 Feb to de-emphasise further the passaging option, which remained only in a stealth mode (v11a and v12b of Appendix A).
18.9 Farrar asks for some further watering down of the deliberate engineering scenario (17 Feb, 6:02 pm UK)
On 17 Feb, at 6:02 pm UK, Farrar asked the co-authors for a change: emergence through laboratory manipulation was to be re-qualified from ‘unlikely’ to ‘improbable’. Exactly nine minutes later (6:11 pm UK), Andersen (who was by then back into leading the editing process) sent a revised version of the manuscript with the modification asked by Farrar. The version on virological.org was also changed.
For Andersen, the change would seem minor. Rightly or wrongly, the co-authors had rejected a deliberate gene manipulation, so ‘unlikely’ or ‘improbable’ did not matter for them. In contrast, as we have already noted, the passaging option was still open in the version uploaded on virological.org, albeit in a stealth mode, after a bit of a back and forth in emphasis between the versions on 15 and 17 Feb.
18.10 Farrar asks to remove the manuscript from virological.org to better push it at a press conference first (17 Feb, before 6.10 pm UK)
We learn from a message by Rambaut, posted at 6:10 pm UK on 17 Feb, that Farrar had asked the co-authors to remove their manuscript from virological.org, so that he may instead get a press conference organised, for maximum impact. Most likely, he meant a press conference with the WHO.
This would have seemed to Farrar to be an excellent way to hit back at the very recent bioweapon allegations (see 18.5). It might have raised questions, though, as to why a non peer-reviewed manuscript was being platformed by the WHO, and only released after their press conference. At the very least, this strongly suggests that Farrar considered that he ‘owned’ the P.O. manuscript, and indeed he did, to some extent, as it had commissioned and shaped it for the WHO.
18.11 Andersen does not want to remove the manuscript (17 Feb, 6:19 pm UK)
In response to Farrar’s demand to take down the manuscript from virological.org for the benefit of a (WHO) press release, Andersen stated that the manuscript should stay there and that there should be no press release (17 Feb, 6:19 pm UK)
A few minutes later (6:40 pm UK), Andersen explained that he was concerned about the fact that putting the manuscript on preprint site such as bioRxiv, instead of on a board like virological.org, would make it more official, so that people would construe it as meaning that the co-authors were acting on behalf of their institutions, which effectively meant that they would have to ask for authorisation first.
18.12 Farrar asks for the preprint release date to organise some publicity (17 Feb, 6:22 pm UK)
While acknowledging reception of the revised manuscript from Andersen, with the requested change from ‘unlikely’ to ‘improbable’, Farrar asked when the manuscript shall be released as a preprint so that he may organise for some publicity for it (6:22 pm UK).
Asking the same question again a few minutes later (6:35 pm UK), Farrar explained that is objective was to coordinate ‘press briefings etc etc… to make sure the key messages are reasonably reported’.
We note that this was in full line with the aggressive stance that the WHO was taking to push its messaging actively, through both traditional and social media, as Dr. Tedros had explained in his WHO statement on 8 Feb (see 14.3).
18.13 Andersen tells Farrar that bioRxiv can’t take their manuscript (17 Feb, 6:30 pm UK)
Soon after (17 Feb, 6:30 pm UK), Andersen told Farrar that bioRxiv does not accept perspectives or opinion pieces, but only original research. So, since the manuscript was very much a commentary or perspective, releasing it as a preprint bioRxiv was normally not possible. Andersen nevertheless suggested that he might use his contacts there to try to get an exception to the rules, if there was a strong justification for it. (Andersen clearly had contact there: as he told Farrar, he was leading the effort at bioRxiv for the screening of outbreak-related preprints.)
We note that Andersen did not share all the objections he had for a preprint with Farrar, such as his concern that more work was still needed. He instead just used a technicality to get out of it.
18.14 Farrar pushes the manuscript on Twitter (18 Feb)
On 18 Feb, 7:49 am UK, Farrar retweeted a thread by Andersen that discussed the manuscript on virological.org. In doing so, Farrar tagged the main WHO officials on the pandemic: Dr. Tedros; Soumya Swaminathan, who was the WHO Chief Scientist; Maria Van Kerkhove, who was with the WHO mission in Wuhan at the time; Gabby Stern, who was responsible for the WHO global communication strategy.
18.15 The NIH requests an update on publication (19 Feb)
Three days after Farrar had shared the preprint to Fauci (NIAID) and Collins (NIH) and got their approval (see 18.3), on 19 Feb, Amanda Christine Coleman, of the NIAID Office of Scientific Coordination and Program Operation, contacted Reed Shabman, of the NIAID Office of Data Science and Emerging Technologies (cc: Liliana Brown, Office of Genomics and Advanced Technologies), to ask if the P.O. manuscript — described as a ‘Covid-19 preprint of interest’ — had been submitted to a peer-review journal and possibly already accepted.
Shabman asked the question to Andersen and sent it to Coleman, whereby Andersen told him of the submission to Nature and recommended that NIAID coordinate with Chris Emery, VP of Marketing and Communications at Scripps Research (Andersen’s institution), if they were interested in pushing the paper on publication.
Main takeaways:
- From the overall exchange, it appears that the NIH and NIAID were strongly interested in seeing the manuscript published, with the seal of authority of a peer review (as had been suggested by Fauci during the Farrar call).
- From Andersen’s email, we can see that the communication strategy being devised by Scripps Research, with the help of Wellcome, targeted both ‘the public and policy-makers’.
Key Insight: Saving Face
For all purposes, the February WHO 2020 mission to China could have been done via conference calls from Geneva and elsewhere. The rest was essentially just messaging that allowed both China and the WHO to save face.
Indeed, all the praising piled by the WHO on China, to get that second mission going, resulted in very limited access: once in China, the full team spent the vast majority of its collective time in Beijing (1,100 km north from Wuhan in a straight line), Shenzhen / Guangzhou (830 km to the South), and Chengdu (1,200 km to the West), very far away from the locked down city of Wuhan, or even from Hubei province. As a map shows, it would have been difficult to pick up major cities further away from Wuhan in these three directions (while including the necessary Beijing).
No access to Wuhan was agreed until around the 20 February, well into the mission, which was initially meant to finish on 22 February, with return flights late that day or the following day. In the very last programmed days of the mission, the WHO managed to extract a very limited access: a visit of two Wuhan hospitals and some discussion with Hubei province officials, over only one day (23 Feb). For that purpose, the WHO extended the mission to the 24 Feb.
The Wuhan contingent was made of only 3 international members (out of 13) selected by China and paired with 3 Chinese members. Weigong Zhou and Clifford Lane — the two US members of the full WHO team in China, who, at the time, were in contact with staff of the Beijing office of the US CDC (namely Robert J. Simonds and Alexander J. Millman) — were suggested by the Western side for the Wuhan contingent, but were not chosen. Bruce Aylward, Chikwe Ihekweazu (Nigeria) and a third unidentified person were selected instead [footnote 96].
For reference, this was even less than the 2 days spent in Wuhan by a 5-member WHO team during the little advertised first WHO mission, precisely one month earlier, on 20–21 Jan 2020. The purely symbolic access was accepted by the WHO because it was felt that the second WHO mission would have little credibility without a visit of Wuhan, how skeletal it may be.
In fact, in a perfect act of Chinese imperial diplomacy, on 23 Feb, during his visit to Wuhan with the two other WHO members, Bruce Aylward did present the findings and conclusions of the WHO report, largely drafted by end of day 22 Feb 2020 (the initially planned last day of the WHO team stay in China), to the Chinese Minister of Health, Dr Ma Xiaowei. In other words, the visit to Wuhan on 23 Feb was largely out of the scope of the report, and could have no incidence on it, as the report was delivered to the Chinese Minister of Health during the Wuhan visit, and thus could not be changed in its findings and conclusions.
It was purely face-saving, or an ‘absolute whitewash’ as per Lawrence Gostin’s words:
Yet, even before the full team gathered in Beijing on Feb. 16, the World Health Organization had ceded ground, according to two people who were on the mission, diplomats and others. It agreed not to examine China’s early response or begin investigating the animal source, they said. It could not even secure a visit to Wuhan.
“We weren’t there to look at the animal origins,” said Dale Fisher, a National University of Singapore medical professor who took part in the mission. [..] Once in China, team members agreed the mission would not be credible unless they went to Wuhan, said H. Clifford Lane, a clinical director at the National Institute of Allergy and Infectious Diseases.
Wuhan was sealed, so six members — three Chinese and three international experts — took a special train to the city. They stayed for about a day and visited two hospitals. They did not go to the market. [..]
“It was an absolute whitewash,” said Lawrence O. Gostin, a professor of global health law at Georgetown University. “But the answer was, that was the best they could negotiate with Xi Jinping.”source: NY Times, ‘In Hunt for virus source, WHO let China take charge’, 2 Nov 2020
The quote above confirms again, if need was, that the Feb 2020 WHO mission was not meant to look at the origins. As we have noted before, this was in line with the official Chinese media definition of the purpose of the mission after Xi Jinping had agreed to it at the end of Jan (see 2.6.d), and as repeated by Dr. Tedros’ right arm on 9 Feb (see 14.7): it was about the international community having a chance to learn from a benevolent China how it was fighting the epidemic.
The same Lawrence Gostin was quoted in an article published in Science on 10 Feb (so likely written on the crucial days of 8 or 9 Feb as the WHO was trying very hard to get an advance team into China), where he discussed the impossible balancing act that the WHO was finding itself in:
Yet the crisis has put Tedros “in a near-impossible situation,” says Lawrence Gostin, director of the O’Neill Institute for National and Global Health Law at Georgetown University.
If Tedros wants WHO to stay informed about what’s happening in China and influence how the country handles the epidemic, he cannot afford to antagonize the notoriously touchy Chinese government — even though it is clear the country has been less than fully transparent about the outbreak’s early stages, and perhaps still is. [..]
source: Science, ‘The health career’, 10 Feb 2020
The same article quoted Farrar asking for more epidemiological data and sequences, with some rather wishy-washy wording about what he expected to actually receive from China:
Farrar says China has done reasonably well so far in a difficult situation, but the country should release more basic epidemiological data and viral sequences quickly. “To me, those are the two big gaps because those two allow you to track the epidemic,” he says.
Tedros urges patience because China is overstretched. “So we get some information, we may not get other information. It may not be complete. But we understand that.”
source: Science, ‘The health career’, 10 Feb 2020
19. In parallel: Daszak’s pre-emptive strike (6–18 Feb)
19.1 The need to close the gaps in the NASEM letter (5 Feb)
On 5 Feb, Daszak started drafting his ‘Statement of Support’ that would become the ‘Lancet letter’ and circulated it by email.
On 7 Feb, when discussing his latest draft, Daszak explained that he would add a reference to the NASEM letter (that was finalised the previous day (see 8.3.g) once that letter is published. He also explained that the latest version of that letter was making the point that ‘The initial views of the experts is that the available genomic data are consistent with natural evolution and that there is currently no evidence that the virus was engineered to spread more quickly among humans’.
Daszak immediately added: ‘I think this is a bit too specific, because there are other conspiracy theories out there’.
In other words, Daszak explained that the draft NASEM letter was going some way to deny a non-natural origin, by addressing the engineering hypothesis, but that it also left some gaps; ‘other conspiracy theories’ as he called them, which meant (i) passaging (which is normally distinct from engineering) and (ii) a research-accident involving a natural (non-altered) virus.
He then stated that his proposed ‘Statement of Support ‘neatly refutes most of them’ by saying that ‘We stand together to strongly condemn conspiracy theories suggesting that 2019-nCoV does not have a natural origin. Scientific evidence overwhelmingly suggests that this virus originated in wildlife, as have so many other emerging diseases’. [footnote 75]
Daszak thus made it absolutely clear that the purpose of his letter was to close the non-natural origin gaps that were left in the NASEM draft letter.
Unfortunately, unbeknown to him, the ‘available genomic data are consistent with natural evolution and that there is currently no evidence that the virus was engineered’ that he saw (and likely worked for) in the draft NASEM letter, has already been struck out in the final version (see 8.3.g).
We can only imagine how Daszak, who was already fretting about some remaining gaps in the draft NASEM letter, must have felt when he found that the door was actually left wide open in the final version of that official US scientific statement at the time, to which intel had contributed (through the FBI & ODNI, as per 8.3.c).
19.2 Daszak decides to hide his role after talking to Baric and Linfa Wang (6 Feb)
According to Baric’s House testimony, when Daszak was starting to write his Statement of Support, or possibly the day before (4 Feb), he called Baric and asked him if he would be willing to be a signatory. Baric stated that he declined and told Daszak that both of them would have a conflict of interest if they signed it, because of their work with the WIV. Then, after talking to Linfa Wang the next day (5 Feb), Daszak decided that he too should not sign the Statement of Support.
That rather clean version of the events unfortunately does not fully match what the FOI’d email conversation between Daszak and Baric shows: there, Baric comes up as still supporting the Statement and was not really trying to distance himself from it (except by not signing it). Nor did Baric seem to object to the fact that Daszak would still push the Statement without signing it, when Daszak wrote to him about it. Moreover, only a week later, Baric had no problem at all contributing to the text of a very controversial EMI article that also denied a non-natural origin, while being careful not to sign it. [footnote 80]
If Baric was actually worried about conflicts of interest, he should have told Daszak to drop the whole Statement idea altogether, or to quickly let others take care of it and have nothing more to do with it. The best that can thus be said for Baric is that, at least, his own practical role in the Statement was near null (he did not edit it or got people to sign it) so that the conflict was theoretical for Baric, while it was very practical for Daszak.
In the end, Daszak did sign his own Statement, while still not revealing his central role in its genesis and coordination, so not changing anything to his conflict of interest. It seems that Daszak worried definitely more about getting the appearances as right as possible, than about the fundamental problem. [footnote 36]
The conflicts of interest of Baric, Daszak and Linfa Wang — whether theoretical or much too real — were only made worse by the fact that the three of them were involved in the key DEFUSE proposal back in 2018, a proposal that was meant to enhance new coronaviruses with some FCS insertion, with the virus discovery and culturing work done at the wholly unsuitable BSL-2 level, in Wuhan. None of Daszak and Linfa Wang disclosed the DEFUSE proposal until it was exposed by a whistleblower (through DRASTIC) [footnote 62].
19.3 Daszak amplifies Dr. Tedros words about ‘trolls and conspiracy theorists’ (9 Feb)
Following the WHO update of 8 Feb 2020, in which Dr. Tedros made some borderline dystopian statement about censure (see 14.3), a remarkably responsive Peter Daszak decided to include part of it in his latest draft:
I’ve cited the statement from the WHO DG which was released today saying “At WHO, we’re not just battling the virus; we’re also battling the trolls and conspiracy theorists that push misinformation and undermine the outbreak response.”
source: Daszak, Biohazard_FOIA_Maryland_Emails_11.6.20.pdf, 9 Feb 2020
That rather over-the-top quote would eventually be replaced in the final version of the Statement of Support by a milder: ‘We support the call from the Director-General of WHO to promote scientific evidence and unity over misinformation and conjecture’.
19.4 The co-authors see a ‘preventive strike’ in the Statement (10 Feb)
On 10 Feb, Holmes mentioned that according to Lipkin (who just contacted him), Daszak was writing a pity letter in support of the WIV (the ‘Statement of support’).
He’s about where we were a week ago. He’s for escape.
He also said that Peter Daszak, grand wizard of EgoHealth, and some others were writing a piece saying the Wuhan lab were being persecuted.source: Holmes, Proximal_Origin_Emails.pdf, 10 Feb 2020.
Garry replied that Daszak’s purpose there may be to try to counter the possibility of passaging (the ‘scenario 3’ in the draft of P.O.), which had been picked up by intel and could be very damaging to his work with the WIV, given that higher-ups and intel are involved.
Not surprised Ego krewe (maybe Fouchier too) writing some sort of counter to the white paper with the allusion to scenario 3 “passage.” Preemptive strike?’
source: Garry, Proximal_Origin_Emails.pdf, 10 Feb 2020
Taking more swipes at Daszak and his ego, Holmes called Daszak the ‘Grand wizard of EgoHealth’, Garry referred to the ‘Ego krewe’ (EcoHealth Alliance [‘Ego’] and the Dutch [‘krewe’] opposition of Fouchier and Koopmans), and Andersen joked that when ‘Butt Lesion’ (Lipkin) and Daszak get into the same room, one gets a ‘thermonuclear ego explosion’.
19.5 Farrar gets Daszak’s letter in Lancet (16 Feb–18 Feb):
a. From a petition on change.org …
Initially, Daszak’s idea was to get his Statement circulated via social media and emails, to drive potential signatories towards an online page (change.org) where they could add their signature.
The plan was also for Daszak to present it at the 19th International Congress on Infectious Diseases in Malaysia (ICID), programmed for 20 February, where Hume Field and Peter Daszak (both of EcoHealth Alliance), plus George Gao (head of the Chinese CDC), Linfa Wang, and Sylvie Briand (who was part of the WHO trip to Beijing with Tedros and Ryan on 28–29 Jan) were supposed to be part of a panel discussion on the outbreak.
Daszak, 6 Feb:
Once we’ve got a good group of around 20 well-known people, I will then circulate this via social media and email, with a link to a webserver for others in the greater science community (and interested public) to sign on to this statement. I will then present this to the ISID meeting in KL, Malaysia in 2 weeks and I think we’ll get a big impact from that community and it should then take off.
Please note that this statement will not have EcoHealth Alliance logo on it and will not be identifiable as coming from any one organization or person, the idea is to have this as a community supporting our colleagues.source: OSU-1.12.23–1265-pages.pdf, p.274
b. ... To a letter in Lancet
Daszak planned his Statement as an online petition, that he would then push in some conference in Malaysia. That event got postponed on 10 February (and eventually cancelled) because of the COVID-19 outbreak. Nevertheless, thanks to Farrar, the Statement of Support got published in a journal, and of all of them in the most prestigious one (Lancet).
The involvement of Farrar was disclosed in an email that Daszak sent on 16 Feb to ask the Statement signatories if they had any issue with a publication in Lancet, on top of the existing publication on change.org. There, Daszak mentioned that Farrar had offered to arrange for the publication of his Statement of Support in Lancet ‘to get our statement across directly to the senior leaders in the governments of China and around the world’.
Daszak, 16 Feb, 3:48 pm UK:
For these reason, and as a way to get our statement across directly to the senior leaders in the governments of China and around the world, Jeremy Farrar (Director of the Wellcome Trust, and co-signatory) suggested that I submit this letter to the editor of The Lancet for possible publication. I have done so just now, out of a sense of urgency, and await his response (I have also asked if he is willing to sign).source: OSU-1.12.23–1265-pages.pdf p.365
So, on 16 Feb, not only was Farrar was working hard for the release of the manuscript of Proximal Origin on virological.org the following day, and for facilitating its prompt consideration by Nature (sent there on 17 Feb), but he was also working at the very same time on quickly getting Daszak’s Statement of Support in Lancet (published on 18 Feb). [footnote 93]
c. Additional implications
A few additional elements are worth noting regarding Daszak’s email:
- It is clear that Farrar’s purpose in getting the Statement in Lancet was to appease China, to facilitate access for the WHO.
- Farrar was 100% aware that Daszak was the promoter of the Statement of Support, even if he never mentioned it until this was disclosed via FOIs.
- Farrar had accepted to sign the letter, even if he was not yet included in the latest version circulated by Daszak with his email, which already included Golding and Turner from the Wellcome Trust.
Essential Insight: Farrar’s offensive
As we saw in section 15.3 (‘Essential Insight’), the dice were cast on 8–10 Feb, when Farrar decided (‘ordered’ may be a better word) that the report of the co-authors should be sent for publication, and not just reviewed privately within an ad hoc WHO group.
Already on 9 Feb, Schwartländer (Tedros’ right arm) had described the coming WHO mission as the way for WHO to learn about the outbreak and the way China was fighting it, with Wuhan not being necessarily the best place to do so because of the health emergency there. There was zero mention of a non-natural origin, and Schwartländer was open to the possibility of no visit to Wuhan at all. This was the price to pay for the mission to be able to fly to China.
As the NY Times wrote:
Yet, even before the full team gathered in Beijing on Feb. 16, the World Health Organization had ceded ground, according to two people who were on the mission, diplomats and others. It agreed not to examine China’s early response or begin investigating the animal source, they said. It could not even secure a visit to Wuhan.
source: NY Times, ‘In Hunt for virus source, WHO let China take charge’, 2 Nov 2020.
The 16 Feb 2020 is the second key date in the shaping of the zoonosis narrative. While the WHO mission had just arrived in China, there was still no guarantee yet of its smooth proceeding, especial as regarded access to data and isolates. So, on that day, Farrar made sure that he had two assets to try to facilitate the work of the WHO team that was now in Beijing.
These two assets were:
- A ‘scientific go to paper’, to be released on virological.org the next day (17 Feb), with hopefully a WHO press-release, ahead of a publication in Nature. Farrar did not only order its publication, but precisely shaped it as a product (tone: ‘definitive’, objective: ‘scientific go to paper’).
- A statement of solidarity with Chinese scientists and medical professionals, to be published in Lancet without delay (it took only two days). There, Farrar upgraded Daszak’s Statement from a dull presence on change.org to a prompt publication in the most prestigious medical journal.
Both assets conveniently denied a non-natural origin while demonstrating the determination of the WHO and of Western scientific publications to endorse that line. The publication of both was organised by Farrar, who contacted the right people at each publisher.
There is no escaping the conclusion that, at the time, Farrar and the WHO were quite happy to sacrifice any proper investigation of the origin, for the sake of getting access to some epidemiological data and isolates, and be able to demonstrate that the WHO was still relevant. Moreover, with the non-natural origin evacuated, there was less pressure to go to Wuhan, which the Chinese had clearly no appetite for.
With this done, the WHO mission could seem manageable.
19.6 More examples of Daszak’s upset at intel involvements (16–18 Feb):
On 15 Feb, Senator Jon Cotton dared state on Prime TV that the origin of the virus was unknown and that it could be either natural or the result of a research accident. This was followed the next day (16 Feb) by a strong attack on Cotton in the Washington Post (see 17.2).
In the email of 16 Feb to his signatories, discussing Farrar’s offer to get his Statement into Lancet, Daszak mentioned the interview of Tom Cotton on prime TV the previous day. In the same email, Daszak also asserted that the ‘conspiracy theories’ had been given credence by ‘discussions at a very high level with government in the USA’.
Daszak, 16 Feb, 3:48 pm UK:
Secondly, I want to mention that the situation as regards conspiracy theories has worsened over the last few days, having been given credence through reporting yesterday in a UK tabloid (Express), regurgitation on prime time lV yesterday in the USA by US senator Tom Cotton, and discussions at a very high level within government in the USA and China. At the same time, some colleagues in China have received violent threats to their families and themselves.
What Daszak likely meant there, was that the non-closing of a non-natural origin possibility in the NASEM answer to OSTP had allowed the question of the origin to start spreading within the US government.
This was in perfect agreement with what Lipkin had told Holmes on 11 Feb, when discussing the concerns of ‘higher-ups’ and intel (see 16.1.a). It is also consistent with the email that Daszak sent to Linda Saif on 18 Feb (see Insight under 8.4):
“Unfortunately this is the exact same group that elevated the potential that this was a lab release all the way to the White House two weeks ago.”
19.7 Publication of the Statement on change.org and in Lancet (18 Feb)
a. Publication
On 18 Feb, 10:51 am UK, Daszak sent another email to his signatories, announcing that, at the express request of Farrar, Richard Horton of Lancet had agreed for publication on that day, with a Mandarin version too. This was followed by another email at 2:10 pm UK, confirming the timing, and adding that Lancet itself would be contacting journalists (who may in turn contact the signatories), to push it further.
Accordingly, on 18 Feb, ~3 pm UK, the Statement of Support was published simultaneously on change.org and in the prestigious Lancet. As part of the effort to push the Statement through the media, Lancet itself contacted journalists, in coordination with Daszak.
b. A few points about the signatories
- Charles Calisher, the first signatory in alphabetical order, and not Daszak, is corresponding author.
- Three of the signatories are from the Wellcome Trust: Farrar, Jose Golding and Mike Turner.
- One signatory, Bart Haagmans, is from Erasmus, but neither Koopmans nor Fouchier (also from Erasmus) signed it.
- Haagmans (who signed) and Koopmans (who did not sign) were recommended by Linda Saif on 7 Feb, who at the time was co-author of an extremely controversial EMI paper that broke basic editorial and peer-review rules. [footnote 80]
- Baric and Linfa Wang, who were some of the very first people contacted by Daszak, are not signatories, in line with their objections.
- Daszak, Baric and Linfa Wang were all key parts of the DEFUSE proposal to enhance coronaviruses in Wuhan (with virus discovery and culturing done at BSL-2), which they never disclosed.
- Drosten got actually quite upset about that non-disclosure, saying that he would at least have asked questions before signing if he had known. [footnote 84]
19.8 Farrar pushes Daszak’s Statement on Twitter (18 Feb)
Farrar tweeted about Daszak’s Statement in Lancet on 18 Feb, 6:01 pm UK. This happened only 10 hours after Farrar had tweeted about the release of the P.O. manuscript on virological.org, when he also tagged Dr. Tedros, Soumya Swaminathan, and Maria Van Kerkhove (see 18.14).
Two hours later (18 Feb 8:17 pm UK), Daszak himself tweeted about it, while passing himself as just a signatory of the Statement, and not its initiator.
Funny enough, Farrar replied to his tweet (8:30 pm UK) with a mention of the leadership of Daszak in the Statement, something a bit unfortunate since Daszak was precisely trying to keep his role in it undisclosed.
Nevertheless, true to form, Daszak answered in an artful way, which to anybody would sound as if Farrar had just generally discussed their shared ‘need to speak out’, and not about his factual role in the conception of the Statement, and which to Farrar would sound like a ‘thank you’ for his (also undisclosed) key role in getting the Statement in Lancet.
Josie Golding from the Wellcome Trust, Marion Koopmans, as well as the then Chief Scientist of the WHO, Soumya Swaminathan (since then replaced by Farrar), soon also tweeted about it:
Insight: Roll of Honour
The people most responsible for the downplaying of the lab-accident hypothesis, via the two key pieces released in 17 Feb (P.O.) and 18 Feb (Daszak’s Statement), are:
- Farrar, motivated by what he considered a supreme objective (securing WHO access).
- Holmes, for various reasons, including an unconditional support to his ‘handler’ (Farrar) and his parallel involvements with Chinese teams on pangolin papers which were pushing a zoonosis narrative.
- Indirectly Fauci, for bringing attention to the US capacity and desire to look into lab-product scenarios (7 Feb), at the worst possible time for the WHO, following the NASEM call and letter (see 8.4).
To which we should add the contributions of:
- The NASEM call (3 Feb) and the letter to OSTP (6 Feb), which, against the best efforts of Daszak, and with the likely support of Andersen and the agreement of the intel elements involved (FBI Quantico and the ODNI representative), did not kill a non-zoonotic origin. (See Insight under 8.4)
- The relentless demonising of Tom Cotton, with strong US political overtones, that overshot its target and only made the task more difficult for the WHO (see Insight under 18.7).
▶ ️Farrar:
Given the essential role that Farrar played in the publication of both P.O and of Daszak’s Statement, and of their platforming, Farrar should rightly be considered as the scientist who most actively and wilfully contributed to the shaping of the zoonosis narrative.
Farrar did so for the purpose of gaining access to China for the WHO:
- His decision to go for publication, and its first circulation of the draft report, coincided with the departure of the WHO advance team to China, where they were to negotiate the visit of a full WHO team.
- His push for the quick publication of P.O. (virological.org on 17 Feb), and his decision to publish Daszak’s Statement of support in Lancet (18 Feb), coincided with the Tom Cotton’s controversy on 17 Feb and the start of the work of the WHO full team in China (16–24 Feb), when access to Wuhan had not even been agreed and the team was kept at best 800 km from there (see Insight under 18.15).
I leave it for others to argue whether the decision made sense or not. What is sure is that the decision was the logical result of the weak hand that the WHO had, partly due to its structural inability to take a stand, that had gone only worse since SARS-1.
What is also certain, is that the decision will haunt the WHO and Farrar for ages. Farrar’s writing of Spike may have been a cathartic attempt, but it would have been more effective if Farrar had no intentionally left a huge blank in his book where the story of drafting of P.O. should have been, and if he had acknowledged his role in publishing and platforming Daszak’s Statement of Support.
▶ ️Holmes:
Holmes, a British citizen and a known world expert whom Farrar had immediately gone to when spurred by British intel services, thus played a key role, by facilitating Farrar’s key move.
But after gifting Farrar with a good reason to go for publication, Holmes should have, in fact, excused himself from the drafting process as he was working on natural-origin papers with Chinese teams (starting with Lam et al., sent for peer review to Nature on 7 Feb). That would have been the right thing to do.
But, first, Holmes is also known for putting his name on a surprisingly large number of papers, and, secondly, he seemed to have had no problem persuading himself that his collaboration with pangolin paper teams, on top of his ongoing relation with the Chinese CDC (which had already delivered, see 1.2), only made him an exceptional and central scientist, fully justified in whatever he chose to write and publish.
▶ ️Fauci:
As for Fauci, there is no question that a natural origin determination would have relieved him, and that he went on to masterly play the cards he was handed.
Still, I believe that he was not yet fully decided on how to position himself until that second week of February, when he must have started fully understanding the implications for NIAID and himself. In fact, Fauci definitely further complicated the situation for the WHO, when he highlighted the capacity and desire of the US to look into the origin question, in his ABC interview of 6 Feb.
After Farrar changed course and tried to kill a lab-product origin, Fauci simply did not need to do much or take much risk; Farrar had done the job. Fauci was left with tidying up his own house, one step remote, via intermediates like Morens.
▶ ️What about Andersen, Garry, Rambaut and Lipkin?
These four were not the marionettists of the narrative, they were the marionettes. They reacted to events, tried to dodge the bullet of being called conspiracy theorists, and in the end, under pressure, largely agreed to go with what Farrar needed, duly relayed by Holmes in their drafting group.
Andersen’s efforts to keep the passaging option alive in the version on virological.org may seem rather trivial given how well hidden that hypothesis was there, but at least he did it. And still, that stealth mention of passaging was enough to trigger the Ego ‘krewe’ into some nasty backstabbing before its final publication in Nature Medicine.
20. The pangolin claims implode (18 Feb–20 Feb)
20.1 Wrong alert: the Lam et al. sequences (17 Feb)
The co-authors were hoping to soon see the SCAU pangolin sequences on GISAID so that they may learn whether these had any FCS, and better understand the <up to> 99% claims.
The SCAU preprint was deposited on bioRxiv on 17 Feb, based on its DOI (10.1101/2020.02.17.951335). It was released by bioRxiv on 20 Feb, a typical 3 days after being deposited. It is quite likely that Andersen, Holmes or Rambaut would have known of the imminence of the SCAU preprint being sent there.
They got a false alarm on 17 Feb, at around 5:51 pm UK, when Peter Bogner, manager of GISAID, shared with Rambaut some new pangolin sequences, which had just been deposited on GISAID (a normal step when depositing a manuscript on a preprint site). A quick examination showed that these sequences had no FCS. Holmes explained later that day that these sequences were, in fact, from the pangolin preprint he was working on with Tommy Lam (Lam et al., see 20.3), which was soon due for release on bioRxiv. Holmes confirmed that these new GISAID sequences had no FCS at all.
20.2 SCAU Pango99 sequences disappoint (18–19 Feb)
On 18 Feb, 11:52 pm UK, Rambaut shared an analysis of some new sequence that had just been uploaded to GISAID (EPI_ISL_410721.fasta). This time, he was able to recognise these as being the awaited ‘pango99’ from SCAU based on the GISAID attributes. The sequence had no FCS, and <up to> 99% really meant an average of 90.3% similarity.
Andersen’s reaction was rather clear:
Andersen, 18 Feb, 11:56 pm UK:
The more pango sequences I see the less likely I find that they are intermediate — I think they’re just one of many animals with SARS-like CoVs.
Holmes summed up succinctly the situation when getting back in the Slack conversation for the first time in more than 24 hours:
Holmes, 19 Feb, 4:11 am UK:
The new pango virus is almost identical to ours [meaning Lam et al.]. They [SCAU] totally over-hyped in the press release. Mind you. Universities always over-hype these things.
20.3 Two pangolin preprints and a funeral (18–20 Feb)
In the end, the ‘pango99’ hopes were dashed by the publication of a pair of preprints on bioRxiv:
- On 18 Feb: ‘Lam et al. (‘Identification of 2019-nCoV related coronaviruses in Malayan pangolins in southern China’). That’s the preprint Holmes had been working on with Tommy Lam.
- On 20 Feb: the SCAU preprint, Kangpeng Xiao et al. (‘Isolation and Characterization of 2019-nCoV-like Coronavirus from Malayan Pangolins’), with a virus very similar to one already published in Lam et al.
The pangolin viruses described in these preprints had no FCS at all and a fairly distant homology to SARS-CoV-2. They were not the very close relatives that the false 99% claims had suggested, and they thus left the origin question still open, despite the hasty claims made during the SCAU press conference on 7 Feb.
20.4 Still undecided while waiting for Nature’s answer (19 Feb)
When Andersen talked to Clare Thomas (Senior Editor at Nature) on Tuesday 18 Feb, she had mentioned a likely peer review feedback by the end of the week, which would be by Friday 21 Feb at the latest. Meanwhile, the co-authors tried to understand the likely consequences of the punctured SCAU pangolin claims, in terms both of the likely origin but also in terms of the possible perception of their manuscript, by either the readers on virological.org, or by the reviewers at Nature, especially as concerned the possibility of passaging.
a. An intriguing RBD — Passaging is ‘quite a high probability event’
With the absence of an FCS and the low overall homology with SARS-CoV-2 of the just released Malayan pangolin sequences, except in the RBD section, Andersen started focusing on the RBD instead.
One possibility that was considered, for that near perfect conservation of the RBD in SARS-CoV-2, was a recombination across animal hosts.
But at the same time, Andersen questioned if a pangolin virus could have been grown or passaged in a laboratory to see if it could infect human cells, especially since these Malayan pangolin viruses were collected around January 2019. In doing so, Andersen qualified the passaging scenario as ‘quite a high probability’ event, just two days after mentioning that possibility only in stealth mode in the P.O. text on virological.org, and after Farrar, Daszak and others used that text to deny any lab origin.
Andersen, 19 Feb, 00:06 am:
Question is — did they recently realize that pangolins carry CoVs and then grew them in the lab to see if they could infect human cells? This is quite a high probability event.
Clearly none of these pangolin sequences were the source though.
The RBD is very intriguing — if it’s not lab, then definitely recombination (also high probability event)
b. Checking if the manuscript should be altered (19 Feb)
Next, Andersen and Garry discussed whether they should modify the manuscript to mention something about recombination, in light of the RBD. They did not discuss reinforcing the passaging option.
The discussion took an interesting turn, with Gary arguing that reinforcing the recombination scenario, at the detriment of the passaging one, may not be necessary, given that close to 99% of people did not make a fuss about the stealth possibility of passaging in the version on virological.org.
Unless, that is, their manuscript was sent for review to someone who would want to see any possible argument used, or abused, to dispel any whiff of a non-natural origin, as eventually happened.
Garry, 19 Feb, 0:28 am UK:
[..] the twittering has been closer to 99% [positive] than the pangolin sequence. A few diehards might object to even whiffing at the possibility of a lab escape, but I didn’t get the sense from the public reactions that that was offensive to most. Clearly stating no bioengineering seems to be the take home, plus that it is well done and needed.Andersen, 19 Feb, 0:37 am UK:
I think there are two camps in the interpretation of the paper:
(1): definitely didn’t come from the lab,
(2) they [the authors] said they can’t rule out it came from the lab so it definitely came from the lab.
c. An odd deletion that could point to passaging
As part of these discussions, Rambaut mentioned that one of his collaborators had found a deletion just before the FCS in about 50% of the reads from a sample, and was wondering whether this could be induced by passaging:
Rambaut, 20 Feb, 2:38 pm UK:
Basically a collaborator has found this deletion in about 50% of the reads from a sample. I guess it is possible that it is a cell adaptation (removing the glycan sites as well).
d. Dazed and confused
Garry and Andersen replied by expressing their incapacity to land at the time on one specific origin, constantly moving from natural to research-related accident:
Garry, 20 Feb, 3:18 pm UK:
Indeed — that PRRA insertion is the most perplexing aspect of the entire genome. It’s likely “out-of-frame” actually, but seeming inserted like a scalpel into a very constant region. If that region is or can be put under some selection pressure would be good to know.Rambaut, 20 Feb, 3:20 pm UK:
This whole thing is doing my brain in, I literally swivel day by day thinking it is a lab escape or natural.Andersen, 20 Feb, 3:25 pm UK:
Haha, my brain has been a badly calibrated MCMC. I’m hoping I’ll start converging at some point… [footnote 35]Garry, 20 Feb, 3:26 pm UK:
All our brains are in a bit of trouble — hopefully you’ll don’t get rear-ended anytime soon… Hopefully also we hear something positive from Clare SOON — then we’ll all likely be facing the lab escape or natural question head-on and should have a consistent response.
Andersen also pointed that Farzan had mentioned that such deletions could be a sign of passaging in a laboratory:
Andersen, 20 Feb, 3:09 pm.
Interesting with that deletion. I should say that Mike Farzan mentioned that any deletions around this site would be a red flag for him the that furin site had initially come about with (T/C) [thymine/cytosine] passage — and then with slower passage in humans, might be modified’.
Insight: Just like a Greek tragedy
At this stage, one question imposes itself:
Given that the emerging false pangolin narrative had convinced Farrar and Holmes to go for publication, why didn’t the authors of the manuscript revisit their draft after the SCAU imploded on 20 Feb, to re-emphasise passaging, especially considering the questions that they were now asking about passaging based on the RBD?
The answer is fairly simple: That ship would not turn at that stage.[footnote 31]
- The priority for a large part of the US and European scientific community was to fight the outbreak, not to start an internal fight at home and a diplomatic one with China.
- Changing course would have run the risk of rendering extremely complex the necessary pandemic work with China, and any further mission there. Health diplomacy prevailed, not science for the sake of science.
- Too much virtue signalling had occurred in the US media to walk that back without damaging one’s reputation. Especially given the politicisation that accompanied it.
- Holmes himself, so instrumental in providing the excuse that Farrar needed to go for publication, was too taken in the pangolin theories to walk that line back. Beyond ‘Lam et al.’, he was invested in a few other pangolin papers, absolutely convinced that there lied the solution to the origins of SARS-CoV-2.
Another scientist, Susan Weiss, provided a fully similar experience. At the very same time as she was expressing privately her serious concerns about the SARS-CoV-2 FCS, after the disappointment of the SCAU publication on 20 Feb (Kangpeng Xiao et al.), she was still happy to put her name on a paper in EMI which strictly denied the rationality of such concerns. [footnote 80]
In the end, scientific integrity has its limits — especially when imperatives are driven by health diplomacy issues. Furthermore, opposition to the main scientific narrative may come at a huge personal cost; most scientists working for public institutions know that very well and will not voice their concerns in public, especially not in these uncomfortable situations where it would matter most. Instead, they typically go with the flow, and if the flow is defined by health and science diplomacy concerns, with an added domestic political colouring, then all the better — they’ll find even fewer reasons to object.
21. Rejection by Nature (20 Feb)
21.1 Reasons given for the rejection (20 Feb, 4:52 pm UK)
On Thursday 20 Feb, 4:52 pm UK, Clare Thomas got back to Andersen. Unfortunately, the news were not good: Nature had turned down publication. She highlighted two specific arguments for the refusal:
- Reviewer #2 considered that the manuscript (which still quietly left open the passaging option) could actually feed the ‘conspiracy theories’ instead of squashing them.
- Both Nature and Reviewer #2 felt that the perspective could be quickly become obsolete, as more data was expected to come in.
➤ Argument #1
The first argument could be expected: only the day previous day, Andersen had summarised the polarised reactions that the piece may steer:
Andersen (19 Feb, 0:37 am UK):
I think there are two camps in the interpretation of the paper:
(1): definitely didn’t come from the lab,
(2) they [the authors] said they can’t rule out it came from the lab so it definitely came from the lab.
🤷♂️
➤ Argument #2:
The second argument is backed by Nature, not just reviewer #2. However, the reasoning of Nature and reviewer 2 are very different.
For Nature, a fundamental issue is that the Perspective format is not really suitable, since, with the constant flow of new information, nobody can see very far. Said otherwise, a perspective (literally ‘see-through’) should reach further to be useful. In fact, Clare Thomas had already mentioned on 17 Feb that the draft she had received was likely already outdated due to some just released publication).
We will discuss the reasoning of reviewer #2 in the next section. But suffice to say here that Nature’s answer was not totally surprising, even if one could have rightly expected much more understanding, given the intervention of Farrar who got right to Magdalena Skipper (the Chief Editor).
Clare Thomas concluded her email by stating that ‘Nature Medicine are interested in publishing it either as a Comment or a Correspondence, which compared to a Perspective format could indeed make more sense. To summarise, it seemed that Nature did not want the exposure for some reason, while not wanting to look uncooperative either (hence the referral to Nature Medicine).
21.2 Quick answer back from Andersen (20 Feb, 5:48 pm UK)
Within one hour, at 5:36 pm UK, Andersen shared the news that Clare Thomas at Nature had turned down their manuscript. His Slack message included her email.
Before anybody had the time to answer on the Slack group, Andersen sent a quick acknowledgement email back to Clare Thomas, that thanked her for her effort and gently but firmly argued back against the second argument of reviewer #2 (5:48 pm UK).
In all truth, that second argument was rather weak as formulated by reviewer #2. Reviewer #2 pre-supposed that the origin must be zoonotic and relied greatly on the pango99 claims that had been discredited since the review was written. Hence, reviewer #2 could not have indeed been more wrong when writing ‘Once the authors publish their new pangolin sequences — a lab origin will be extremely unlikely’.
As Andersen correctly asserted in his answer to Clare Thomas:
The possibility [of a lab origin] must be considered a serious scientific theory (which is what we do) and not dismissed out of hand as another ‘conspiracy theory’. We all really, really wish that we could do that (that’s how this got started). but unfortunately it’s not just not possible given the data.
source: Proximal_Origin_Emails.pdf, p. 118, 20 Feb 2020, 5:48 pm UK
21.3 The co-authors discuss Reviewer #2 (20 Feb, ~6 pm UK)
After checking the refusal email and the full reviewers comments, the co-authors briefly discussed the objections of reviewer #2 and started formulating a response.
In the words of Garry, reviewer #2 ‘set up a straw man that our paper was [meant] to refute SARS-CoV-2 as a bioweapon then shot it down’.
Garry was correct to point out that the manuscript was meant to consider all non-natural theories, including passaging, and not just the extreme bioweapon theories (based on targeted engineering and not passaging), even if, as we have seen, that passaging option was, in fact, included in a rather stealthy mode, to avoid potential blowbacks.
We may notice the logic of reviewer #2 was very close to the reaction of Drosten when he checked the draft for the first time (see 13.7). It was also very close to the position of Koopmans and Fouchier, who first made sure that the targeted engineering option was rejected in the Farrar call, and then balked at the passaging option that Andersen, Garry, Holmes and Rambaut came back with, by saying that there was no need to discuss that extra concern as it was not in the public domain (see 15.2).
Holmes, not incorrectly, added that ‘Reviewer #2 is clearly of the Fouchier mindset’. The convergence of arguments and logic was indeed rather surprising:
Garry and Holmes went further and suspected that Reviewer #2 sent some very critical private comment to the editor, scaring Clare away:
In full contrast, reviewer #1 seemed rather balanced and perfectly comfortable exploring lab-escape scenarios. That reviewer also had a fairly good knowledge of SARS lab-escapes, as it corrected two mistakes about these in the manuscript, and s/he even asked why the authors did not at least briefly mention the possibility of a lab-escape of a natural virus (see Appendix C).
Sideline: A very odd review
One paragraph of reviewer #2 comments is rather odd:
“It’s not clear why the authors do not refute a hypothetical lab origin in their coming publication on the ancestors of SARS-CoV-2 in bats and pangolins. The tree showing diverse pangolin viruses has kindly been made available by some of the authors in GISAID. Once the authors publish their new pangolin sequences, a lab origin will be extremely unlikely. It is not clear why the authors rush with a speculative perspective if their central hypothesis can be supported by their own data. Please explain.”
source: Proximal_Origin_Emails.pdf, p. 130
One cannot help but feel that there was a bit of a confusion between publications, as reviewer #2 seems to mentally merge the P.O. manuscript he was reviewing with Holmes’ paper with Tommy Lam, sent to Nature for review on 7 Feb and released as a preprint on 18 Feb.
But more importantly:
▶️ First, reviewer #2 accusations are severe. S/he is saying that the authors know that the passaging option is invalid, based on their own data which they are not presenting here, and still decided to keep that option alive in an unnecessarily speculative manuscript. It is actually quite difficult to think of a more damaging comment, as, if true, the offence could be qualified as fraud (pushing results clearly inconsistent with the data available to the authors). As Garry put it, “Actually this is rather freaking insulting to say the least.”
Accordingly, faced with a reviewer happy to play the fraud card, Andersen addressed that point straight ahead in his answer to Clare Thomas, when he explained to her that: ‘Had that been the case, we would of course have included that — but the more sequences we see from pangolins [..] the more unlikely it seems that they’re intermediate hosts’..
▶️ ️Second, it is not clear how reviewer #2 reached that conclusion, and whether s/he is being honest with it. Lam et al. was released on bioRxiv on 18 Feb, and the data shared on GISAID the day before (see 20.1). The included phylogenetic tree that reviewer #2 likely checked was very far from showing a close enough similarity between SARS-CoV-2 and various pangolin sequences, to be able to claim that the case was closed. So, was reviewer #2 supposing that Holmes had access to more conclusive sequences? Or was s/he using that tree as a pretext to try to sink the P.O. manuscript?
21.4 Deciding on a strategy (20 Feb, late evening UK)
The co-authors immediately started writing some rebuttal of the objections of the reviewers and of Clare Thomas. The idea was to see if Nature would reconsider, and if it would not, to give the rebuttals to Nature Medicine so that they may publish the piece with some minimal modifications.
To try to convince Nature to reconsider, Andersen suggested that Farrar should get back to Magdalena Skipper (Nature’s chief editor) and explain to her why reviewer #2 was badly wrong.
Andersen, 20 Feb, 6:32 pm UK :
The only potential door still open with Nature would be for Eddie and Jeremy to get hold of Magda. Reviewer 2 in general doesn’t understand what going on [..] and no, sadly, the pangos don’t solve it. I get a sense that Nature might be a little gun shy though — hence we need to go all the way to the top.Garry, 20 Feb, 6:36 pm UK:
Good idea — let Jeremy know and give him the rationale why Reviewer 2 was full of it.
Holmes backed that suggestion, especially since he was afraid of having to tune done the conclusions further and/or having to cut heavily through the text to get it published in Nature Medicine. Within a few minutes (6:44 pm UK), Andersen had a basic version of some rebuttal arguments (‘Nature rebuttal’) that was good enough to be used by Farrar when talking to Magdalena. As the one with the privileged contact to Farrar, Holmes forwarded him that initial version of the rebuttal.
Holmes, 20 Feb, 6:58 pm UK:
I forwarded to Jeremy. Reviewer #2 is clearly of the Fouchier mindset. I am very surprised at Nature here.. rejecting it then recommending another Nature journal [..]. My worry about transferring to Nature Medicine [is] that they will want the text hugely reduced for a Comment/Correspondence section. Also I think we should stick to our guns about the message and not tone it down just to get it published. I’m pretty sure Cell would take it.
By the end of the day (21 Feb, 0:33 am UK), a plan had been agreed for the next steps, whereby the co-authors would polish the rebuttal so that it could be sent to Nature, Holmes would email it to Magdalena Skipper and request a call with her to discuss the situation. In parallel, they would do some final edits in the P.O. manuscript to get it ready for either resubmission or to be sent to Nature Medicine.
22. RmYN02 and its consequences (21–25 Feb)
22.1 Holmes hears of some bat virus that may point to an FCS (21 Feb)
On Friday 21 Feb, Andersen and Holmes worked further on the rebuttal. When he was finished, Holmes explained that he was worried about argument #2 of the rejection (the risk that their manuscript may get quickly out of date, see 21.1), since he was already ‘hear[ing] rumblings’ that seemed to point to the discovery of a related bat virus with a Furin Cleavage Site.
He advised to wait for a few days (so until Monday 24 Feb at least), which was agreed. As part of the discussion that unfolded, both Andersen and Garry nevertheless pointed out that an FCS in a somewhat distant bat virus would not refute any of their three origin scenarios, but would indeed be important information that would need to be reflected in the revised manuscript before sending it with the rebuttal letter to Nature.
As we shall soon (see 21.7), the ‘ramblings’ that Holmes had heard were likely about the yet-to-be-disclosed RmYN02, a still distant bat virus with some inserts in the right place, but not a working FCS.
22.2 Clear headwinds ahead (22–24 Feb)
a. Possible games
On 22 Feb, Andersen created yet another new Google Doc (!) to finish the rebuttal letter (‘Nature rebuttal letter’).
As they were working on it, Garry made an interesting comment:
- One of the reasons reviewer #2 gave against publishing the manuscript, was that there was no real need to address the supposedly very remote issue of a lab product, since (i) the few accusations of bioweapons were neither credible nor that common at all, (ii) opening the lab product scenario, including passaging, in a journal could only make things much worse by conferring some credibility to these.
- But Garry pointed that, in fact, the manuscript had been already uploaded on virological.org and had not made things worse, quite the contrary. Garry went further and sensed some play in the background, with some likely forcefully worded private comments to the editor.
Garry, 22 Feb, 7:51 pm UK
Yeah — damn good — I agree about the ‘fan the flames by adding speculation’. It would not surprise me that the reviewer wrote a VERY strong private comment to the editor that effect to scare the hell out [of her]. Again reviewer #2 wrong about everything. 50K+ views and probably 10s of thousand of tweets and retweets — I did not detect fanned flames — on the contrary.
To the point made by Garry, it is also worth noting that the concerns of reviewer #2 position were absolutely not reflected by reviewer #1, who, on the contrary, would have welcomed some mention of the possibility of a research-accident involving a natural (unaltered) virus, which at that stage was totally off the radar screen for most people.
b. Racaniello’s partial treatment of the P.O. manuscript
On 20 Feb, Racaniello, a virologist, posted a quick summary of the Proximal Origin manuscript on his virology blog (virology.ws). The summary itself is close enough to the original and mentions the passaging option in its toned down mode.
The next review of Proximal Origin by Racaniello was much more partial. On 23 Feb (10:15 pm UK), Rambaut shared the latest episode of the TWiV (The Week in Virology) podcast by Racaniello, recorded that day. Racaniello effectively ended up turning the deliberate ambiguity of the manuscript, with its toned down mention of passaging, into a overblown case against a lab escape [footnote 55].
Maybe that podcast, in its convivial atmosphere, was not necessarily conducive to the best science, resulting in that rather one-sided episode. Or possibly, Racaniello, a fervent supporter of GoF who had objected to the GoF moratorium [footnote 46], had personal reasons to ignore the passaging options:
And I tell you, it really bothered me that there were some suggestion that this was a lab construct, because, if it turned out to be true, that would bother the hell out of me. Not just because people dying and so forth, but it’s kind of an indictment of the field, right?
source: Racaniello, on TWiV 588 (43:02)
c. Bad signal from Nature
Soon after, on 23 Feb 11:35 pm, Garry shared a more serious bad signal, this time from Nature itself, in which they were trying to get their manuscript reconsidered: Garry pointed to a note, recently added to an old Nature article about the WIV, that stated that ‘scientists believe the most likely source of the coronavirus to be an animal market’, and was rather surprised by the position taken by Nature on the role of the Wuhan market:
Robert Garry, 23 Feb 2020, 11:35 pm UK:
Wow — not sure Nature is correct on this.
_____
Editors’ note, January 2020: many stories have promoted an unverified theory that the Wuhan lab discussed in the article played a role in the coronavirus outbreak that began in December 2019. Nature knows of no evidence that this is true; scientists believe the most likely source of the coronavirus to be an animal market.
_____Robert Garry, 23 Feb 2020, 11:58 pm UK:
Nature seems to be getting some bad advice — did reviewer #2 strike again?Robert Garry, 24 Feb 2020, 3:51 pm UK:
Someone should tell Nature that the fish market probably did not start the outbreak.
The co-authors were now facing a strong official narrative, that absolved any research activity and was instead pointing to the Wuhan Seafood market. With that narrative being platformed through supposedly reliable sources such as TWiV, and in Nature itself where they were trying to get published, it is difficult to see how the co-authors could keep taking a stand for passaging.
22.3 Holmes shares RmYN02 and starts working on yet another publication (24 Feb)
Precisely at that time when a zoonotic narrative was being emphasised through ‘expert’ avenues, an important bat sequence was shared with the co-authors. Based on that sequence, the P.O. co-authors veered temporarily towards a zoonotic consensus, with the help of some half-hearted deficient logic.
The sequence, RmYN02, was shared by Holmes on Monday 24 Feb (7:41 pm UK), when it was still not officially released. RmYN02 had been discovered in bat samples collected between May and October 2019 in Yunnan, by a local team, which included Alice C. Hughes. It would be the subject of a paper published in May 2020 (Zhou et al.), with Holmes (credited as having worked on data interpretation) and Alice Hughes as co-authors, alongside a team from Shandong and Di Liu of the WIV [footnote 54].
Holmes, as we have just seen, had explained 3 days earlier (Friday 21 Feb) that he could hear some ‘ramblings’ about a new bat sequence with an FCS. The image below, confidentially shared by Holmes on Slack on 24 Feb (7:41 pm UK), is essentially the same as in the bioRxiv preprint sent for publication a few days later, on 2 Mar (released on 5 Mar). It’s also very similar to the one in the final paper. We can thus infer that Holmes must have joined the Zhou et al. team over the weekend (22–23 Feb).
22.4 The FCS argument (25 Feb)
a. From the correct analysis ...
After some back and forth analysing that new sequence, the co-authors converged towards the correct analysis that, even if it was rather distant and without a working FCS, RmYN02 showed at least that some insertions were possible at the S1/S2 site in sarbecoviruses, right where SARS-CoV-2 (also a sarbecovirus) had that remarkable razor-sharp PRRAR insert.
This indicated that a path towards the natural evolution of the FCS of SARS-CoV-2 was theoretically possible, even if RmYN02, or any other known relative of SARS-CoV-2, did not actually constitute a proper stepping stone on that path, being way too far in evolutionary distance. What RmYN02 provided is a demonstration that a natural scenario for the SARS-CoV2 FCS was possible, without any of the specifics.
b. ... To the wrong conclusion
Unfortunately, Holmes, and to a lesser degree Andersen, seized on RmYN02 to accomplish an unjustified logical jump, whereby inclusion of a research-related accident as a valid hypothesis was now supposed to required necessity (sine qua non) instead of plausibility; that is to say, they would only retain the already artificially toned down passaging hypothesis if it was a strict requirement for an explanation of the FCS. And since RmYN02 showed that it was not, they were suddenly quite happy to purge the already half-hidden FCS origin hypothesis from their paper.
We can see that Andersen was a rather divided, with a difficult to follow logic (see highlighted sentences below), that is eventually betrayed by another deliberate use of quotation marks (‘natural’, on the model of ‘environmental’):
Andersen, 25 Feb 5:03 am UK:
I don’t think this data necessarily argues against accidental infection/release, it shows something very important — insertions at this site can happen in nature, making the need to reach for a non-natural explanation much diminished. This is new important knowledge that would need to be introduced in our commentary and lends significantly stronger support to the ‘natural’ scenarios we’re describing. I say we have to wait for this to come out — at a minimum on the bioRxiv. It doesn’t go against (or prove/disprove) the scenarios we’re describing, however, it is very important knowledge for a reader to know.@Eddie Holmes — what’s your take on how we handle this? I think we should wait until this is out, update the commentary, and then put that back in via Nature/Nature Med with some significantly stronger conclusions about this being ‘natural’. Thoughts?
Holmes, who had a conflict of interest in his co-authorship of Kangpeng Xiao et al. (which was precisely in review at Nature at the time) and Zhou et al. (soon to be sent for publication), was all for ditching the passaging option, particularly since he also believed that more important data would be released that week.
Holmes, 25 Feb 5:53 am UK:
I’m now very strongly in favour of a natural origin. The component bits of the virus are more or less there in a tiny sample of wildlife. Plus there is more to come (this is not Zhang’s data), I don’t see why we need a lab origin on these data. I agree we have to hold back for bioRxiv. Hopefully something will be submitted this week. I’m actually at a meeting with Clare next week.
22.5 Efforts to ignore a research-accident with a natural virus (25 Feb)
a. Some poor arguments
As part of the discussions on that busy 25 Feb, some faulty arguments were tried by the co-authors to wrap up the work and additionally kill the scenario of a lab-escape of a natural virus handled in a Wuhan lab:
Arg #1: One can’t say anything on a research-accident with a naturally occurring virus, ‘so we shouldn’t even mention it’ (Rambaut).
👉🏻 Absurd logic:
It’s not because one can’t say anything determinant about an hypothesis, that one should just ignore it! This is totally non-scientific. The normal scientific handling is instead to mention it clearly as a limitation in the paper.
Maybe because of the obvious limitations of Arg #1, a second argument was suggested soon after, which attempted to move from ‘let’s not mention it’ to ‘let’s mention it to better kill it’.
Arg #2: ‘the potential lab exposures pale in number to natural exposure’, so one can actually mention it to better ‘shoot it down’.
👉🏻Doubly wrong:
a. After the SARS escapes in two BSL-3 labs and one cabinet BSL-4 lab, with one outbreak (the Beijing-Anhui SARS outbreak of 2004), the WHO correctly stated that the risk of research-accident outbreaks was becoming essential [footnote 60]. Daszak himself reiterated that point in his GVP pitch [footnote 61]. This confirms that the top experts considered the risk of outbreak from a lab escape to be at par with the risk of outbreak from a zoonosis.
b. Not only were the risk factors of zoonosis vs. research accident already flagged as comparable, but one also needs to include the location of the outbreak and its timing in any proper evaluation of the relative probabilities of both hypotheses. Which practically means weighting the research risk factors against the zoonotic risk factors, all applicable in Wuhan at the time [footnote 62].
Put simply: the right question to ask is whether an outbreak, apparently starting in Wuhan (and nowhere else [footnote 63]) at a time when scientists there were working on coronaviruses at BSL-2, is more likely to be a research accident or a standard zoonosis. This is a basic law of probabilities that has proven its value in solving complex investigations. Ignoring it just non-scientific.
Rambaut, 25 Feb, 12:46 pm UK:
The only thing left on the ‘conspiracy’ side of things is that a researcher became infected through handling, sampling bats or culturing bat viruses [..]. But we don’t (and cannot) address the actual nature of the zoonotic event from the evolutionary/genomic event, so we shouldn’t even mention it.Garry, 25 Feb, 12:49 pm UK:
Agree — and as the last response to Rev#1 the potential lab exposure pale in number to natural exposure. [..]Andersen, 25 Feb, 2:35 pm UK:
Yes, I agree with this — mention it (because it must), but then shoot it down, that’ll be the most powerful way of countering this.
b. The rebuttal letter is adjusted:
Following this desultory exchange about a possible research-accident with a natural virus, the next version of the rebuttal letter (29 Feb) flipped from Arg #2 to Arg #1. The probabilistically ignorant ‘it is obvious that the latter is much less likely than the former’ was replaced with the non-sequitur ‘We would prefer not to speculate’.
22.6 A new plan is agreed (25 Feb)
a. Lowering of the lab-product hypothesis while waiting for RmYN02 publication
A new plan was quickly agreed by the co-authors, and validated by Farrar:
- Holmes to accelerate the release of the RmYN02 paper (that he was now working on) as a preprint, promising to ‘get the bat paper sorted ASAP’.
- Holmes to talk to Clare Thomas to explain what they are doing.
- The co-authors to do some revisions to the P.O. manuscript, to reference RmYN02 and lower the lab-product hypothesis, based on the assumption that RmYN02 made it less likely (see ‘Insight’ below).
- Then one should wait for the RmYN02 preprint to be up on bioRxiv (with the help of Holmes).
- Once this is done, the co-authors may then reference it directly in their revised version of P.O. and share it with Nature or Nature Medicine.
As Garry noted, the lowering of the lab-product hypothesis should ‘make Nature etc even happier’. Unfortunately, a large part of the corresponding Slack discussions related to the edits to the P.O. manuscript were not included in the FOI, so we likely are missing some important additional elements [footnote 58].
When validating the plan, Farrar asked for the SARS-CoV-2 name, which had been decided by the responsible ICTV committee on 5 Feb and ‘taken note of’ by WHO on 11 Feb, to be replaced by the unofficial HCoV-19. HCoV-19 was the name that China pushed back with, culminating in a publication in Lancet by Zhengli, Gao and other on 19 Feb. Holmes Slack message made it clear that the decision to use HCoV-19 came from the WHO.
The draft was changed accordingly by Holmes, who had previously been rather happy to use SARS-CoV-2, although he knew that ‘China will hate it’. Holmes had even changed the name back to SARS-CoV-2 when Lipkin had changed it to NCoV-19 on 12 Feb. One may wonder about the way the WHO was happy to accommodate the Chinese side against the official ICTV. (See Appendix B)
Andersen, 25 Feb 6:31 am UK:
Sounds good — I too think we should wait until this is out and then we can do a quick turn-around — I think we’ll still have a paper to publish by then and in fact, I think it’ll be even stronger as it’ll have much less of an open ending (again, it doesn’t rule out lab infection/release, however, there is now no longer any ‘mysteries’ to explain — we see the optimized RBD in pangolins and part of the furin site in bats [which is pretty cool!). Generally speaking, I also don’t think we want to rush, If you can please grab Clare when you see her, then that’d be great.Holmes, 25 Feb 9:04 am UK
Agreed. I promise to get this pushed out ASAP. I need to talk to Jeremy in a little while. Clare wants to talk about stuff so this will clearly be on the agenda.Holmes, 25 Feb 9:20 am UK
Jeremy agrees with this plan. I’ll get the bat paper sorted ASAP. They want to call the human virus HCoV-19🤷♂️.Rambaut, 25 Feb 10:00 am UK:
OK. To return to the paper — so are we going to:
1) Renuance it to explicitly lower our bet on the lab passaging scenario on the basis that both cleavage site insertions and the full RBD exist in nature. This leaves just having the source virus in the lab and someone being infected with it which is just an alternative human exposure hypothesis without any evidence.
2) Lower our odds on the pre-circulation in humans because of reasons above, and lack of evidence of cases.Garry, 25 Feb 11:03 am UK:
Agree with 1). This will make Nature etc even happier I think — so yes re-nuance. [..]
b. Some remaining paradox (25 Feb)
In the process, Garry and Rambaut could not help but notice the paradox that if they believed that the RBD and FCS evolved in a bat or some animal host, it would have been quite well adapted to humans when it jumped to them, meaning that ‘it should have circulated as soon as it arose’. By, paradoxically, ‘we don’t see any genetic variants that are likely older than Autumn 2019'.
Rambaut, 25 Feb 11:46 am UK
We still have the paradox — if the virus is human adapted, it should have circulated as soon as it arose. But we don’t see any genetic variants that are likely older than Autumn 2019.Garry, 25 Feb 12:03 pm UK
[..] Bottom line for me — the scenarios in the current draft don’t change, except lab escape unnecessary (we said this but can be further nuanced) — the new data refines the analysis considerably sharper, particularly re recombination, which is a major upgrade.
Yes — paradox still in full force.
Key Insight: A critical loss of perspective:
There are major issues with the co-authors’ treatment of RmYN02’s significance.
First, the question was never whether the SARS-CoV-2 FCS had to be a lab construct, but rather how likely it was for it to be a lab construct, all things being considered. Which meant looking at biological plausibility, technical capability, intentions and funding for such work in Wuhan, then crossing that with the specific circumstances of the outbreak, while weighing it all against the possibility of a zoonosis.
Secondly, given that laboratory research activities in Wuhan were precisely about predicting these evolutionary paths (through natural inserts and recombination), one could argue that, by definition, any lab product would seem to be somewhere on a possible evolutionary path, and would thus look perfectly natural and expected, if produced by no-see-them technics and/or passaging — which were perfectly accessible to Chinese scientists.
Hence, unless someone had left a deliberate signature in the genetic sequence, or used uncommon codons, any lab construct from such research activities would look exactly like it did NOT ‘need to be a lab construct’.
It thus looks as if the co-authors seized on an excuse to dump the passaging hypothesis, given the headwind that a non-natural hypothesis was now getting in Nature itself, and the likely realisation that RmNY02 would provide endless fodder for attacks from fellow scientists.
As Amy C. Sims, senior research scientist in the Chemical and Biological Signatures Division of the National Security Directorate at the DoE Pacific Northwest National Laboratory (PNNL), wrote on 15 May 2020 about the final version of P.O. [footnote 57]:
Amy C. Sims (PPNL), 15 May 2020:
To me (I only know one of the authors) this was written to provide reasonable scientific arguments to quiet the voices that were screaming about the pandemic being caused by a lab released virus. I do appreciate this doesn’t make the arguments true but they are plausible.source: FOI by Tobias, p. 5.
Key Insight: Lying by omission, or worse
That loss of perspective is all the more regrettable that at the time the co-authors were unaware of the DEFUSE proposal of 2018, which purpose was to precisely insert FCSs into unpublished viruses with a roughly 80% homology with SARS, and to polish these with some passaging, at the wholly inadequate BSL-2 level, precisely in Wuhan and precisely in 2019.
Given their assiduity to look into previous research associated to Baric and the WIV, with 6 Baric’s papers included in their presentation for the Farrar call on 1 Feb, the P.O. co-authors would likely have been very interested to know about that DARPA proposal, and this would have very likely totally reshaped their analysis. Indeed, even if it was eventually rejected by DARPA, the proposal offered a window on the directions of research in Wuhan, and also meant that some of the proposal preliminary work would likely have been partially done by the time it was drafted. On top of this, the WIV simply did not need DARPA money to pursue that research of interest, no more than Baric’s lab at UNC Chapel Hill needs Chinese money to pursue its own research objectives [footnote 59].
To add insult to injury, neither Baric nor Daszak came forward to disclose the DEFUSE proposal during the key NASEM call on 3 Feb, to which Andersen also participated, despite both having taken central roles in drafting it. It would have been straightforward for them to mention the proposal during the call, or to talk to Andersen on the side. They never did it. And as we have already seen (Lancet letter, section 19), and shall see again (section 24), that NASEM call seems to have only convinced Daszak to redouble his efforts to torpedo the P.O. co-authors.
Key insight: Was the approach doomed?
We can also see that one issue with the whole enquiry started by Andersen was likely doomed, due to the narrow angle it took and its asymmetry of information; Andersen had raised the alarm based on the assumption that the sequence included tell-tell signs of genetic manipulation. Once this was discarded (rightly or wrongly) by Fouchier during the Farrar call, Andersen came back with the passaging hypothesis as an explanation for the FCS.
In the absence of any knowledge about DEFUSE (contrarily to Daszak and Baric), with the RmYN02 showing that sarbecoviruses might eventually evolve a Furin Cleavage Site, Andersen simply folded, even if he later tried to dress this as a normal progression. Holding to the passaging hypothesis was just not worth it any more, given the strong headwinds:
- it would ago against the scientific narrative that was imposing itself (starting in Nature itself),
- it could likely mean rejection from Nature Medicine,
- it would go against all the efforts of Farrar,
- it would go against Holmes' current involvement in pro-zoonotic pangolin and bat papers,
- it would likely be met as a conspiracy theory, since it could not harness any hard evidence (at this stage), only strong probabilistic considerations (location, timing, and known work at BSL-2).
Further, the asymmetry of information (DEFUSE) allowed Daszak to run circles around P.O. manuscript, and to even later attempt to finish the authors off, in order to recover his suspended R01 grant and to optimise his chances of large GVP grants.
23. End-Game: Publication in Nature Medicine (26 Feb-17 Mar)
23.1 The RBD, or a rather fuzzy proximal origin (26 Feb)
After addressing the FCS, the co-authors spent some time discussing the RBD. Some very complex recombinations seemed to be at work in the SARS-CoV-2 genome, with similarities in some parts to RaTG13 (a bat sequence), and in other parts to the pangolin sequences. To illustrate that point, Rambaut had worked out some intricate recombination figure, which he interpreted as supporting a bat origin for the RBD and wanted to include in the manuscript.
Holmes was strongly opposed to pushing a bat origin for the RBD. According to him, the RBD may have instead evolved in an intermediate animal, but more importantly (and he was 100% right on this) the picture was very fuzzy with simply not enough data. Also, postulating a bat origin for the RBD would contradict the other papers Holmes was working on in parallel with other teams — which he could not accept. Holmes ended up threatening to remove his name from the manuscript if one went ‘into detail saying which bit of sequence came from where’.
The co-authors agreed to remain unspecific, Rambaut’s figure was not included, and Holmes took charge of writing about recombination in the manuscript.
Insight: Theory vs. Data
The whole RBD episode is interesting as it clearly shows how the lack of actual data turned into a key issue. Holmes was right in arguing that there was no way the co-authors could afford to be too specific. In the end, the co-authors had to resort to either very moot arguments (such as zoonotic precedents) to set the tone, or to the recombination Swiss-knife for some necessary heavy lifting, or to some evasive hint of an eventual possible evolution in a sarbecovirus (provided by RmYN02) for the FCS, etc.
All of these arguments may individually seem rational, but the only way they could amount to a demonstration of a likely zoonotic origin, and not just an ad hoc aspirational patchwork, was data.
In the absence of data, of which there was effectively very little, these arguments only contrived some plausible scenarios. Not a bad result per se, as long as other plausible (non-zoonotic) scenarios were not ignored based on a symmetrical absence of data — which unfortunately was not the case.
23.2 Decision to go with Nature Medicine (27 Feb)
On 27 Feb, Rambaut, Garry and Holmes, while wondering what to do next, thought that the ‘the window of opportunity for publishing this in the form it is, is vanishing quickly’.
Andersen was not so worried and thought that even if their paper was ‘going to be of decreasing interest as focus moves to pandemic control’, it was still going to be interesting enough. Nevertheless, the co-authors decided to try to accelerate the publication. This meant immediately addressing two questions:
a. Choice of journal for a fast publication
Because Nature itself would take too much time (they would have to persuade Clare Thomas to reconsider), they concluded that it was better to start looking elsewhere. Andersen was confident he could ‘sell’ the manuscript to Sri Devi Narasimhan, the Deputy editor at Cell, without having to try instead to convince the peer-reviewers, so that Cell would then take it with a light peer-review very quickly.
Nevertheless, Nature Medicine (which Thomas had suggested) had a higher impact factor. So the decision was to check with Nature Medicine if they could just reuse the peer-review from Nature — which would also ensure a prompt publication.
b. How to handle RmYN02 and Zhou et al.
The co-authors had initially agreed to wait for Zhou et al. to be released as a preprint before re-submitting their manuscript for peer-review, which would allow them to properly reference RmYN02 in their manuscript. Holmes had promised to expedite the release of Zhou et al. Unfortunately, some re-sequencing by the Zhou et al. team was now delaying that release. But if the co-authors did not wait for the preprint release, and if Zhou et al. was released before their own manuscript was ready for publication by Nature Medicine, they risked being told by Nature Medicine that their manuscript was already dated.
Andersen suggested that ‘Instead of directly pointing to [RmYN02], we’ll make it clear that stuff like this happens all the time and that “we’d expect to see animals harboring CoVs with similar insertions as research is ongoing” — and then add a few more points to e.g., furin in human CoVs and flu.’. With the added advantage that this ‘will make us look wicked smart when the RmYNO2 paper comes out too…’
Rambaut, 27 Feb, 2:27 pm UK:
[..] The only thing that is currently unpublished, and that we need for this, is the cleavage site insertion in a bat. But the window of opportunity for publishing this, in the form it is in, is vanishing quickly.Andersen, 3:34 pm UK
I’m not too worried about not being able to publish this — yes, it’s getting to be of decreasing interest as focus moves to pandemic control, but it’s still of interest. Here’re my thoughts:
1. If the additional figure brings in too much ‘raw’ data/analysis that could be controversial, then yes, we probably shouldn’t include for a commentary
2. I will focus on reshaping / finishing the manuscript Monday/Tuesday, assuming the half-furin data will be published shortly(ish)
3. I’ll reach out to Sri at Cell to sell the story to her — that way we don’t deal with the reviewers and Cell is more likely to take it
4. We either reference to a new study showing half-furin from Eddie’s figure, OR (if that isn’t going to be out anytime soon) point to other viruses saving that furin stuff happens all the time, and we predict we’ll see the same here…!
That way we can keep the message strong, without actually citing the study — if the study comes out in the meantime, then we’ll throw a citation in. [..]Holmes, 8:11 pm UK:
Things have been a little delayed with the bat paper…they done some re-sequencing. Doesn’t change anything but it is slower. I agree with the window is closing. Why not just send to Nature Medicine today as is? That will the fastest.Holmes, 10:25 pm UK:
Not sure about the fastest. Will Nature Medicine want a review? If not them — Kristian — should we ask Sri?Andersen, 10:30 pm UK:
Hey folks. Sorry, in constant meetings today (at UCLA) and tomorrow -driving back from LAX tonight. I’l be able to find a couple of pockets of time, so let me use that to first write Nature Med to see what they’d need -review, then let’s go with Cell, Otherwise, let’s try Nature Med first — seems like most folks leaning that wayGarry, 10:35 pm UK:
I actually think the revision is not in bad shape but does need some help with transitions and the new references. I’ll stop but it needs several passes by the rest of the team. Not a long process.
Kristian, just remember — write drunk but edit sober — I need a beer or two.
Should not need a full review at NatMed — all points of the prior review addressed ~ mostly — I thinkHolmes, 10:45 pm UK:
Nature Medicine then. I’ll go over the new version of the paper this morning.Andersen, 27 Feb, 11:35 pm UK
We’ll leave out RnYN02. Instead of directly pointing to it, we’ll make it clear that stuff like this happens all the time and that “we’d expect to see animals harboring CoVs with similar insertions as research is ongoing” — and then add a few more points to e.g., furin in human CoVs and flu. Will make us look wicked smart when the RmYNO2 paper comes out too…
23.3 Submission to Nature Medicine (28 Feb — 6 Mar)
On 28 Feb (0:34 am UK), Kristian Andersen contacted João Monteiro at Nature Medicine to discuss submission there (without including the latest version of the manuscript), and informed his co-authors about it five minutes later on slack. [footnote 65]
Within two hours, Monteiro, who was at a conference for the day, quickly typed an answer expressing some strong interest. In his partly gibberish answer, Monteiro confirmed that there was no need for any new additional peer review. Effectively, he could just transfer the Nature review over.
Thanks for reaching out. Yes, we are very interested in the comment[ary], and since it’s been already peer reviewed, we were hoping to [move] ahead with [it] fairly quickly. I’m at a conference right now, back to the office tomorrow. In the meanwhile, could you send me the revised version you’re working on? I can work [on it?] that Fri[day]. The editorial side, so that when you transfer, we can move ahead with accepting it straight away.
source: Monteiro, in Proximal_Origin_Emails.pdf, p.138, Feb 28, 1:53 am UK.
The next stage was thus for the authors to get their last edits in, then for Andersen to send the latest manuscript over (v16), which he did on 1 Mar, after some last edits by the team (0:22 am UK time, confirmed 1 minute later on Slack).
On 1 Mar, Holmes was just starting a conference in the ski resort of Tahoe city (CA), possibly the ‘Global Virome in Health and Disease’ (1–4 Mar), where he still hoped to meet Clare Thomas and maybe convince her to take the manuscript in Nature [footnote 64]. By the end of the second conference day, it was clear to Holmes that Thomas was not there (unbeknown to him, she had cancelled her trip), so this plan did not pan out. The co-authors then decided to keep moving along with Nature Medicine and to just send an open-ended email to Clare Thomas, telling her of their submission to Nature Medicine, just in case.
Next, Monteiro came back on 3 Mar (6:02 pm UK) and asked for the manuscript to be cut back by about 25% before he could properly consider it for publication. That sized down version (v17) was sent by Andersen six hours later (4 Mar, 0:27 am UK). The last step was to transfer the manuscript over to Nature, which Andersen took charge off (Holmes was in LAX airport on his way back to Sydney).
By 5:29 pm (UK) on 5 Mar, the manuscript had been transferred to Nature Medicine, with some minor last edits (v18a). At 6:17 pm UK, Monteiro wrote to Andersen and Holmes that all was in order.
Last, Andersen spent some time polishing the press releases (first drafted on 24 Feb), which he shared again on Slack late a few hours later (5 Mar, 11:07 pm UK).
23.4 Sharing the news with Farrar, Fauci, and Collins (6 Mar)
The next day, 6 Mar, Andersen shared the news of the manuscript acceptance with Farrar, Fauci, and Collins and thanked them for their “advice and leadership”. He included in his email the manuscript and his press release.
23.5 Daszak is cautiously confident (17 Mar, 10:03 am EST)
On 17 Mar, a few hours before the publication of P.O. in Nature Medicine, and before the chorus of condemnations following Trump’s choice of word the previous day (‘Chinese virus’, see 24.5), Daszak sent an email to Baric, in which he ostensibly informed him of the latest developments within the OSTP (Office of Science and Technology Policy of the White House) regarding COVID-19.
In his ill-timed email, Daszak touched on where he thought the White House stood on the origin question:
➤ He had joined a new OSTP committee, the “Standing Committee on Emerging Infectious Diseases & 21st Century Health Threats”, while offering not to take part in any origins discussion that may take place. [footnote 87]
➤ He was nevertheless confident that the new Committee would not look again into the origin question, due to the White House satisfaction with:
- The ‘earlier meeting’, likely meaning the NASEM call of 3 Feb, or alternatively the first virtual meeting of the new committee on 11 Mar.
- P.O. now due to be published in a scientific journal,
- the general comments within the scientific community (to which the release of P.O as a preprint on 17 Feb would have contributed significantly).
One must remember that the OSTP is part of the Executive Office of the President, and has a broad mandate to advise the President on key science policy issues. Whether the president pays attention to the analysis to the OSTP is a different matter, but at least the recommendations of the NASEM Standing Committees that inform the OSTP represent some important consensus. In short, Daszak was quite confident that the danger was over and that the origin question would not invite itself at the White House again.
His prognostic was unfortunately way off, since — as we shall soon see — the situation was just about to change that very same day (see 24.3).
23.6 Publication in Nature Medicine (17 Mar, < 3pm EST)
On 17 Mar, a bit before 3 pm EST, Proximal Origin was published by Nature Medicine. On the same day, 5:29 pm UK, the Chief Editor of Nature Medicine hailed it on Twitter:
Monteiro (Nature Medicine), 17 Mar 5:29 pm UK:
‘Let’s put conspiracy theories about the origin of SARS-CoV-2 to rest and help to stop spread of misinformation — great work from @K_G_Andersen’.
In his tweet, João Monteiro quoted a Nature tweet posted half an hour earlier (4:56 pm UK), which incidentally called Proximal Origin a ‘paper’, when in fact it was published as a correspondence. A key sentence of Proximal Origin was also highlighted in Nature’s tweet:
Nature official account, 17 Mar 4:56 pm UK:
New paper in @NatureMedicine: ‘Our analyses show that SASR-CoV002 is not a laboratory construct or a purposefully manipulated virus’.
Insight: A question of timing
The success of Proximal Origin (P.O.) in its first week, as well as some of the reactions that followed (such as Monteiro’s email above), should be considered in the light of the ‘Chinese virus’ controversy that started on the very day of publication (17 Mar).
P.O. was the right paper at the right time to contrast Trump’s annoying utterances with the alleged purity of a scientific analysis. The ‘Chinese virus’ controversy thus acted as great accidental marketing.
Incidentally, when reviewing the Altmetric score of P.O. on 19 Mar, Andersen could not help but make a joke about the controversial ‘Chinese virus’ tweet from 16 Mar evening’.
“Hey @realdonaltrump, here’s the evidence you have been looking for — it’s totally the Chinese Virus! #MAGA”. Yeah?
23.7 The NIAID Office of Communications is informed (17 Mar, 3:01 pm EST)
Also on 17 Mar (3:01 pm EST), following the request from Coleman on 19 Feb (see 18.15), Shabman let Coleman know that the paper was now published by Nature Medicine. Coleman in turn alerted the NIH/NIAID Office of Communications.
23.8 Scripps Press Release (17 Mar)
On the same day, the Scripps Research Institute issued a press release about Proximal Origin that left absolutely no ambiguity about the necessary natural origin.
In particular, it quoted Andersen:
“By comparing the available genome sequence data for known coronavirus strains, we can firmly determine that SARS-CoV-2 originated through natural processes.”
“These two features of the virus, the mutations in the RBD portion of the spike protein and its distinct backbone, rules out laboratory manipulation as a potential origin for SARS-CoV-2”
The contrast between Andersen’s statement and the headline of the press release is to be noted:
Andersen’s words:
‘[W]e can firmly determine that SARS-CoV-2 originated through natural processes.’Press release headline:
‘The COVID-10 coronavirus epidemic has a natural origin, scientists say.’
Andersen’s words were about the origin of the virus itself, while the Scripps press release are about the origin of the epidemic.
In particular, Andersen’s choice of words left entirely open the possibility of a lab escape of a natural virus handled there, or an infection of a researcher on a sampling trip, which are a form of lab origin and were simply not covered in P.O. (see 22.5). Hence, even before judging the scientific validity of Proximal Origin itself, the headline of the Scripps press release was already problematic.
Sideline: Something for everyone
Immediately following the publication of P.O. in Nature Medicine, the situation offered something to all the main actors:
👉🏻 Farrar had already received his ‘go-to statement’ on 17 Feb, in time to facilitate the work of the WHO mission that arrived in China on 16 Feb. With the publication in Nature Medicine on 17 Mar, his efforts were eventually rewarded by an official stamp, albeit later than it would have expected having initially counted on a fast publication in Nature (by the end of February at worst). But meanwhile, he had at least arranged for Daszak letter to be published in Lancet on 18 Feb, in just two days. [footnote 86]
👉🏻 Daszak could start feeling safer, after so much defensive work (the Lancet Letter and the NASEM answer to OSTP), when it seemed that passaging would still be left in the final version of PO.
👉🏻 Holmes had been truly essential in January 2020, and now had his name on multiple pangolin papers. He had shown his value as an intermediate between the Western and Chinese scientific communities, and had diligently helped Farrar getting his ‘go to statement’, occasionally gently herding the co-authors in the right direction and at the right pace.
👉🏻 The Fouchier ‘krewe’ now had both the deliberate engineering and passaging options closed.
👉🏻 Fauci got a convenient disculpating research piece, which he could then use to try to hide any skeleton in his NIAID cupboards. He most likely did not plan for this exact turn of event, but expertly exploited the situation and vulnerabilities of the other actors. Soon he would make the most of it, especially when he memorably cited P.O. from the very pulpit of the White House, at the invitation of Trump, in March 2020 (see Key Insight under 26.4)
24. CONTEXT: White House vs. China Wolf-Warriors (6–17 Mar)
24.1 Trump shows little interest in the origin question (to 12 Mar)
For context, we must make it clear that president Trump had been rather conciliant with China until that specific point in time.
As we saw in 3.1.c, until early March Trump was focussing on trade deals with china, which — as usual for Trump — meant waxing lyrical about Xi Jin-Ping if need be, and not being ‘distracted’ by the outbreak. If Trump was being criticised in the second week of March, it was instead for down-playing the risk of a pandemic, not at all for bludgeoning China with it.
The person who had been the object of accusations of conspiracy theories about the virus origin was Tom Cotton, not Trump (see 17.2) — and that was in early February, with the agitation at the time contributing to the publication of Daszak’s letter in Lancet and to the release of a P.O. on virological.org. Since then, the Cotton controversy had died down, and not much had happened on the domestic ‘conspiracy’ front.
24.2 Pompeo gets frustrated by China’s lack of transparency (6 Mar)
As the pandemic started taking its toll on the US, and as the chances of a trade deal faded away, the second week of March saw some tensions between China and the US. A typical blame-game started replacing the cooperative game played by Trump until now.
Justly frustrated by the inability of the US to obtain proper data from China, despite the patience it had shown in its support of the WHO to which its efforts had been redirected (see 8.8), Pompeo used the term ‘Wuhan virus’ on March 6.
In a CNBC interview, Pompeo said it has proven “incredibly frustrating” to work with the Chinese government to obtain data on the coronavirus, “which will ultimately be the solution to both getting the vaccine and attacking this risk.”
“Remember, this is the Wuhan coronavirus that’s caused this, and the information that we got at the front end of this thing wasn’t perfect and has led us now to a place where much of the challenge we face today has put us behind the curve,” Pompeo said in an interview with CNBC’s “Squawk Box.”
“That’s not the way infectious disease doctors tell me it should work. It’s not the way America works with transparency and openness and the sharing of the information that needs to take place.”[..]
“We have high confidence, too, that there was information that could have been made available more quickly and data that could have been provided and shared among health professionals across the world,” Pompeo added. “It’s most unfortunate.”
source: CNBC article, 6 Mar 2020
Whatever we may think of him, Pompeo’s whole argument about the lack of transparency from China was factually correct and timely relevant. The main problem is that the public at that time was not fully aware of the details of Chinese obfuscations, of the unsuccessful US efforts to get isolates and proper data, of the even more frustrated WHO that had compromised itself so much for an elusive access. It would take much more time for this to come to the surface.
Meanwhile, China would play its best defence, which was to go on the offence, with the help of some rather biased reporting in US media.
24.3 China wolf-warrior Lijian Zhao blames the US (12–13 Mar)
China went on a full offensive, when, on 12 Mar 2020, Foreign Ministry spokesman and wolf-warrior Zhao Lijian suggested that the US army might have brought the epidemic to Wuhan’ and asked for the US ‘to be transparent’. His words were relayed extensively by the Chinese official media.
The next day (13 Mar), Chinese ambassador Cui Tiankai was summoned to the State Department for some official protest.
24.5 Trump starts slowly shifting (16–17 Mar)
As China upped the ante, the first sign of a potential change in Trump’s attitude happened on the evening of 16 Mar (6:51 pm EST), when for the first time Trump had tweeted about the ‘Chinese virus’:
Trump explained soon after that:
“China was putting out information, which was false, that our military gave this to them. And rather than having an argument, I have to call it where it came from. It came from China.”
source: Telegraph UK article, 17 Mar 2020
24.6 A chorus of condemnations (from 17 Mar)
The Chinese government would react angrily on 17 Mar. The US media immediately joined in a chorus of condemnations, with the NY Times, the New Yorker, the Atlantic, ABC, and many others indiscriminately brandishing Trump’s tweet as an example of every day anti-Asian racism. To make things worse, China’s Foreign Ministry announced that same day (17 Mar) that it would expel US journalists working for the NY Times, Wall Street Journal and Washington Post.
25. The Nature Medicine version
25.1 Comparison with the virological.org version
Here we shall quickly review the main changes between the version that was released on virological.org on 17 Feb (12b), and the text published in Nature Medicine exactly one month later (v19).
a. FCS feature analysis:
In line with the plan agreed on 25–27 Feb (see 22.6 and 23.2), since the RmYN02 preprint (Zhou et al.) was still not yet released when Andersen sent the cut-down version of P.O. (v17) to Nature Medicine on 4 Mar for final approval [footnote 68], that version (and subsequent ones) contained a new sentence in the ‘Natural selection in an animal host‘ section, which hinted to the possible natural evolution of the FCS without mentioning RmYN02:
‘Mutations, insertions and deletions can occur near the S1–S2 junction of coronaviruses, which shows that the polybasic cleavage site can arise by a natural evolutionary process’
b. RBD feature analysis:
Two important changes were made:
b.1 The ‘Receptor-Binding-Domain’ section was changed in a way that gave it a rather different meaning:
‘This is strong evidence that SARS-CoV-2 is not the product of genetic engineering’
became
‘This is strong evidence that SARS-CoV-2 is not the product of purposeful manipulation’.
That was an unfortunate change, at it made it sound as if this argued against passaging alongside deliberate engineering, when the supposed evidence in that paragraph did not apply to passaging. [footnote 85]
b.2 The ‘Selection during passage’ section was changed, to use the RBD as an argument against passaging, by stating that recombination and mutation was a ‘much stronger and more parsimonious‘ acquisition process for the RBD, compared to passaging.
c. Conclusions section (RBD + FCS):
Last, the ‘Conclusions’ section was also altered to reflect the changes to the RBD and FCS features analyses. The section in bold below was added (starting with v16):
‘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.’
25.2 Confusion in the Conclusions
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.’
That sentence above was the climax of Proximal Origin, and has raised some strong criticism ever since. It is difficult to think of a more cited and downloaded paper where the most important sentence is actually so vague, so illogical, and interpreted in multiple ways by scientists trying to deal with its limitations.
Lipkin, as a co-author who went through the editing process, put these issues nicely in his House testimony:
‘There are some nuances in the way the paper was, you know, finally assembled and printed that are, I think, less than ideal.[..]
a. Bad logic
As previously noted in Key Insight, 22.3, the main issue with the above sentence is that laboratory work on coronavirus in China would precisely aim to address the pathogenic risk from existing features of related coronaviruses.
There is indeed no point developing a vaccine or some drugs against some viral features that are very unlikely to evolve naturally. There was also no point developing a vaccine purely against SARS-1, since that virus had never reappeared in 16 years, and would have likely evolved away since then.
Vaccines/drugs developments are both health and commercial decisions: they won’t find a market if there is no real application. What makes sense instead is to develop vaccines or drugs against the likely evolution of the pathogenic front, based on known pathogenic features in related viruses; an evolution that may have already been realised but was never sampled, or which may happen within a few years.
Hence, this anticipated evolutionary front target, materialised via genetic construct and/or passaging, would typically look like a composite of features present in existing coronaviruses, and would most likely remind one of a mix of optimisations and complex recombinations.
In simpler words: observing ‘all notable SARS-CoV-2 features’ across various coronaviruses in nature (even if in a very rudimentary way for the FCS) is consistent with both the zoonosis and the research accident hypotheses. Per itself, that does not make one hypothesis more likely than the other one.
b. Backbone: when Holmes knew better
Another issue with the above sentence has to do with the treatment of the backbone:
It is improbable that SARS-CoV-2 emerged through laboratory manipulation of a related SARS-CoV-like coronavirus. Furthermore, if genetic manipulation had been performed, one of the several reverse-genetic systems available for betacoronaviruses would probably have been used. However, the genetic data irrefutably show that SARS-CoV-2 is not derived from any previously used virus backbone.
source: Proximal Origin (Nature Medicine).
We have already noted that most of the viruses sampled after 2016 have not been reported, as has been confirmed by some knowledgeable scientists [footnotes 55, 56]. What is rather striking is that Holmes himself was well aware of that fact, and hence of the weakness of the backbone argument. His doubts were somehow reserved to private conversations:
c. Dancing around the issue
It seems that at some stage the co-authors were actually aware of the limitation of their reasoning, as v17 was more in line with the Slack discussion, when it substituted ‘necessary’ for ‘plausible’: ‘we do not believe that any type of laboratory-based scenario is necessary’.
‘Necessary’ is fundamentally correct, but — unfortunately — it would clearly not allow arguing against a non-natural origin, since ‘non-necessity’ has never stopped an accident from happening. Probably for that reason, it was quickly changed back to ‘plausible’ in version v18, despite its own limitations.
d. Scenarios ignored
Last, as noted previously with the Scripps press release (23.7), P.O. cannot exclude a sampling trip infection, or a lab escape of a natural virus, two additional forms of research-based accidents.
These scenarios were purposefully neglected based because they were deemed to be indistinguishable from a zoonosis, in the sense that the responsible virus is to be found in nature. But neglect does not magically transform these occupational research hazards in random zoonotic events.
e. A Rorschach test
Beyond the scenarios that were ignored, there is a fundamental ambiguity in the ‘any type of laboratory-based scenario’ when it comes to passaging. Lipkin himself would later argue that P.O. was badly worded and could only reject one type of laboratory-based scenario: deliberate engineering.
That one of the authors of the paper contradicts so much ink later splashed in the media by journalists and supposed experts is rather telling of the battleground that P.O. has become. (Lipkin would not stop there, and just like Baric, he has also been very dismissive of the market jump(s) hypothesis later pushed by ‘armchair epidemiologists’ [footnote 27])
‘There are some nuances in the way the paper was, you know, finally assembled and printed that are, I think, less than ideal.[..]
So we excluded the possibility of deliberate design for the reasons I’ve just said. [..] Given the number of animals that were present in the market, this would be a higher probability event. And the possibility that somebody — that it might have occurred as a research-related accident was open and could not be entirely excluded.’
source: 2023.04.06-Lipkin-Transcript.pdf
25.3 Confusion in the Approach
The finding of SARS-CoV-like coronaviruses from pangolins with nearly identical RBDs, however, provides a much stronger and more parsimonious explanation of how SARS-CoV-2 acquired these via recombination or mutation.
The parsimony claim above, introduced in the ‘Selection during passage’ section, is illustrative of a methodological shortcut that is worth exposing, given how often that error is made.
👉🏻 The ‘much stronger and more parsimonious’ argument is a statement about an uninformed rough estimate of relative odds of the two hypotheses (zoonotic and research-related accidents).
👉🏻 A proper evaluation of these odds should instead start from (i) the risk factors attached to both hypotheses (as generative processes) and (ii) the particular details of the outbreak (the informed/posterior probabilities).
In contrast to that proper investigative approach, the parsimony claim of Proximal Origin is a semantic honey trap that lures people towards a wrong conclusion along their path of least resistance:
- One shall first ignore a-posteriori probabilities, attached to the actual location, timing, and risk factors of the research being done, by smugly declaring that they are useless coincidences.
- Then using only the much more intuitive and incomparably simpler uninformed a-priori probabilities, and going back to time immemorial (which is totally irrelevant to the actual balance of risk factors), one can focus purely on the zoonosis hypothesis and declare that it is the obvious natural order of things.
In the end, the liberal use of ‘much stronger and parsimonious’ without any attention to the actual risk factors, or to all other information crucial to the posterior probabilities, is either accidental bad science, or of a deliberate sleight-of-hand.
26. Post publication doubts (14 Apr–18 Apr)
26.1 Initial euphoria (17 Mar–14 Apr)
Immediately following the publication of Proximal Origin, the co-authors started paying attention to its Altmetric score. Andersen was first out of the gate, checking the score before the end of his day (17 Mar, 11:42 pm California), then again the next evening as the score started picking up strongly.
By the end of the month, the paper had been accessed three million times, making it exceptionally successful. Until the 14 April, the otherwise now quiet Slack discussion channel became largely an outlet for the euphoria of the co-authors, but also for their annoyance at being bombarded by cranky emails from an odd assortment of readers.
26.2 The Embassy cables (14 Apr)
On 14 Apr, the Washington Post published an Opinion piece that related how some State Department cables were adding to the conviction, within that part of the administration, that a lab-origin was plausible, if not probable.
Here were essentially the same questions as the ones raised by Tom Cotton earlier, but this time coming from within the administration. And instead of being shot down in flame as the Post had done earlier for Tom Cotton, and despite the resounding success of Proximal Origin, the Post was giving some proper considerations.
“The idea that it was just a totally natural occurrence is circumstantial. The evidence it leaked from the lab is circumstantial. Right now, the ledger on the side of it leaking from the lab is packed with bullet points and there’s almost nothing on the other side,” the [senior administration] official said.
[..] We don’t know whether the novel coronavirus originated in the Wuhan lab, but the cable pointed to the danger there and increases the impetus to find out, [Xiao Qiang, a research scientist at the School of Information at the University of California at Berkeley] said.
“I don’t think it’s a conspiracy theory. I think it’s a legitimate question that needs to be investigated and answered,” he said. “To understand exactly how this originated is critical knowledge for preventing this from happening in the future.”
source: Washington Post piece by Josh Rogin, 14 Apr 2020
26.3 Collins and Fauci react to the White House news (16–17 Apr)
On April 16, 2020, Collins shared with Fauci a report by Fox News host Bret Baier, which stated that ‘multiple sources’ believed that COVID-19 emerged from a lab in Wuhan, based on classified and open-source elements presented during some origins briefing. Collins deplored that Proximal Origin had not stopped this ‘very destructive conspiracy’, against his hopes, and wondered if the National Academy (NASEM) should not step up again.
Fauci responded on April 17, saying, ‘I would not do anything about this right now. It is a shiny object that will go away in time.’
26.4 Fauci pushes P.O. from the White House podium (17 Apr)
Later that very day, April 17, immediately after Trump expressed some well-deserved criticisms of China for being non-transparent, and for the WHO for being too weak in its handling of China, Fauci was asked about a possible non-natural origin and took the occasion to pitch Proximal Origin from the White House podium.
Trump:
We have a lot of discussions going on with China. Let me just put it this way: I’m not happy, okay? I’m not happy. And I spoke to them. And this could have been shut down a long time ago. They knew it. And we couldn’t get in. And, in all fairness, World Health couldn’t get in, and that’s why I wish they took a different stance. They took a very pathetic stance and a very weak stance. But they say they couldn’t get in.
But, ultimately, they got in; they got in much sooner than anybody, but they didn’t report what was happening. They didn’t report what was happening inside of China. No, I’m not happy with China.Trump. to the next reporter:
Yeah, please.Reporter (Kevin Freking):
Mr. President, I wanted to ask Dr. Fauci: Could you address these suggestions or concerns that this virus was somehow man-made, possibly came out of a laboratory in China?Trump, to Fauci:
Want to go?Reporter, to Fauci:
You studied this virus. What are the prospects of that?Fauci:
There was a study recently that we can make available to you, where a group of highly qualified evolutionary virologists looked at the sequences there and the sequences in bats as they evolve. And the mutations that it took to get to the point where it is now is totally consistent with a jump of a species from an animal to a human.
So, I mean, the paper will be available — I don’t have the authors right now, but we can make that available to you.
Key Insight: So many layers of deception
It is remarkable that Fauci managed to undermine the White House enquiry into the origin, just two days after it was disclosed, right at the White House, and standing close to the president who had invited him to answer a question about the origins. It made the White House and the president look like a bit like fools.
There is also some irony in the fact that Trump — whatever his many faults — was not particularly wrong about the ‘very pathetic and [..] very weak stance’ that the WHO took until it was eventually allowed in China on 16 Feb, except — that is — that he had himself been praising China all the way to middle-March, with an eye on some trade concessions.
We may further wonder what the president’s reaction would have been, had he known that the very paper that Fauci was promoting from the White House podium, at his invitation, on that 17 Apr, had been quickly written and first released on 17 Feb, precisely to appease China and facilitate the WHO pretend access to the country, with full support from Fauci, who also had some secondary benefits to draw from it.
To make things worse, part of that need for appeasement was originally triggered by an inopportune remark of Fauci himself about the ability and readiness of the U.S. to look into a non-natural origin, in an ABC interview published on 6 Feb (see 8.4)
In other words, the Nature Medicine piece, that Fauci was now pushing, was essentially a by-product of the failed WHO strategy that Trump had rightly criticised a few seconds earlier, to which Fauci himself had contributed to, both willingly and accidentally.
26.5 Andersen expresses some doubts (17 Apr)
On 16-17 Apr, the P.O. co-authors shared many articles on the subject of the White House renewed interest into the origins.
Soon after (17 April 9:57 pm UK), an important conversation took place on Slack; in the process of trying to make sense of the latest news, and in particular to evaluate if the White House ‘might be onto something’, Andersen articulated clearly that he was still wondering about (1) passaging, and (2) an engineered virus.
The analysis he offered on that occasion mirrored rather well some of the arguments that have since then been put forward by an ever-growing number of scientists:
- On engineering, after initially dismissing it (his case 1: ‘totally off the table’), Andersen backtracked in that last paragraph and correctly pointed to the possibility of leaving no signs by using Gibson assembly and setting up a reverse genetic system, both fairly easy.
- On the possibility of the virus having been collected in a wild, and before infecting a researcher handling it (his case 2), Andersen seemed to be aware that the main argument that P.O. rolled out to exclude that possibility, was very weak per se (i.e.: ‘one would hope’ that they had very little chance of coming across the virus in the wild), even before considering other issues. [footnote 66]
- On passaging in cultures (his case 3), it is very significant to see Andersen pointing to the fact that the glycan argument, that was used in Proximal Origin (P.O.) to weigh against passaging, was actually weak (‘they could actually play a different role’, and indeed that glycan argument has been invalidated since then).
Referring to what he clearly saw as more virtue-signalling than science, he added that, like so many others, he ‘really really want[ed] to go out there guns swinging saying “don’t be such an idiot believing these dumb theories — the president is deflecting from the real problems”, but that he simply could not do so.
In other words, Andersen was unable to rule out any of the lab-product scenarios at all. The doubts that had publicly re-emerged with the embassy cables article could not be easily dismissed.
With the entry on the scene of the intel services, some with excellent resources in epidemiology and biotechnologies, and with so much of the media showing a keen interest in the debate, Andersen may also have started to feel the heat. What had been a narrative initially meant to support Farrar and the WHO, and then weakened further to allow for publication in Nature Medicine against some strong headwinds, was starting to look a bit shaky under the investigative spotlights.
Andersen, 17 Apr 2020, 9:57 pm UK:
Okay, so about the current news, Is there any reason to believe that they might be onto something, or is it all smoke and mirrors? [..] The main things from my perspective:1. Bioweapon and engineered totally off the table
2. If there is no engineering and no culturing, then it means that somebody magically found a pre-formed pandemic virus, put it in the lab, and then infected themselves. The prior on that vs somebody coming into contact with an animal source infected with the virus is as close to zero as you can get. Humans come into contact all the time with SARS-like CoVs, but the likelihood of somebody finding exactly that pandemic virus and infecting themselves is very very low (make no mistake — if they did find that pandemic virus, then they would get infected if they grew it in the lab — but the likelihood of them finding it in the first place is exceedingly small (or so one would hope — otherwise, good luck World avoiding future pandemic).
3. But here’s the issue — I’m still not fully convinced that no culture was involved. If culture was involved, then the prior completely changes — because this could have happened with any random SARS-like CoV, of which there are very many, So are we absolutely certain that no culture could have been involved? What concerns me here are some of the comments by Shi in the SciAm article (“I had to check the lab”, etc.) and the fact that the furin site is being messed with in vitro. Yes, it loses it, but that could be context dependent. Finally, the paper that was shared with us showing a very similar phenomenon (exactly 12bp insertion) in other CoVs has me concerned [..]
I really really want to go out there guns swinging saying “don’t be such an idiot believing these dumb theories — the president is deflecting from the real problems”, but I’m worried that we can’t full disprove culture (our argument was mostly based on the presence of the O-linked glycans — but they could likely play a different role [..].
We also can’t fully rule out engineering (for basic research) — yes, no obvious signs of engineering anywhere, but that furin site could still have been inserted via gibson assembly (and clearly creating the reverse genetic system isn’t hard — the Germans managed to do exactly that for SARS-CoV-2 in less than a month)
26.6 Holmes tries to push back, but fails (17–18 Apr)
a. First attempt:
To try to address Andersen’s concerns, Holmes replied soon after (17 Apr 11:38 pm UK) with five new arguments, on top of the initial ones he had already deployed after learning of RmYN02 (see 22.4). None of these new arguments makes much sense; they instead show an odd mix of conceit, motivated reasoning and wishful thinking:
- George Gao (Chinese CDC director) is ‘too dumb’ to set up a complex cover-up. Hence, the virus truly first appeared in Wuhan, at the market.
- Engineering, if any, would be done by Baric in the US, not in Wuhan.
- Passaging viruses in culture can’t make them more human adapted.
- It is possible that SARS-CoV-2 evolved in nature. So one should not discuss a non-natural origin ‘because’ it is not needed.
- If one allows for the possibility of a non-natural origin, Ebright, Lipsitch and Bergstrom will try to use this to push for a control of GoF research.
To his credit, Andersen replied with patience to that child’s talk. He simply reiterated the key point that culturing SARS-like viruses at BSL-2 was a main concern — ‘no question about it’. He then proceeded to give various example of such culturing work at the WIV (17 Apr, 11:51 pm UK):
Andersen, 17 Apr, 11:51 pm UK:
Shi didn’t do any GOF work that I’m aware of — but GOF work isn’t the concern here. She did A LOT of work that involved isolating and culturing SARS-like viruses from bats (in BSL-2) and that’s my main concerning scenario (we cite several of those in the paper — if you have a lock at those original publications, it’s definitely concerning work, no question about it — and is the main reason I have been so concerned about the ‘culture’ scenario).source: Proximal_Origin_Slack.pdf, p.129
b. Second attempt:
Answering Andersen, Holmes tried yet another approach (18 Apr, 0:05 am UK):
Holmes, 18 Apr, 0:05 am UK:
Let’s face it, unless there is a whistleblower from the WIV who is doing to defect and live in the west under a new identity we are NEVER going to know happened in that lab. Never.
In other words, Holmes argued that, unless one had a chance to clearly prove the origin hypothesis, one should not discuss it and get in the middle of a very controversial field, however plausible the hypothesis may be; better take the safe and easy road. This was just another rewording of his argument #4.
Andersen, once again, was not very impressed by that brand of rationalism, and immediately replied that one should be careful not discount too quickly the embassy cables that had started the news cycle on 14 April. It seems likely that Andersen was afraid that cables could reveal irresponsible BSL-2 work at the WIV.
Andersen, 18 Apr, 0:06 am UK:
‘That’s my thinking too. But that’s why I’m a little worried about these ‘cables’ — because is it possible that they might have something? I’m putting all of this to typical Trump BS smoke and mirrors (and just plain idiocy), but I’m not quite willing to die on this hill,Holmes, 18 Apr, 0:48 am UK:
Yes, I’m not dying on a hill either.
c. Third attempt:
A few hours later, still undeterred, Holmes came back yet one more with some arguments against a research-related accident, all equally weak when not fundamentally wrong:
- Shi and her group have published the sequences of all the viruses they have collected.
- There is no evidence of an initial outbreak at the WIV, and the WIV would not be able to hide an outbreak.
- A wet market is not the logical place for a research-related outbreak, it should instead be a much more crowded area.
- Why would Shi publish RaTG13 if the WIV is the source of the outbreak?
Andersen was not convinced and simply politely reiterated his reservations, while trying to sound conciliatory.
Andersen, 18 Apr, 4:58 pm UK:
[..] Again, I’m pretty damn sure this is all smoke and mirror, but I’d need to see those actual cables before I put my head on the block 🤪.
Sideline: Holmes’ hurt feelings vs. Andersen’s realism
Looking back at this tit-for-tat exchange over two days, what is rather obvious — especially compared to the remaining of the Slack — is how conflicted, if not hurt, Holmes felt when Andersen detailed his doubts about the main thesis of Proximal Origin, and would not back off.
There is a bit of a diva in Holmes — he certainly did not take well to pointed criticisms. Maybe Holmes had issues with the fact that, spurred by his own involvement in pangolin papers with mainland scientists [footnote 33], he had obliged and provided a perfect reason for Farrar to decide to go for publication on 8 Feb — a decision that did not look so smart after the embassy cables news.
As to the valid points that Andersen made in his replies to Holmes, we can only imagine what his reaction would have been had he known at the time about DEFUSE, and specifically about Daszak’s intention to see dangerous coronavirus work, including passaging, being done at BSL-2, precisely in Wuhan, and precisely at the right time [footnotes 25, 26, 59, 61].
Andersen was on something.
26.7 Altmetric High (end of April)
That uncomfortable conversation between Holmes and Andersen ended up with the co-authors opting to focus instead on more positive thoughts, namely the Altmetric score of P.O., which seemed to be on its way to reach #1 again.
In the end, P.O. had indeed become the wished ‘go to scientific statement to refer to’. There was no simply no turning back, and whatever the doubts they would keep resurfacing, these were not allowed to disrupt a very successful narrative.
26.8 Recap: How did we end up there?
The co-drafters started with the idea of finding possible signs of engineering, then went on to write a report on these to initiate some face-to-face discussions at the WHO level. However, they quickly found themselves in a dead-end:
- It is possible to engineer a virus without leaving traces (or at least obvious ones, using Gibson assembly (26.5)), so arguing for engineering based. on the genome becomes particularly difficult (at the very least, it would require special forensic skills and expertise).
- They were somehow not considering a mix of gene editing and passaging, which is precisely the recipe recommended by Baric.
- It is very difficult to tell the difference between natural evolution and passaging. Again, only forensic skills may be able to give some colour there.
- An accidental release of a natural virus is not considered either, which unless a whistleblower comes up, would also be near impossible to substantiate
In the end, what Andersen, Garry, Rambaut and Holmes leant, and what Fouchier knew perfectly already, was that it would be very difficult to find clear proofs of engineering or passaging. At most, they would be hints. From that point, the co-drafters did find themselves on difficult grounds they did not expect.
At the same time, they were being overtaken by the blanks, through a pinching movement led by the WHO and NIH/NIAID:
- The WHO was from day one only too happy to give a free pass to China, to get access, which was essential to better understand and be able to contain the pandemic, and without which no origin may well never be proven anyway. The key person who relayed the WHO imperative was Farrar, whose role was essential in shaping P.O., in giving a platform to Daszak’s Statement of Support in Lancet, and — in the end — in durably shaping the zoonosis narrative.
- NIH/NIAID progressively started signalling its preferred narrative, with Morens (Fauci’s right arm), pushing out a paper defending a pure zoonosis, and Fauci joining Daszak on a podcast where he blurred the lines, then coming out gradually in support of the natural origin narrative.
To get out of that kill-box, either consciously or after persuading themselves, the co-drafters set two very distinct burdens of proofs for the two main hypotheses:
- A zoonosis should be retained as the only possible origin if it could be approximately reconciled to the genome and little data available.
- Conversely, a research-related accident should be considered as a possible origin if the genome and little data available strongly pointed to that possibility, away from a zoonosis.
That drift in the burden of proof is the main issue, and has created some logical confusion in the origin debate.
For instance, many zoonosis proponents argue correctly that the Slack conversation shows a progressive willingness to ditch the passaging argument, before submitting again to Nature Medicine. However, as proponents of a possible research accident have pointed out, Andersen’s doubts actually persisted past publication. But more importantly, and that’s where the key is, that removal of passaging was essentially caused by an asymmetrical burden of proof, largely imposed by the circumstances.
And naturally, given how asymmetrical that burden was, Andersen would keep coming back and wondering about passaging, independently of whatever assurances were published in P.O. He basically kept mentally rewriting the paper that could not be published. Hence, the oft-repeated interpretation of the Slack from the zoonosis side is technically right on that point, but totally wrong on substance.
Reciprocally, the research-accident side often could better acknowledge the growing willingness of the co-drafters to drop the passaging hypothesis as they moved towards publication in Nature Medicine, even if it followed a chaotic path, bouncing off any new rare piece of information. But the research-accident side is perfectly right on the main take-away: in the end, P.O. drop of the passaging hypothesis is wrong (due to much higher burden of proof eventually set on it, and the still limited data that was available at the time), while other accident scenarios were simply not even considered (no mix of engineering and passaging, no proper argument against an accidental release of a natural virus).
Sideline: An evaluation
In the end, the main issue with P.O. is the one that Andersen identified perfectly early on: it is acceptable as opinion piece or as working notes, with limited diffusion, written in a fast changing environment, but it should never have been published, especially not under pressure, and especially not just to be hailed as the definitive ‘go to publication’.
That P.O. was published and pushed aggressively had all to do with immediate external imperatives, and nothing to do with pure science.
The predictable result is that P.O. achieved very little against its initial imperatives (China never cooperated, the NIH ended up compromised), while delivering mostly negative scientific outcomes since its publication four years ago (obstruction of key papers [footnote 44], stifling and radicalisation of the scientific debate), and while promising to do so for a few more years to come (discredit of science and scientists).
27. Egomaniac strikes back (Jul 2020)
27.1 A strange email (25 Jul)
On 25 Jul 2020 at 7:22 am EST, Jon Cohen at Science received an anonymous email entitled ‘The authors who wrote the paper saying that SARS-CoV-2 is not human-engineered first tried convincing Anthony Fauci of the opposite’.
To paraphrase the email, the would-be whistleblower’ explained that, instead of being glorious scientists, the authors of Proximal Origin (save Lipkin) were clueless ones, who had to be educated by some true coronavirus experts invited on the call by Farrar, who proceeded to show them that their lab-product fears were crackpot. From there, any credit for the remarkable success of P.O. should be attributed to these true coronavirus experts, no to Andersen, Holmes, Garry or Rambaut, who are simply impostors.
The whistleblower added that Nature refused to publish Proximal Origin when it was informed of that supposed misrepresentation of the authors’ contribution. The whistleblower’s email shows some second-hand knowledge of the call content, and states that the story was backed by two people (a standard for a journalist).
27.2 A prompt answer (27–28 Jul)
Instead of taking the bite, Jon Cohen shared the email with Andersen and Holmes instead (27 Jul, 6:02 pm EST). Andersen quickly sent it to Fauci with Farrar in CC (27 Jul, 9:05 pm EST).
Farrar sent some questions back to Andersen (28 Jul, 2:54 am EST), before Holmes, Andersen and Farrar drafted a reply to Cohen. In the process, Farrar asked for advice from his ‘comms person’ at Wellcome, who provided some suggestions the same day.
Once the answer was ready, it was sent back to Cohen (28 Jul, 3:43 pm EST). It took a mere 24 hours, with everybody in the loop, including Fauci and Wellcome. And nothing more happened.
27.3 Fouchier and Koopmans: the ‘krewe’
We also earn from the email to Jon Cohen that the denunciator ‘work[s] in the field and ha[s] heard this story from two people who were on the initial call with Fauci’.
In the Slack conversations, Holmes suspected Ron Fouchier was part of these two. In all probability, these were Fouchier and Koopmans, without necessarily foreseeing how their confidence would be misused. Hence, Holmes (and Rambaut) were very likely guessing right.
27.4 Looking for the motive
The key to understand the denunciation motives is to notice that the alleged revelations about the P.O. drafting process would not jeopardise P.O. conclusions. Indeed, the point raised with Jon Cohen is not whether P.O. conclusions are valid or not, but who is actually responsible for the conclusions and should get the credit.
So that leaves only two possible motives:
- Someone who is truly upset by the wrong attribution of the main P.O. thesis to its listed authors, while the main contributors (Fouchier, Drosten) are unlisted.
- Someone who is interested in hurting the P.O. authors by having the authorship exposed, and not in the authorship issue itself.
27.5 Not Baric
On Slack, Holmes suggested that Baric was the would-be whistleblower. Holmes’ suggestion is most likely way off the mark; it should be clear that Baric is a very unlikely candidate for either of the two possible motives. [footnote 79]
1: Baric being vexed about authorship?
That does not sound right at all:
- Authorship issues happen all the time, as in this well-known case where Baric himself is an undisclosed contributor, and nobody is much bothered about that — the rules seem rather elastic.
- Anyway, Baric would be very unlikely to stoop down to that level; it is not in his character and certainly not worth the risk.
2: Baric interested in hurting the P.O. authors?
We have all reason to believe Baric when he denied being the author of the e-mail to Jon Cohen during his hearing on 22 Jan 2024.
Why would Baric do that? He had no axe to grind with the P.O. authors, no reason at all to try to hurt them. And why would Baric ever write an email to Cohen that starts by enthusing about the work done at the WIV, at what Baric knew were BSL-2 conditions, before pushing for a resumption of EHA grants? In fact, Baric would instead soon publicly ask for a proper investigation, including of the WIV, because of their incorrect approach to biosafety in the work done in partnership with EHA.
- In November 2020 he would state on Italian TV that access to the WIV databases would first be required to determine if SARS-CoV-2 was a chimera.
- In May 2021, after the very controversial report of the US mission to Wuhan (to which Daszak and Koopmans took part) he would sign the Science letter asking for a proper investigation of the origins of COVID-19 (published 14 May 2021).
- On 27 May 2021, following the Lancet letter which triggered and an email from Daszak arguing that the WIV was working with adequate biosafety measures at BSL-2, Baric excoriated Daszak in the clearest possible terms.
- In July 2023, a few days after the Biden administration produced a very short declassified summary of the Intelligence Community position on the origin question, he expressed his frustration with the continued lack of thorough investigation, was very critical of the WHO report of 2021, and dubious of the biosafety measures (BSL-2) used in China for coronavirus studies.
27.6 Not an authorship issue
More generally, the problem with the authorship motive is that it does not make much sense.
The call attendants who pushed hard for a purely natural origin were Fouchier and Drosten. None of them really cared much for authorship.
- Fouchier’s objective was to get block any mention of a possible passaging origin, once Farrar ignored his plea not to make this public. That goal was eventually achieved, with the help of a very Fouchier-minded reviewer at Nature who rejected the paper, and effectively constrained the authors to further dilute their conclusions before publication in Nature Medicine.
- Drosten was involved during the call, but very detached after that. His personality and very different concerns at the time do not suggest at all that he would get into an authorship fight.
As to a friend or colleague of Fouchier or Drosten taking it on him/herself to denounce the authorship issue, while Fouchier and Drosten did not care for it, is even less likely. Authorship issues happen all the time — why would an external observer go to that length to expose it? What’s in it for such a person?
Note:
The remaining sections make no firm statement as to who hides behind the anonymous writer to Jon Cohen, as no hard proof exists. They do not represent attribution, but instead illustrate with an example how one may go about such a task.
27.7 A first clue
A first clue about the whistleblower’s identity may be in the title of the email itself: ‘The authors who wrote the paper saying that SARS-CoV-2 is not human-engineered first tried convincing Anthony Fauci of the opposite’.
- Why insist on ‘first tried convincing Anthony Fauci of the opposite’? Why so much attention on Fauci and what he might have thought if the authors had had not been rebuked, when that concern is not present in the body of the email?
- Why not instead focus on what the body of the email actually addresses, with for instance: ‘The author who wrote the paper had to be schooled on a call and did not acknowledge it’, or the ‘The authors who wrote the paper totally borrowed their conclusion from others on the call’
It thus seems that the opinion of Fauci matters an awful lot to the whistleblower, more than the authorship issue (which anyway is rather relative, see 27.6), even if, for some reason, the whistleblower would rather not make that too obvious.
27.8 A second clue
A second and essential clue may be in the date (25 Jul 2020) and the very first sentence of the email: “Given your recent mentions of the origin of SARS-CoV-2, …”.
The previous day, Science had published an article by Cohen, ‘Wuhan coronavirus hunter Shi Zhengli speaks out’. That article was a denial of any lab accident by Zhengli, with strong support from Daszak, and was published shortly after the NIH confirmed the suspension of Daszak’ grants [footnote 95]. The article also quotes Holmes and Andersen, as the logical go-to experts on the origin question, alongside Daszak. [footnote 4].
27.9 A third clue
The email to Jon Cohen took a detour via Ian Lipkin and took care to exclude him from the authorship issue.
One author on the paper was not on the conference call: lan Lipkin. It’s not clear how much of the back-story he is aware of. It might be worth giving him a call to ask, in case you feel like investigating. If his co-authors left him in the dark as to what actually happened and he’s worried about the possible fallout he may want to help.
We know that Lipkin is a friend of Baric, which explains why Holmes may have thought that the ‘whistleblower’’ was Baric. But that is a dead-end (see 27.6).
27.10 Making sense of the clues:
From the first and second clues, we are looking at someone who cares more about the opinion of Fauci than a supposed authorship issue.
From the second clue, we can see that the EHA work with the WIV is in the background. But the juxtaposition of Daszak, Holmes and Andersen as the experts cited by Cohen in the article of the previous day is also interesting: Andersen and Holmes presence must have rattled someone’s cage.
The third clue requires to simply ‘follow the money’, especially since we already know that Fauci’s opinion and EHA grants are likely important: Ian Lipkin was one of the key Bellagio GVP forum attendees, and he published a paper with Daszak, back in 2011, pushing for the kind of GVP broad viral sampling in wildlife. [footnote 42]
Last, a basic profiling indicates that we need to look for someone with elastic moral standards, rather brash, a bumptious Black Adder, not afraid of executing a shadowy plan that is both elaborate and silly.
We’ve got someone, and we have a motive: hurting the P.O. authors, whereby the purpose of the denunciation is the final destruction of Andersen, Holmes, Garry, and Rambaut for their audacity in pushing a non-natural origin. Fauci, god bless him, did not fall for it — but these rascal authors had set in train a chain of events that eventually reached the White House and intel. Part of the problem was the NASEM call that Andersen and Daszak attended on 3 Feb, and the subsequent NASEM letter to OSTP that left passaging open (despite Daszak’s best efforts), which Daszak believed had resulted in the suspension of his R01 grant in May and could endanger funding of as the GVP, the mammoth $1.3 bln project that he hoped to benefit from.
28. CONTEXT: A long feud and some funding woes
28.1 The GVP fight between Daszak, Andersen and Holmes
a. First shot by Holmes in Open Biology (Oct 2017)
In October 2017, Edward Holmes took a strong public stance against the GVP, in Open Biology (a journal of the Royal Society).
b. Next shots by Holmes and Andersen in Nature (June 2018)
As if this were not enough, another criticism of the GVP project was published in the Nature, in June 2018, this time authored by Holmes Andersen and Rambaut, three of the four key authors of Proximal Origin.
The publication of that piece in Nature, a top scientific journal, would have hurt Daszak a lot, as it was a thread to both his reputation and to a substantial funding source for the coming 10 years, with a minimum budget of $1.3 bln (many times more than the whole PREDICT 10-year budget).
c. Twitter fight (Sep 2018)
That feud went on, one article, one paper or then one tweet at a time:
‘That is such a ridiculous and arrogant claim that it doesn’t even merit discussion.’
source: Holmes discussing the GVP claim of economic damage preventions with Andersen (Twitter)
28.2 Andersen’s all-out assault on Daszak’s grants (Oct 2019)
At the end of October 2019, a major article appeared in the NY Times, giving the stage to Daszak to denounce the supposed defunding of his (maturing) PREDICT grants. It was written by Donald G. McNeil, the preferred contact of the GVP at the NY Times, as per their lobbying spreadsheet [footnote 47].
Dennis Carroll, the former director of USAID’s emerging threats division who helped design Predict, oversaw it for a decade and retired when it was shut down. The surveillance project is closing because of “the ascension of risk-averse bureaucrats,” he said. [..]
Congress, along with the administrations of George W. Bush and Barack Obama, were “enormously supportive,” said Dr. Carroll, who is now a fellow at Texas A&M’s Bush School of Government and Public Service.
“But things got complicated in the last two years, and by January, Predict was essentially collapsed into hibernation.”
The end of the program “is definitely a loss,” said Peter Daszak, president of the EcoHealth Alliance, a nonprofit global health organization that received funding from the program. [..]
source: ‘Scientists Were Hunting for the Next Ebola. Now the U.S. Has Cut Off Their Funding’, NY Times, Oct 2019, Donald G. McNeil Jr.
Andersen was visibly seriously irritated by the pity tone of that article, and took to Twitter to decry Daszak’s grants and their ‘hyped marketing claims’. whereby the GVP and even DARPA PREEMPT [footnote 13] were accused of perpetuating PREDICT poor approach to mitigating future outbreaks, by not giving the priority to the cheaper and much less intrusive detection of viruses in populations at risk.
The effect on Daszak would have been rather chilling; with reasoned arguments Andersen and Holmes, both well-known scientists, were out there trying to kill his ‘business model’ of sampling in the wild, which was the foundation of most of his grants.
28.3 Spat with Daszak about credits (12 Jan 2020)
On 12 Jan 2020, Andersen had a very critical exchange with Daszak on Twitter.
The spat was caused by a web page and some tweets published by EcoHealth Alliance that showed the just published SARS-CoV-2 sequence alignment and tree, without acknowledging who actually did the work. That work done by Prof. Zhang, a friend of Holmes, who also did the sequencing, with Holmes and Zhang then working together to release the sequence as China was dragging its feet.
Rambaut also pitched in, and he and Andersen argued that not acknowledging people who actually did the work is what hurts open sequence data, and questioning why EcoHealth itself does not practice open data. [credit: Billy] Andersen went further and accused Daszak of using someone else’s work to deceptively raise funds for EcoHealth Alliance.
A few weeks later, Garry, Holmes and Andersen privately pocked fun at the ‘shameless’ behaviour of the ‘Ego Health’ crowd, when comparing its response to the non-publication of the 99% pangolin sequence to what Eco Health did with Bombali. This refers to the discovered the new Bombali Ebola strain under PREDICT. which was announced by EcoHealth Alliance in July 2018.
That specific criticism of EcoHealth work was also mentioned in the Slack conversation on 11 Feb, with Holmes taking a swipe at ‘Dastwat’.
28.4 Daszak’s anger against the Proximal Origin authors for raising the origin issue (Feb 2020)
We have seen in section 15.4, how Baric, Fauci and Auchincloss were likely involved around the 11 Feb in a review of Daszak grant activities with the WIV, especially as to enhanced pathogens.
We have also seen in section 19.4 how on 10 Feb, Garry described the pity letter that Daszak was writing (his ‘Statement of Support’) as a possible ‘pre-emptive strike’ to try to deflect any possibility of the Furin Cleavage site being the result of passaging, a major concern of ‘high ups’ (White House) and ‘intel’ (intelligence services).
In perfect concordance, on 18 Feb 2020, after Linda Saif sent him the version of Proximal Origin just published on virological.org [footnote 11], Daszak complained to her that the Proximal Origin authors were the same group that ‘elevated the potential that this was a lab release all the way to the White House two weeks ago’:
Daszak (18 Feb, 5:27 am UK):
Definitely — already cited it [GD note: the P.O. version uploaded on virological.org on 17 Feb]. It’s in review for Nature. Unfortunately this is the exact same group that elevated the potential that this was a lab release all the way to the White House two weeks ago.”
Daszak would have learnt of the doubts of the Andersen, Garry and Holmes through Andersen, who was also on the NASEM panel of expert answering the OSTP question about the origin and likely evolution of the virus, and likely through the grapevine.
What is jumping out in that quote by Daszak above, is that Daszak is very annoyed at the question of the origins having reached the White House, and attributes it in large part as the result of the concerns expressed by these three scientists, which would then have triggered the OSTP question.
To make it worse, as we have seen, Daszak was convinced that the Farrar call had triggered a top-level NIH review of Daszak’s grant, working against his own support one level down at the NIAID (where Morens was his strongest asset and his intermediary to Fauci). The conjunction of both the White House getting alerted to it, and the NIH having to move into damage limitation mode, with Collins seemingly detaching himself from his NIAID Director (Fauci), effectively put his grants in a very compromised situation.
The full email of Daszak to Linda Saif also shows that Daszak knew that the manuscript was in review with Nature, only one day after it had been sent there. He would have learnt about the peer-review in Nature via Farrar who exactly at the same time was organising for the publication of Daszak’s Statement in Lancet and signed it.
Since the full Farrar call group had not been told of that Nature submission at the time, it is very likely that Koopmans next learnt about it through Daszak (they had worked together only a week earlier at the WHO Blueprint in Geneva [footnote 40]). From there, her colleague Fouchier would have been aware too, and not too pleased to see that Farrar was executing his plan against their repeated opposition.
28.5 Daszak’s desperate efforts to save his NIH/NIAID R01 grant
a. Punched in April
On April 24, 2020, after the publication of Proximal Origin in Nature Medicine on 17 March 2020, the NIH suspended Daszak’s grant R01 AI110964, through an email to him.
Erbelding was in CC. That email was soon forwarded to Auchincloss (Fauci’s Principal Deputy Director, see 4.1) and others.
b. Support from Fauci
In the evening of 25 April, Daszak and David Morens (Senior Advisor to Fauci, the Director of NIAID) have a phone conversation. The next day, using his private Gmail address [footnote 92], Morens contacted Daszak, cc. Gerald Keusch (Jerry), assuring him that:
There are things I can’t say except Tony is aware and I have learnt that there are ongoing effort with NIH to steer you through this with minimal damage to you, Peter, and colleagues, and to nih and niaid.
[..]
Indeed, I expect we would do everything possible to protect you. I have reason to believe that there are already efforts going on to protect you.
[..]
As a coda, Tony knows about your work. Sometimes around 1 Feb I sent him an email outlining your work over recent year, and why it was so important to science.
Daszak was infuriated by the suspension, blaming it on some White House interference with the NIH (not the NIAID), which supposedly was happily abetting Trump’s conspiracy theories. In the next few days, the press would be relaying Daszak’s accusations, painting the decision to suspend his grant as a pure political interference with essential science, fanned by conspiracy theories.
The Trump administration abruptly cut off funding for a project studying how coronaviruses spread from bats to people after reports linked the work to a lab in Wuhan, China, at the center of conspiracy theories about the COVID-19 pandemic’s origins.
source: Politico, April 27, 2020
c. Trying to get back on its feet
On 29 May 2020, as an important EcoHealth Alliance paper detailing some samples collected up to 2016 in South West China (Latinne et al.) was being published, Daszak forwarded the paper draft to Morens (Senior Advisor to Fauci at NIAID), with some talking point for journalists that he has produced himself.
In one of these talking points, he described the termination of his grant as ‘being “‘for convenience’ following political interference from the White House’.
d. Knocked out in July
In her interview with Jon Cohen (24 Jul 2020), Shi Zhengli discussed the work she had done with Peter Daszak and ended up criticising the termination of his NIH/NIAID grants in April 2020 (with again Erbelding being involved in the process), confirmed for all purposes on 8 July 2020 by setting strict transparency conditions on its resumption.
Peter quickly tweeted about the Shi Zhengli interview, and described the decision to suspend his grant as totally political.
e. Financing Crunch Time
For context, as PREDICT was maturing in mid-2019, EcoHealth alliance was facing a funding crunch; the DEFUSE proposal had been turned down by end 2018, while the Global Virome Project (a mammoth $1.3 bln over 10-year project) was not taking off, leading Daszak and the other GVP directors such as Dennis Carroll to even try — against better judgement — to get China to take the lead of it if the US was not going to sign the cheques.
In short, Peter already needed all his existing NIH grants to keep flowing, and further he needed to keep the GVP chances alive [footnote 88], when the April 2020 suspension of his R01 grant added to his woes. The quiet ‘bridge-financing’ that he immediately received via private foundations was welcome, but it was just a stop-gap measure; he still needed proper government grants with a seal of approval [footnote 89].
Then in early July 2020, Daszak faced a Pyrrhic victory, whereby his NIH grant could be restarted but only under conditions that he could not possibly meet. That would have rattled him — especially when his main GVP scientific opponents — Andersen and Holmes — were by now, like two accidental heroes, basking in the glow of Proximal Origin.
At precisely the same time, Daszak’s hopes for the GVP would have been somewhat raised when Dennis Carroll (the main proponent of the GVP with Daszak) joined the Biden Campaign as an advisor [footnote 90]. All that Daszak needed was to see Andersen and Holmes off.
Essential Insight: Sour Grapes and True Heroes
Andersen and Holmes had started with a frontal attack on the GVP in 2017, and had followed by an all-out assault on Daszak’s grants back in Oct 2019, this time aiming not only at the GVP, but also at his USAID grants (PREDICT) and even at DARPA PREEMPT [footnote 13].
As if this was not enough, Daszak also blamed Andersen and Holmes for privately ‘spreading rumours’ about the origin of the virus at the highest level in the US administration, with that fateful NASEM call in the presence of intel and the subsequent open-ended letter to OSTP. That, despite his best efforts at containing it, started a prairie fire in the US administration that ended up engulfing his R01 grant.
Just after his NIH grant decisive knock-out in July 2020, the contrast between his own fate and the resounding success of the Proximal Origin piece, and thus of its authors, would have been particularly difficult for Daszak to swallow. First, it would also have raised the possibility of further high-profile attacks from Andersen and Holmes, now in a position of major strength, on the GVP and other sample collection grants. Secondly, there was an insufferable irony in the fact that the very people who were now publicly considered scientific heroes for supposedly ‘debunking’ some non-natural origin in Nature Medicine, had actually achieved just the opposite by starting that destructive prairie fire in the US administration — where it really mattered.
The letter to Cohen was meant to address both of these aspects: the danger to the GVP and other grants represented by Andersen and Holmes now in a position of force, and the tragic irony of them getting all the ‘debunking’ rewards for having practically achieved the opposite. Thus, it was not just about ego, it was also about survival of Daszak’s ‘business’ model, both of which had long outgrown the confines of an ordinary Track II game.
As a side product of its reputation demolition aims, the letter to Cohen can also be considered to be a thank-you note to Fouchier, pictured there as the true expert during the Farrar call. Fouchier, with his ‘once every 2.5 times the age of the universe’ claim about biosafety [footnote 46], and his likely role in getting PO rejected at Nature and the passaging option being eventually closed down, had secured a place in Daszak’s pantheon.
Koopmans (who works Fouchier at Erasmus) would again prove to be a great friend when she pushed back with Daszak against Dr. Tedros, in a two-hour call, at the end of the 2021 WHO mission. In that ‘ugly’ call, they both argued that their ‘poll’ of opinions as to a possible lab-origin should be the official conclusion of the 2021 WHO mission, and should therefore evacuate a lab origin once for all. Even though half of the scientists in that poll were Chinese, with all the risks that dissidence would entail, and despite all the background work of the Ego-krewe of Daszak, Fouchier, and Koopmans to impose an exclusive zoonotic narrative going back right to February 2020.
Reminder:
As a general principle, any amount of circumstantial evidence is not a formal proof. So, according to that principle, an accumulation of possible motives, a long existing feud that was only getting worse, the trading of insults, an apparently suitable temperament and modus operandi [footnote 91], as impressive as they may all seem, should not be interpreted as a statement that Peter Daszak (the subject of the methodological example above) is the author of the anonymous email to Jon Cohen.
To properly assess the validity of that general principle, the reader is invited to think of any other person, with similar fitting characteristics (or anywhere to half of them), who could have been substituted to Peter Daszak.
footnotes:
[❖] When the FOIs emails are a clear thread made of consecutive replies, the best is to read them backward, starting from the last one, while noting that the timezone of each email is often the one of the person replying directly to the email, not of the person who wrote the email.
Indeed, each reply may freeze the time of the replied-to email in the time-zone of the person replying directly. This is due to the fact that, when an email is replied to, a simple email client would typically take the ‘sent’ time of the replied-to email in the time-zone of the person replying, and dump it as text in the new email. From there, that text becomes immutable and just get included with each further reply-to as any other text.
When the last email in the thread is not one that somebody replied to, which can often be seen from the pagination, the code in the bottom-right corner of the page typically gives the owner of the mailbox, which then gives the time zone of that email (ie: LIPXXXX refers to Lipkin’s email box). However, some threads seem to be partially reproduced, making it more difficult to understand the timezone of the apparent last FOI’d email in the thread. Of note, REVXXXX, whatever that one is, is UK -8.
Also, pay attention to these YYXXXX codes in the right corner. Some Congressional documents are a mash-up of emails extracted from different mailboxes (different YYs), and reordered in an approximate chronological order. Understanding these users transitions help identify the continuous threads and the timezone of the last email in each thread.
[1] When Zhang discovered a coronavirus in that patient sample on 5 Jan 2020, Zhang and Holmes may have believed that they were the first one to positively identify the new coronavirus. In fact, this was far from being the case. The very same Wuhan Central Hospital had sent a sample of another patient to another NGS company, Vision Medicals in Zhengzhen, on 24 December 2019, not as part of that ’Survey of major epidemic virus sources’ program, but at the initiative of the frontline doctors who were trying to diagnose another case of pneumonia of unknown origin.
By the 27 Dec Vision Medicals had a good quality near full sequence and had already done a fairly good analysis of the SARS-CoV-2 virus. On that day, it shared it conclusion with the local CDC, and also shared the sequence with the Microbiology Institute of the Academy of Medical Sciences (Beijing) and with Tongji hospital (where the patient had been transferred by then). On 27 Dec, Tongji asked Jin Yin-tan hospital to take the patient, since they were the designated pulmonary infection hospital in Wuhan, with a negative pressure ward. In the evening of 27 Dec, Jin Yin-tan next asked Vision Medicals to share the sequence with them so that they could review it before accepting the patient. Jin Yin-tan forwarded the sequence to the WIV, and had a call at 10 pm that day (27 Feb) with the WIV, where the WIV repeated the analysis of Vision Medicals and came up with similar conclusions.
Hence, by the 27 Dec evening the virus had been identified and analysed, the results had been shared with key scientific institutions, and the identification and analysis even repeated by the WIV. But nothing was made public as the subject much way too sensitive.
[2] One may remember for instance, in the context of an already weary public opinion due to various US foreign policy failures (Vietnam War, Bay of Pigs), some House and senate committees and some ad hoc commission went after CIA director William Colby in 1975 (the ‘Year of Intelligence’), with accusation of spying on US citizens on US soil and of attempts on the lives of various foreign leaders. These committees often competed for the latest juicy revelations, in what looked very much like a media circus for unrelated political gains, with some exaggerations and misconstructions thrown in.
However untidy the process was, it still was very much part and parcel of a necessary right of oversight by elected representatives, something that Colby understood and respected, likely saving the CIA in the process.
[3] An e-mail by Holmes, written about 7 months later, stated that the Farrar call agenda included a discussion about the opportunity of elaborating on Andersen’s notes as paper. This is in contradiction of the actual agenda of that call and related emails. For sure, the point had been mentioned by Fauci to Andersen before the Farrar call, but there is no element to attest that it was discussed during the call. That item is instead on the agenda of a call on 8 Feb.
[4] Zhengli’s interview with Jon Cohen contains what may now seem a slightly surprising mention that the market may just be a superspreader event.
This exact topic was also the subject of a Twitter interaction between Marion Koopmans and Peter Daszak, dated 15 June 2020, so a few weeks before Zhengli’s interview.
[5] Missing the necessary distinction between lab-construct, lab-product and lab-escape has resulted in many journalists and scientists producing statements about the possible origins, that one can easily identify as logically false, without having to enter into a complex scientific debate. It is also a perfect way to build some straw-men.
[6] One may add that any risk of misinterpretation of the causal role of the virus (a major issue with SARS-1) could have been a concern that explained the Chinese reluctance to release the sequence. That is, in fact, their official version (while pretending that they did not know of it already on 27 December). Still, that Chinese version does not make much sense: first, that risk — if any — was alleviated by the presence of Holmes as co-author and of by the reviewers, then the manuscript sent for review would have been very careful in drawing any causality between the newly detected virus and the outbreak in Wuhan (the final paper itself only briefly mentions a likely relation between the two, by drawing on a Chinese paper published in early February).
For the context, the initial misattributing of the SARS-1 outbreak to chlamydia, with SARS-1 playing at most a secondary role, was a big loss of credibility for China and resulted at first in a totally inadequate official treatment based on antibiotics. One standard in the precise attribution of an outbreak to a new pathogen is to verify Koch’s postulates, and China has emphatically claimed that such work had to be done before it could release the SASR-CoV-2 sequence, thus causing some delays. Still, Chinese authorities admitted that the pathogen was likely as SARS-like virus on 9 Jan 2020. Also, some experts have argued that in such a dangerous outbreak, the completion of all the Koch criteria may cause very unreasonable delays, and is thus not fit for purpose.
[7] The description of these events in Spike is problematic:
- The book explains that Maria Van Kerkhove contacted him on 10 Jan (UK time), then points to his tweet posted as a result on the afternoon of the same day, which supposedly triggered Holmes to contact him and then started the conversation between the two.
- In the same pages, Farrar dates the beginning of that conversation with Holmes on Thursday 9 Jan (evening, UK Time), not 10 Jan. This is clearly inconsistent with Holmes contacting Farrar after seeing his tweet on 10 Jan (UK time).
- The book dates the publication of the virus as of 11 Jan, around (1 am UK time), which is correct. This would only be a few hours of ultimatum if that ultimatum had been issued on 10 Jan at 9 pm, after first Van Kerkhove contacted Farrar and then Holmes contacted Farrar, and not 24 hours at least.
So, most likely Van Kerkhove and Farrar got in touch on 10 Jan, after Holmes and Farrar somehow had got in touch on 9 Feb, and Farrar’s tweet on 10 Jan was not the trigger for Holmes to contact him, despite what the book says. Which, incidentally, means that we still do not know what that trigger was.
[8] The New England Journal Of Medicine paper is most likely ‘A Novel Coronavirus from Patients with Pneumonia in China, 2019’, DOI: 10.1056/NEJMoa2001017. That paper reported the identification, isolation and sequencing of a new coronavirus, which it called 2019-nCoV, by the China CDC. It was published online on 24 Jan 2020.
[9] This is very much in line with the way Van Kerkhove had called Farrar on 10 Jan 2020 to tell him of the NEJM paper for review (as per footnote 8). The two are careful to share quickly all relevant information.
[10] The NIH headquarters are based in Bethesda, Maryland.
[11] This exchange with Linda Saif happened in the discussion for another key denial paper, fraught with peer review fraud, undeclared contributors (including Shi Zheng-Li and Baric) and the authors mentioning the real possibility of a lab escape and of a unnatural FCS.
[12] During his interview with the House, Andersen explained that straight from the Farrar’s call, Holmes was already focussed on a paper, while he was more keen on doing some analyses, discussing their findings, and seeing where this would lead.
Andersen further explained that he was reluctant to go for publication, not only with the first versions but down to the very time the manuscript was sent to Nature for peer-review. His objection was that the manuscript was just not ready.
Andersen also explained that Holmes is the one that put pressure on sending the Manuscript to Nature. Unfortunately, on that latest point, Andersen’s version of the events does not tell the full story by trying too hard to exculpate Farrar:
➤ Farrar is the one that took the decision to go for publication on 7 Feb, after talking to Holmes about the supposedly 99% homology pangolin sequences. Farrar repeated his decision and made it clear to all that it was final (9–11 Feb), in the process overruling Andersen who had just made the pint that more data was needed (9 Feb, see 13.11). Farrar then pushed for an early release on virological.org on 17 Feb, just at the same time as he was working on the publication of Daszak’s Statement in Lancet.
➤ Even more to the point, Farrar had Holmes send the manuscript to Nature on 17 Feb so that they could get an answer back in the morning of that day as to whether Nature would be fine with a release on virological.org. In fact, Andersen was not back to the connected world when Holmes sent the manuscript to Nature (he was still on his way back from a desert trip). So there was not even a conversation with Holmes at the time, and there would not have been one for the previous three days when Andersen was unreachable.
One may further wonder why Holmes sent the manuscript to Clare Thomas (based in London) at around 2:30 am UK time, when he could at least have waited until 8 am UK time, which would have given time for Andersen to connect back with the world while being the evening in Sydney. As it is, Clare Thomas replied to Holmes at the very start of her day (8:07 am UK), which was a comfortable 7:07 pm Sydney.
One may thus wonder if the idea was not to force the sending of the manuscript without risking some last-minute delays from Andersen, since Holmes and consequently Farrar knew of Andersen’s preference for taking more time. As it is, Andersen did not even have a chance to quickly review the manuscript that Holmes had managed in his absence over 3 days.
[13] In the context of that specific Twitter discussion about EHA PREDICT grant, the mention of PREEMPT does not mean at all that Andersen knew about DEFUSE in Oct/Nov 2019. To start with, a few months later, the Slack conversation never mentions DEFUSE, despite being consistently very critical of Daszak; so at most Andersen may have known that EHA tried to apply for a DARPA PREEMPT grant back in 2018, without the details.
The Twitter conversation then simply illustrates the opposition between Andersen and Holmes’s preferred approach (surveillance of sequencing at the human-animal interface), and Daszak’s approach (sequencing in the lab followed by laboratory work that may include viruses enhancements), which they consider particularly wasteful and risky.
[14] The division of roles between MI5 (UK domestic-focussed intel service) and its MI6 (focussed on gathering intelligence out of the country) can confusing at times, but is actually not that complex to understand, being essentially pragmatic.
First, in the case of the COVID-19 outbreak, both the UK and the US would look at the threat as a domestic one, necessitating a coordination of domestic efforts. It is perfectly in the remit of MI5 and FBI to investigate such domestic threat, possibly coordinating their effort with other nations’ services (through the Five Eyes for instance).
Secondly, MI5 has — by necessity — developed very strong links with the Five Eyes since 9/11 (including directly with the CIA), as the Al-Qaeda threat was international in scope but involving a large amount of domestic investigations and surveillance. In fact, MI5 worked on both ends of the problem: it integrated itself strongly in the Five Eyes while at the same time integrating itself closely with the Metropolitan Police in the UK.
Eliza Manningham-Buller played a key role in that respect. A veteran of MI5 with 26 years of career there, she was deputy director-general of MI5 when, on 12 Sep 2001, one day after the terrorist attacks, she flew on an RAF plane to the US to visit the CIA. On the Vickers VC10 were also Richard Dearlove, Chief of MI6, and Francis Richards, head of GCHQ. From that time on, until she retired and moved to the Wellcome Trust, she has been a well-respected figure within the Five Eyes.
[15] The section of Redfield testimony with that quote is included below. Redfield said something similar but a bit less precise in an interview with Paul D. Thacker (https://disinformationchronicle.substack.com/p/former-cdc-director-robert-redfield), adding that he contacted Fauci in mid-January. Fauci, in a videotaped deposition, could not remember Redfield contacting him in January to discuss a possible non-natural origin.
[16] In 2015, during the US GoF moratorium (2014–2017), Fouchier published a paper where he estimated the probability of a Laboratory Acquired Infection (LAI) in his own BSL3 to be incredibly low, at 0.00001% per person, per year (one accident over 1,000,000 years considering 10 people working in his lab).
Multiplying this by his own estimate of onward transmission, following an infection of someone in a lab, Fouchier concluded with full seriousness that an ‘onward transmission’ caused by his lab (with at least one human-to-human transmission) would happen less frequently than once every 33 billion years, which is about 7 times the age of the earth and 2.5 times the age of the universe.
That rather optimistic estimate, obviously meant as an argument against the GoF moratorium, was promptly taken down by Lynn Klotz, who criticised the calculation and compared it to the actual 0.2% per BSL3 lab per year (1 LAI in 500 years) recorded by the US CDC, based on declared accidents.
There is no other research field, with such potential public impact, where Fouchier’s attitude would be tolerated. When Boeing cut corners and silenced whistleblowers, resulting in a few hundreds unfortunate deaths, the company was nearly brought to its knees, its stock collapsed, its CEO left, all largely due to the pressure from the competitive market it operates in. A pressure, partly formalised by regulators, which has constantly worked to improve air-travel safety, and never stopped it, quite the opposite.
[17] That distinction between a direct insert of the FCS and its production by passaging is also the distinction between a ‘lab-construct’ and a ‘lab-product’. Loose vocabulary unfortunately often confuses the two, when it is not deliberately used for straw-manning arguments against a research-related accident.
[18] Of secondary interest, the members of the Covid commission who contacted LeDuc were Gerald Keusch (‘Jerry’), Danielle Anderson (who has worked at the WIV) and Su Yadana from EcoHealth Alliance.
Following the dissolution of the Lancet Commission Task force by Jeffrey Sachs in late 2021, due to Daszak’s non-declared conflicts of interest, the former members of that task force reinvented themselves as an ‘Independent Task Force on COVID-19 and Other Pandemic Origins’.
Eventually, Keusch, Anderson, Daszak, Koopmans, Supaporn, Yadana, and Linda Saif, would publish in October 2022 a report in the Proceeding of the National Academies of Sciences (PNAS) representing their findings under that new incarnation, stating that the ‘balance of available evidence strongly supports a so-called natural origin of SARS-CoV-2’.
[19] The official summary of that Oct 2019 meeting makes for a very interesting read. During a presentation entitled ‘Advances in Understanding and Altering Pathogen Properties’ (image below), Baric summarised the state of the art in synthetic biology and the creation of enhanced virulence viruses: starting via a now rather cheap full synthesis from good quality sequences, adding some predictive modelling for a few key features, which because of its limitation typically needs to be supplemented by passaging in adapted mice. He then reminded that this semi-predictive and semi-passaging approach was subject to ‘the law of unintended consequences and the possibility of surprise’.
[Dr. Ralph Baric] noted the rapid pace of advance and decreasing cost of nucleic acid synthesis; the first coronavirus to be synthesized cost roughly $42,000, a price that would now be $6,000. The largest genome currently synthesized is a 520kb mycobacterium, indicating that it is now possible to synthesize the genomes of most RNA and DNA viruses. In addition, high fidelity sequences are available for many viruses, rendering it possible to synthesize viral genomes and recover viable virus for many strains.
Studies to alter pathogen properties of viruses can use several approaches, including selection pressure to drive evolution toward a phenotype as well as deliberate design. Potential opportunities might include building chimeric viruses with altered structures for the receptor for viral entry, or those that incorporate changes to other virulence determinants or that modulate host-pathogen interactions. [..] But he cautioned that a combination of techniques, including selection pressure from passaging in mice, is generally needed to generate virulent strains in a host and reminded the audience of the law of unintended consequences and the possibility of surprise.
Dr. Baric also noted that predictive modeling of protein interfaces and the use of such models for structure-guided virus design has recently improved, providing advancing capabilities for affecting host-virus interactions and altering antigenic properties. [..] He also noted the barrier provided by a glycosylation site that makes it relatively easy to create mouse-adapted SARS strains but not mouse-adapted MERS.
[20] The correct way to track the channel name is to look at the change notifications. The proof of the original channel name is thus on page 17 of Proximal_Origin_Slack.pdf:
[21] The facts that politics were catching up with the Track II initiative in 2019, with the clearly all-encompassing Chinese authorities putting obstacles to visits and meetings by NASEM members, should have led NASEM to rethink the validity of that Track II approach. After all, the all ideas of Track II was precisely to try to create a space devoid of such interference.
On the plus side, it may be said that these obstacles were still less vexatious than the treatment that was dished out to the US consulate in Wuhan in 2017, following Trump’s election. During that year, the consulate was subject to a menial but constant barrage of interferences with its normal operations. That the Wuhan consulate was selected for such treatment, out of all US consulates in China, was likely no coincidence.
[22] An alternative was to synthesise components of the virus based on published sequence. This is what UTMB eventually tried to do, by ordering the required components from a Chinese supplier. [source: UTMB-LeDuc-batch-1.pdf, p.4,206 and 4,678]
[23] If need be, the actual opinion of LeDuc as to the possibility of a research-related accident is nicely encapsulated in an exchange between him and Franz in May 2021, discussing the Science letter of Bloom, Baric, Relman, and others, asking for a proper investigation of the possible research-related origins of the virus:
[24] For a more in-depth analysis of the Track II effort towards China, and the competition in that space, especially between CISAC and Inglesby’s JHU, see the ‘CISAC Bio Conference Call Notes’ of Aug 2019 and this medium article.
[25] The backbone point is important, as many scientists have tried, since then, to make the point that only a known backbone would have been used in research. Here, Andersen judged that scenario as ‘most basic’. Others have also correctly remarked that if the objective was to be ahead of the evolutionary front, by trying to push a non-virulent strain in into a human pandemic candidate (as per DEFUSE), there would have been little reason picking up either SARS-1 or MERS which have already evolved into that domain (see Garry’s notes for instance in 5.1b), or some less virulent relatives which have already been studied and experimented on to try to get them there. The ideal backbone may instead be a not so well-studied relative, and potentially one more recent that could thus be the object of some first publication too.
Andersen’s backbone treatment that day is also worth contrasting with his contribution to the NASEM letter discussions (see 8.3.f.3) and the final treatment of the backbone arguments in Proximal Origin (see 25.2.b).
[26] The whole a-priori / a-posteriori argument falls apart when one considers the favourite approach of Baric, which he had explained to Chinese scientists at the Harbin conference of Jan 2019, and which was also included in the DEFUSE proposal he worked on with Daszak: targeted insertion of the FCS via editing, some more possible point mutations, then passaging to polish up the result and make it hopefully sharper and stable.
[27] The actual position of Lipkin on the origin debate is well exposed in the transcript of his House hearing.
On Proximal Origin, he laboured to explain that the paper, if read carefully, still left the door open to non-engineered research accident scenarios, based on his understanding of both zoonosis and research risks:
‘There are some nuances in the way the paper was, you know, finally assembled and printed that are, I think, less than ideal. But if you look at the beginning of the paper where we describe the constraints with which we had to work and trying to understand what was going on, and if you look at what we said there at the very end where we say, you know, it’s possible that if we get accumulated additional information, our views would change, the same three potential models remain viable [1: adaptation to humans, 2: selection in animal host, 3: selection during passage] — the same models — not three models, but the latter two models still remain viable. So we excluded the possibility of deliberate design for the reasons I’ve just said. [..] Given the number of animals that were present in the market, this would be a higher probability event. And the possibility that somebody — that it might have occurred as a research-related accident was open and could not be entirely excluded.’
He also explained that he later tried to “steer down the middle”, when he signed the Bloom/Chan Lancet letter calling for a proper investigation of the origins. So, when Holmes invited him to sign Worobey et al. (2022), one of the two Science market papers, he simply refused, considering that:
“I don’t think that I can conclude based on this paper that the origin is the seafood market. The only thing that I conclude from that paper is that it was a site for amplification. That doesn’t mean that it wasn’t the site. It just means that I don’t find it, you know, what we call dispositive evidence.”
and:
“Our colleagues fuelled this with armchair epidemiology based on unverifiable data sets and terms like ‘dispositive evidence.’”
After his refusal to sign Worobey et al., Holmes did not invite him again to join on a paper.
In the end, Lipkin, just like his friend Baric, comes out quite well in that tragicomedy, and for the same reasons. It’s not often that someone is frank in a senate or House testimony, in Lipkin’s case to the point of effectively exposing his/her occasional work for the CIA, which would likely seriously limit such future opportunities. During his testimony, he even mentioned that he thought that the damage his frankness on that subject had caused was likely irremediable.
[28] Strictly speaking, this ignored other forms of research accident, such as an infection with a natural virus during research activities (in a lab or in the wild). But these formed too subtle a distinction for the vast majority of people, including a large majority of experts, so their omission would not change in any way the overall perception.
[29] During his Jan 2019 Harbin trip, Baric had clearly explained to his Chinese hosts that a combination of deliberate design and passaging was the best way to succeed in creating host-adapted virulent strains (albeit with some possibility of surprise). So having a category for pure deliberate engineering in the draft is a rather artificial construct, and the arguments for and against are more complex than the one listed. See also footnotes 19 and 26.
[30] Interestingly, passaging would typically require a fairly close virus, clearly unknown, to start from. But the absence of a known backbone is listed as an argument against deliberate engineering (gene editing), and not passaging, which does not seem very logical. Anyway, at least other arguments than the backbone are listed against (pure) deliberate engineering, whether good or not.
[31] One might have been able to build a partial counterargument to this section, in line with some earlier remarks of the co-authors: the finding of a virus with a 99% homology with SARS-CoV-2 without an FCS should not have affected the weighing of the various hypotheses in the first place, since it also provided the ‘missing’ backbone for Passaging and Deliberate Engineering.
So the right question should instead be: Why did Holmes and Farrar consider it as an argument to go for publication anyway? But we already know the answer to that question, and it’s largely the same as the one provided here (see Insight under 19.8).
[32] ‘Superficially is the keyword here:
- For intentional design, if one could hope to close the rumours of RaTG13 having been used to design SARS-CoV-2, as part of some malevolent program (since it would make little sense to use when a closer relative existed), unfortunately anyone could then claim that SARS-CoV-2 was intentionally built using that 99% relative.
- For passaging, it does not matter if natural evolution looks more likely when there is a claimed 99% relative. What matters is the behaviour of the relative probabilities of natural to non-natural. Unfortunately, the probability of passaging may just be as much affected by finding a closer relative, being just an accelerated form of ‘natural’ evolution, which also needs to start from something (the closer, the better).
[33] On 10 Feb, Holmes would suggest that the issue came from them having mentioned that pangolins were illegally trafficked into China, something that was generally not to be reported in international media as China could lose some face as a result of it. And especially not in these outbreak circumstances.
[34] Holmes would keep pushing first the pangolin hypothesis, putting his name on a flurry of pangolin papers in 2020. When that did not catch up (no pangolin was sold at the market, as per Xiao et al., and, as Daszak himself explained, they are ‘extremely rare in markets’), Holmes switched to the raccoon dog as intermediate host. He did so largely based on a photo of a lonely raccoon dog in a cage, that he took there in 2018.
With no more scientific element to substantiate that case, and with the photogenic appeal of the poor animal eventually exhausted, the identity of the potential intermediate host has since reverted to an open question, as part of a very contested market zoonosis hypothesis.
[35] A MCMC model is a Markov Chain Monte Carlo simulation. The convergence of the MCMC means that the chain has reached a stationary distribution that properly approximates the true posterior distribution of the corresponding model. One application is the production of estimates for missing data. Here Andersen is effectively saying that, based on the insufficient available data, his conclusion keeps bouncing back between a zoonosis and a research-related accident. He basically has problem converging.
[36] At some point in his quest for the best superficial solution to the problem of being the instigator and author of the Statement of Support without disclosing it, Daszak hit with the idea of pretending that Linda Saif, Jim Hughes, William Jaresh, and Hume Field were the authors of the statement, and himself just a helpful signatory. The side by side comparison with an email sent 4 days earlier, where he enjoined the very same Linda Saif, Jim Hughes and Hume Field to support and sign the Statement he had drafted and was sending them, is quite comical, but par for the course for that character:
[37] That very same point was later repeated by Baric in an incendiary take-down of Daszak’s defence of the WIV biosafety practices:
As for the ‘state determined rules’ that Baric mentioned above, his House transcript shows that he got a confirmation from Shi Zhengli:
Their regulations state pretty clearly that they don’t consider culturing bat viruses at BSL-2 as a biosafety concern. I also had that verbally confirmed by Zhengli Shi at a meeting in Harbin [Jan 2019], when I was telling her she should move it all to BSL-3, and the reasons why. So I know that. And she also in that meeting said that all animal work is done at BSL-3.
source: Baric-TI-Transcript.pdf, p.93
I have also previously shown that the guidelines published by the WIV, once available on their website and now deleted (but translated and saved here), had BSL-2 as required level for work on rodent and bat coronaviruses.
[38] Bruce Aylward (Canadian), a senior adviser to Dr. Tedros, was the Team Lead of the WHO-China Joint Mission on COVID-19 (16–24 Feb 2020) for the international side. The Chinese side was led by Wannian Liang.
FOI’d emails have shown that Aylward was willing to heap praise on China’s actions in Wuhan, in the official report of that mission, to the point of being told to dial it back a bit, by two members of the mission, Chikwe Ihekweazu (Nigerian CDC) and Tim Heckmanns (RKI, Germany). These emails show that he was also very concerned about anything that could upset China, such as using SARS-CoV-2 for the virus name.
A month later, on 27 Mar 2020, Aylward made the news again when he pretended not to hear the question of an HK journalist who was interviewing him, then hanged up on her, before dismissing her question once more when she dialed him back. The question she dared ask was if the WHO would reconsider Taiwan’s membership.
China must have been very pleased.
[39] One direct implication of this is that the WHO would have been fully aware of the hidden role of Daszak as promoter of the Statement of Support, when Daszak was selected by China for the WHO mission to Wuhan in Jan-Feb 2021.
[40] About 350 scientists convened (online or in person) to that WHO Blueprint meeting in Geneva on 11–12 Feb 2020. Farrar (co-chair of the meeting), Rambaut, Daszak, Drosten, Koopmans and Embarek were there in person. The last three were also all part of the animal/environmental origin breakout group (in which Daszak played a leading role), and would be selected by China later that year, to be part of the WHO mission to China in Jan 2021.
During that 2021 WHO mission, Daszak would come back as leader of the Animal / Environment Working Group, Koopmans as leader of the Molecular group, and Embarek as Mission lead (Western side). This was a complete throw-back to the 11–12 Feb 2020 Blueprint setup, which must have made China rather comfortable with these names.
[41] The absurd non-reporting of cases during the first two weeks of January 2020 is well illustrated by the official Chinese CDC numbers below, used at the WHO meeting in Geneva on 11–12 Feb 2020, during which hundreds of scientists convened:
[42] The restriction site mentioned is the BamHI restriction site at the end of the spike sequence. Fouchier had conveniently dismissed it away as a pure chance mutation in his notes to Farrar on 2 Feb (see section 5.2).
[43] The role that P.O. ended up playing in actually setting back the origin debate, and hurting even Chinese scientists, is perfectly illustrated by the story of Xiao et al.
In June 2021, a key paper was published in Nature: Xiao et al. ‘Animal sales from Wuhan wet markets immediately prior to the COVID-19 pandemic’.
That paper offered an essential description of the animal found in the market stalls at the Huanan Seafood market and two more markets in Wuhan, over 30 months, and sold there legally or not. That study was based on monthly open discussions between the stall-holders and a couple of Chinese researchers that they had learnt to trust (Xiao Xiao and Zhao-Min Zhou).
This story was first reported by Chan and Ridley in their book, ‘Viral’.
A recent FOI of David Morens’ emails (Fauci’s Senior Advisor) fully confirms it, via the draft of a Bloomberg article that its author Jason Gale, a Bloomberg journalist, sent to Morens on 5 Aug 2021. The article was eventually published on 17 Aug, but without some of the important details that can be found in the draft and in ‘Viral’.
Here are the key elements:
- The draft of Xiao et al. was sent to a journal for publication right up in February 2020 (before the State Council gag order of 5 March 2020). The authors (who also included three Western scientists, two from Oxford University) expected a swift review and publication, given the extreme importance of the data that their manuscript included.
- Instead, the peer-review process did cast doubts about the quality of Dr. Xiao’s surveying and the resulting representativity of his dataset.
- The key issue was quite simply that Xiao et al. very clearly stated that no pangolin was sold at the market. This, right in the middle of the pangomania which was driving most of the arguments behind the first version of P.O. published on virological.org on 17 Feb 2020, and then the Nature Medicine piece on 17 Mar, was simply not welcome news. [footnote 34]
- After submitting a second version and some more delays, the paper was eventually rejected outright in September 2020. “They did not think that it would have widespread appeal”, as Chris Newman (Oxford Uni.) told Jason Gale.
- This was a blow to Dr. Xiao and Zhao-Min Zhou, who felt that scientists were not ready to take their data seriously.
- Still trying hard, the authors revised their manuscript a third time, to include data on China’s pangolin trade networks, even if no pangolin were ever found by Dr. Xiao at any of these 3 markets over 30 months.
- After all that bending backward, the manuscript was revised to develop a very different angle, and was renamed ‘Pangolin trading in China: Wuhan’s alibi in the origin of Covid-19'. In other words, the absence of pangolins being found by Dr. Xiao over 30 months in three markets in Wuhan (including the seafood market) was openly questioned by being described as an alibi.
- The revised manuscript was sent to ‘Scientific Reports’, which did not publish it immediately.
- Eventually, the authors were able to revise the manuscript again, trim the ‘alibi’ angle, drop the pangolin trade element, and go back to their straight descriptions of the sales in these three markets. In that version, the piece was eventually published by Scientific Reports on 7 June 2021, a full 13 months after being first sent to a journal for publication.
The publication of Xiao et al. was still not the end of some rather dubious preprints and of some bad editorial decisions associated to that market data. Indeed, the data of Dr. Xiao, if it did not include any pangolin, included sales of (often wild caught) raccoon dogs. So, soon after Xiao et al. was published, the P.O. authors, especially Garry and Holmes, did fully pivot and published papers asserting that some raccoon dogs, not some pangolins, were the likely intermediate hosts. In doing so, they drew back on Xiao et al to make their point.
Non-content with that léger de main, some of the P.O. authors would later blame Chinese scientists for supposedly ‘hiding’ the sales of raccoon dogs at the market, when P.O. itself was the primary reason for that data to be rejected by Western journals. (For reference, the sale of raccoon dog at the Huanan seafood market was also noted in the Terms of Reference of the WHO mission negotiated in August 2020 and published in November 2020, but went unnoticed during the pangomania.)
In a repeat of that conflicted and extremely patronising attitude towards Chinese scientists, some of the P.O. authors joined force again in March 2023, for a pre-emptive strike that (mis)interpreted CDC sampling data, snatched ahead of the publication of the accompanying preprint by Chinese scientists. That preprint, Gao et al., was published a few weeks later (4 Apr 2023) and reached the opposite conclusion that no animal sold at the market carried SARS-CoV-2. Eventually, a detailed analysis of the data confirmed Gao’s et al conclusions.
It is difficult to imagine why Chinese scientists would want to share key animal data, even if they could: their data is either ignored (Xiao et al.) and/or, worse, hijacked (Xiao et al., Gao et al.), if it does not fit the narrative developed by the authors of P.O., and by other conflicted Western scientists after them, all with the help of Western scientific journals and Western media. It would take a saint to keep trying.
[44] Marion Koopmans had been considered as a potential member of the GVP Board of Directors, back in Dec 2018. As we know from footnote 40, Daszak and Koopmans were also part of the animal/environmental origin breakup group at the WHO Blueprint meeting in Geneva on 11–12 Feb 2020.
[45] The version on virological.org at that precise time was the live one, derived from ‘Andersen.Nature.Perspective.Final_v2.docx’ which had been started by Holmes in the absence of Andersen, based on Andersen previous Google Doc version. This made editing very complex: the Word version was not an easy way to work collaboratively and track changes in real time, it was lagging behind the virological.org version anyway. So later that day, Andersen would tidy that up by putting all the changes back in a Google Doc version and get everybody to work with Google Doc again (17 Feb, 8:53 pm UK).
Anybody who has had to manage some complex collaborative document editing would recognise the pain of moving away from working on an online version to emailing around Word versions. Holmes is old school indeed.
[46] In 2016, Vincent Racaniello took a clear stand against the Gain-of-Function moratorium, in a PNAS article. Some of his words deserve to be reproduced:
The critics of gain-of-function experiments frequently cite apocalyptic scenarios involving the release of altered viruses and subsequent catastrophic effects on humans (8). Such statements represent personal opinions that are simply meant to scare the public and push us toward unneeded regulation. Virologists have been manipulating viruses for years — this author was the first to produce, 35 y ago, an infectious DNA clone of an animal virus — and no altered virus has gone on to cause an epidemic in humans. Although there have been recent lapses in high-containment biological facilities, none have resulted in harm, and work has gone on for years in many other facilities without incident. I understand that none of these arguments tell us what will happen in the future, but these are the data that we have to calculate risk, and it appears to be very low. As shown by Menacherry et al (2) in PNAS, the benefits are considerable.
[Author’s note: the Menacherry paper mentioned is the first one referenced in the presentation for the Farrar call on 1 Feb 2020.]
source: ‘Moving beyond metagenomics to find the next pandemic virus’, PNAS, Mar 2016, https://doi.org/10.1073%2Fpnas.1601512113
[47] Donald G. McNeil made the headlines in March 2024 when he denounced the way Andersen mislead him while drafting Proximal Origin, as was evidenced by the Slack conversations. Previously, in May 2021, Donald G. McNeil had written an article on Medium explaining how he became convinced that the lab leak theory was perfectly valid questioning of a rushed consensus.
All things considered, while he did play a role promoting the GVP and Daszak, to his credit, he eventually saw through the manufactured zoonotic consensus, and did not back off.
[48] The way the glycan question was answered provides a very good illustration of the roles of the different key characters. That question had been asked by Collins to Farrar on 4 Feb, 10:58 am UK, when discussing the rough draft sent. It was forwarded to Holmes, who then relayed it to his co-authors on 4 Feb, 9:59 pm (when Holmes started his day in Sydney). Garry and Andersen answered Holmes by 10:56 pm UK, which may have been too late for it to reach Farrar so that it could forward it back to Collins and Fauci that day.
In the end, the question went back and forth via the two obligatory intermediates (Farrar fronting NIH, NIAID and WHO, and Holmes, fronting the expert scientists working on their analysis) in less than 20 hours.
[49] The last sentence of Fouchier’s email states that ‘I made some additional comments in the attached pdf, also in line with Andrew’s comments.’
This means that he had already made some comments by the time Fouchier answered. But since Rambaut is one of the very authors of the draft, he most likely circulated some clarifications at some stage, and clearly not some feedback to his own draft.
[50] Baric is an exception among the key scientists involved in the origin controversy, as he made it perfectly clear in an interview that nobody could really tell which coronaviruses were in Chinese labs.
[51] It is not clear whether Holmes learnt these points directly from SCAU after asking them about the FCS, or whether he got the news via his back-channels (he has many collaborators in Guangdong province).
[52] We can tell the time the manuscript was sent to Nature from the fact that Andersen came back to the connected world at 2:41 am UK on 17 Feb (7:41 pm 16 Feb in CA), on his way back from a few days in the desert, which, according to Holmes, was just about 15 minutes after he had sent the Manuscript to Nature in the absence of Andersen.
[53] On 17 May 2020, the very day before the WHO Health Assembly, the WHO Chief Scientist, Soumya Swaminathan, played down any research-related origin while showing that China was — once gain — using the access card to steer the WHO in a convenient direction. China would have been dismayed by the audacity of Australia.
What we do know is that this is a naturally occurring virus, that it was not artificially synthesised in the lab. I think that scientists have been quite clear about (that), because there are markers in the genome that would have given it away if it had been a synthesised gene.
We know that most likely this came from a bat. What we don’t know is how and when it actually transferred from the bat to the human, and what this intermediate animal was, and when those events happened and whether it was a single event or a series of events where it jumped. So those things still need to be investigated by scientists.
I hope that there can be some international collaborative effort on this because it will definitely help inform future guidelines and steps that one would need to take to prevent such a thing from happening again.
For those things, we’re in discussions with the Chinese government about having an international team work with the Chinese researchers actually, to understand this much better. There’s been a lot of active collaboration on many, many topics, not on this one specifically.
[54] Zhou et al. was eventually then sent to bioRxiv on 2 Mar, where it was released as a preprint on 5 March, before being published as a paper in Current Biology in May 2020.
[55] Racaniello’s main argument against passaging, in his podcast, was simply that there was no known similar virus in Chinese labs which could have been passaged. But while the Proximal Origin draft, behind the lines, left the question open about what viruses were in Chinese labs — Racaniello was happy to take an absence of evidence as evidence of absence, even calling it ‘the science’.
They point out in this Proximal paper that there’s no known laboratory work done on corona[viruses] that could have been the ancestor of this virus, which is just not close enough, I know some people would not believe that still, but that the science that says it, right?
Source: Racaniello, on TWiV 588 (59:37)
As it is, the evidence of absence simply points to many unpublished samples and viruses in the WIV fridges and databases [footnote 56]. The toned down mention of the passaging option in the Proximal Origin manuscript loaded up on virological.org was meant to avoid too much visibility, it was not meant at all to be evidence of absence.
[56] Francisco de Asis has done some remarkable work, recording all the sampling and sequencing of bat viruses by WIV teams. He was able to point to the lack of published sampling and sequences for collection dates starting from 2016, just as WIV teams were actually sampling broader in South East Asia.
The two to three years delay between collection and publication is rather normal, as the teams involved take the time to process the sample, sequence them, try to isolate some viruses or build them up from scratch from a good consensus sequence, then study them. The enabling technologies, New Generation Sequencing and cheap synthetic constructs, started kicking in precisely around 2017 with strong adoption (NGS) and advances (cheap synthetic constructs) in China.
A prominent Chinese-American scientist with some knowledge of the WIV confirmed on 16 February 2020 (the very day before P.O. was sent to Nature) that the WIV “had many bat samples not yet worked. out or results published”.
Baric, the top coronavirus expert in the world, who worked extensively with the WIV, gave the same message in an interview: “You can engineer a virus without leaving any trace. The answers you are looking for, however, can only be found in the archives of the Wuhan laboratory.”. Baric would later sign a letter in Science asking for a proper investigation of the origin.
Daszak himself highlighted that the WIV had many CoVs not fully sequenced, and 15,000 samples in its freezers, in which logically many more CoVs were waiting to be identified.
[57] In May 2020, another DoE laboratory, the Lawrence Livermore National Laboratory (LLNL), which has expertise on methods of detections of lab-enhanced pathogens, prepared a report considering a lab escape plausible and worth investigating.
Sometime between the Biden intel report summary of Oct 2021 and 11 Aug 2022, the DoE decided to send a revised classified assessment to the Office of the Director of National Intelligence whereby a lab-escape was the most likely origin, with low confidence. That revision was not disclosed until February 2023, when it was incorrectly presented as very recent, despite being at least 6 months old.
[58] The latest Slack message from Rambaut had 9 replies, which were not displayed in the Slack FOI. This happens in other places in the document. FOI handlers (and recipients) may want to be more alert to the problem of properly including replies to messages in the FOIs.
[59] It may be worth asking if the scientists that seriously state that the clear research focuses revealed by DEFUSE could not be pursued in China in 2018–19, without US money, or US technology, are not daydreaming in some form of colonial fantasy.
As I have explained elsewhere, DEFUSE was actually an attempt for the DTRA program to stay in touch with Chinese research in Wuhan, at a time when the US was getting seriously preoccupied by the very rapid development of synthetic biology in China; a development that was likely beyond Chinese ethical and biosafety guardrails. The DTRA intentions were right, but another institution in the DoD, DARPA, correctly considered the risk too high. Hence, the US never managed to get a seat on that Chinese train.
[60] See for instance the WHO SARS Risk Assessment and Preparedness Framework published in Oct 2004:
[61] The GVP deck by Daszak for the DoD is well worth reading. That deck was never disclosed until revealed by a FOI. It put the risk of lab-escape on the same footing as the risk of natural zoonosis, and highlighted the increased risk represented by Gain of Function research. The deck, written in September 2017, was used by Daszak to try to interest the US DoD into backing the Global Virome Project, a $1.3 bln project over 10 years.
I leave it to the reader to consider the implications of another FOI involving thee DoD, discussed in footnote 73, showing that the very same Peter Daszak mentioned privately a few months later, that the WIV would be doing some of the DEFUSE Gain of Function work on coronaviruses at the totally unsuitable BSL-2, at the very same time as he was telling DoD DARPA that the work would be done in the US (where BSL-3 is used instead).
[62] This is exactly what I did back in summer 2020. Because the molecular argument was rather fuzzy at the time (very limited data), and because that lab-product debate carried with it the bitter taste of the never healed battles that punctuated the GoF moratorium of 2014–17, I decided instead go back to basics. I teamed up with a French engineer, who had spent time in Yunnan and installed BSL-3 labs in China (also Mandarin speaker and reserve officer), and used a simple Bayesian approach, based on a review of the risk factors in Wuhan in 2019. In doing so, we specifically made no assumption as to whether the virus was engineered or not, ignored the BSL-4, ignore BSL-2s (while noting that there were documented work on coronaviruses at BSL-2 there), ignored sampling-trip infections, and found that one could not — under any reasonable prior — ignore the possibility of a research-accident in Wuhan, against a zoonosis event first detected there.
See 10.5281/zenodo.4067919.
[63] The usual counter-argument that a zoonotic event that started somewhere else, for instance in a Yunnan farm, could very well have lead to infections being first detected in Wuhan, based on the SARS-1 experience, is unfortunately incorrect: the question is not whether such a scenario is possible, but how likely it is compared to the alternative (i.e. whether it is plausible).
Unfortunately, Wuhan is just one of many cities with wildlife markets where the end products of some infected farm(s) would be sold. So the probability that a zoonotic outbreak would be exclusively detected there, and not anywhere else along the vast supply chains and many supply destinations, is very small.
SARS-1 actually offers the perfect demonstration of the vacuity of the counter-argument, with well documented little outbreaks along many supply destinations, something not seen at all with SARS-CoV-2.
Some scientists then go further by attributing that lack of detected secondary outbreaks, on the way to Wuhan, to a massive cover-up by the Chinese CDC. There is no proof of such cover-up, but it is clearly not impossible. However, the same scientists cannot at the same time denounce the research-related accident hypothesis as requiring a massive cover-up.
Something is sure: a cover-up happened. The only question is whether the Chinese government, via the Chinese CDC, covered up some animal infections in some farms and their distribution networks, or whether the Chinese government covered up a few original human infections in a Wuhan laboratory, or related to a laboratory activity (for instance through waste / used animal handling).
[64] Less than 6 weeks earlier, on 24–25 January 2020, Holmes was attending the ‘30th Challenge in Virology’, an annual conference organised in the Swiss ski resort of Grindelwald, where he did a programmed 30 min presentation on 25 Jan, but also delivered ‘an impromptu session on the new coronavirus’ (source: Spike).
[65] While Clare Thomas was based in London, João Monteiro was based in New York. We nevertheless stay with UK time in this section for simplicity.
[66] The odds that matter are not the relative chance of a villager, somewhere in China, getting infected by SARS-CoV-2 in his surroundings, against a scientist, also somewhere in China, getting infected by SARS-CoV-2 on a sampling trip or in a lab.
The odds that matter are the chance of an outbreak detected in Wuhan, and nowhere else, being caused by a natural virus in some research-related activity involving a Wuhan lab, against some villager getting infected in his surroundings somewhere in China, with first detected infections in Wuhan and nowhere else.
[67] The date of publication of the ABC article is 6 Feb, not 7 Feb. We can see that from the first save on archive.org. Also, if you are very careful, you can see ‘February 6, 5:35 PM’ showing up for a very short time before being changed to ‘February 7, 11:35 AM’.
[68] On 4 Mar, Andersen sent P.O. v17 to Nature Medicine for final approval, just one day before the RmYN20 preprint (Zhou et al.) was released (5 Mar). It would have been difficult for the text to be changed just after submission — this would have only made it sound very ephemeral, when in fact, exactly as planned, the authors could argue that their text already offered documented clues towards the possibility of an eventual natural evolution of an FCS in a sarbercoronavirus (their references 16 and 17), so that no alteration was required.
[69] Strictly speaking, the decision to declare a Public Health Emergency of International Concern is taken by the members of an ad hoc International Health Regulations Emergency Committee, convened under the IHR agreement signed in Geneva in 2005, which is supposed to bind 196 countries. Marion Koopmans and Linfa Wang have been regular members of ‘IHR Emergency Committee for Covid-19’.
[70] Despite the tensions between the Trump administration and the WHO, the US was still the third-largest donor to the WHO for the 2020–2021 period, with about $700 mln, behind Germany and the Bill and Melinda Gates foundation. China came 11th at $168 mln, behind Canada, Japan or the Rotary International.
[71] Fauci’s nemesis, Richard Ebright, did a good job documenting that extraordinary fact. See https://www.theblaze.com/news/lab-wars-inside-one-democrats-20-year-crusade-to-save-the-world-from-anthony-fauci-part-2-2014-2020.
[72] Here again (see footnote 7 for a precedent), Spike could have done with better editing and some careful checks by Farrar. Based on her notes, Farrar’s ghost-writer (Anjana Ahuja) gives the impression that the pushback by Fouchier and Koopmans against Andersen and Holmes happened in some discussion ahead of the call. That is not the case.
[73] DEFUSE was a proposal by EcoHealth Alliance to the PREEMPT program by DARPA, the DoD Advanced Research Projects Agency. DEFUSE was not funded due to gross deficiencies in risk management that ran totally contrary to DARPA guidelines, and despite repeated attempts to reverse their decision. One aspect of DEFUSE was the creation, in volume, of enhanced coronaviruses based preferably on new backbones (and not existing ones) with precisely the divergence of SARS-CoV-2 to SARS-1, to test their ability of infecting humans, via the insertion of an FCS, some point mutations and some passaging. With the passaging and pathogenicity tests to be done by the WIV in 2019.
DEFUSE (just like the GVP pitch to DoD) was never disclosed by Daszak, even to the people who signed his Statement of Support. Instead, DEFUSE came to light through a US Marines whistleblower who worked with DARPA and shared the relevant files with DRASTIC. After weeks of careful validation and analysis, DRASTIC released their summary and the original documents.
One particularly railing aspect of DEFUSE was Daszak’s intention to see some of the work (discovery and culturing for instance) done at P2 in Wuhan instead of P3 in the US, for convenience and speed, against what he was writing at the time in his proposal. A suggestion that strongly upset Baric himself, also part of DEFUSE, as came to light later in some FOI:
As to whether similar work was done eventually in Wuhan, if not already programmed or partly already done when the proposal was made (which would be standard for a large research project that is in perfect continuity of existing work), Baric was quite right when he told a House Committee:
‘I don’t know what China did, and I don’t know what their grant funding was subsequent to this grant’ (Baric-TI-Transcript.pdf, p. 83).
[74] As of the time of writing (Aug 2024), Schwartländer is back in Beijing, as the Global Health Envoy for the German Ministry of Foreign Affairs.
[75] One may despair of the constant confusion in such general statement as ‘Scientific evidence overwhelmingly suggests that this virus originated in wildlife, as have so many other emerging diseases’.
Viruses and diseases are not equivalent; that a virus may originate in wildlife does not mean an outbreak of the related disease must originate in wildlife too. A biosafety failure in a laboratory may do just do as well. This is a good example of Daszak’s art of muddying the water with superficial generalities.
[76] The request is stated clearly in the last paragraph of the letter. The two first paragraphs are some introduction and context-giving. Unfortunately, one sentence there is not very well written:
‘I am writing to ask the National Academies of Sciences, Engineering, and
Medicine (NASEM) to rapidly examine information and identify data requirements that would help determine the origins of 2019-nCoV’
It is not to ‘examine information’, but to identify information requirements, alongside data requirements, as is clearly stated in the key last paragraph. A second sentence in the introduction section is correct: ‘[..] that manuscript highlights the need to determine information and data requirements as quickly as possible to better perform and validate such analyses of origin.’
[77] Based on Fauci’s official diary, Emily Erbelding organised a meeting between Fauci and Baric for 11 Feb at 2:30 pm (room 7A-18 of Fauci’s NIH building).
We have already come across Erbelding in 4.1, when she provided some quick information about the NIH involvements with the WIV, ahead of a briefing of Senators by Fauci on 24 Jan 2020. Back in July 2018, she had also been involved in the lifting of the 2017 GoF term-of-award restrictions on Daszak’s R01AI110964 grant, based on the determination that the grant was not subject to the new HHS P3CO Framework.
[78] The position of Baric on the subject is much better. He understands perfectly the limitations of present research regulations, with their grey areas, as well as the risk-benefit dilemma of finding the optimal level of regulation in the context of biosecurity and of the emerging biotechnological age, the likely battlefields of the 21st century. But, contrary to the likes of Franz and Gigi Gronvall, Baric prefers to leave that problem to the civil society and elected representatives to solve (however difficult that may be):
[..] So over-excessive regulatory restrictions on emerging pathogens or high containment research can be equally disastrous to the U.S. in the future. So there’s a risk-benefit ratio. And if that risk-benefit ratio is wrong, the risk to the competitiveness of the United States could be impacted more than the benefit that would ever occur from the restrictions. And, unfortunately, you guys have to figure that out. I don’t have to figure that out, but you guys have to figure it out.
source: Baric-TI-Transcript.pdf, extracts from p.77
[79] Baric’s House interview contains an amusing section about the suspicion of him being the author of the letter to Cohen:
[80] On 12 Feb (1:12 am EST), Lishan Su contacted Baric for his comments on a manuscript he had prepared with Shan-Lu Liu and Linda Saif for in EMI. Saif contacted Baric a few hours later (11:49 am EST) to tell him that even if she understood from Daszak (who was also thought out for commentaries) that he (Baric) may not want to see his name listed as co-author, it would still be nice to have his comments. Ralph did some comments, made some changes and thanked Lishan Su for his initiative.
So clearly Baric had no problem working as a co-author with a conflict of interest, as long as he could hide it by not mentioning his name.
He was not the only one. That EMI paper that denied any SARS-CoV-2 engineering, was, in fact, written at the instigation of Shan Lu, Editor-in-Chief of EMI, who made significant contributions to it while hiding his role in it by conspiring with Lishan Su (of UNC Chapel Hill, just like Baric) and Shan-Lu Liu, who both never let any of the other co-authors in top their little secret. Then, to cap it all, Shan Lu used his prominent position at EMI (which did not know of his contribution) to get the piece approved immediately for publication without a real peer-review.
As of August 2024, a request for the retraction of that fraudulent EMI paper has still not been answered.
[81] As per his Wikipedia page, James Chau’s WHO appointment by Margaret Chan (then the WHO DG) ‘attracted attention due to his role in presenting forced confessions while working for Chinese state-run broadcaster CGTN’. See for example this FT article and the actual video of the forced confession of British citizen Peter Humphrey presented by James Chau on CGTN in 2013. Eventually, after multiple warnings and statutory penalties for confession broadcasts covering three individuals (Humphrey, HK Simon Cheng, Swede Gui Minhai), CGTN lost its UK broadcasting licence in Feb 2021.
A joint appeal for the WHO to cut its ties with James Chau was sent to Dr Tedros on 3 June 2020, arguing that ‘Armed with his U.N. badge of legitimacy, Mr. Chau has been articulating a one-sided narrative portraying Beijing as heroic, on Twitter, Weibo, YouTube and the broader web since the beginning of the outbreak.’
James Chau is certainly not shy of interviewing reputable figures such as Kofi Annan, Muhammad Yunus, or Jimmy Carter, but also more controversial ones, such as Robert Mugabe, Paul Kagame (the president of Rwanda), and Chinese First Lady Peng Liyuan (the wife of Xi Jin-ping). James Chau is still WHO Guest Ambassador, and Schwartländer had yet another softball interview with him in April 2023, on his return to China as the German Global Health Envoy.
[82] When contacted by the Washington Post, Ebright only pushed back against the bioweapon rumours, not against a possible lab escape. As with many similar early articles on the origin question, the Washington Post piece was eventually expunged of the conspiracy / debunk vocabulary a year later.
[83] As per the history of edits of the site, it seems that Rambaut may have loaded a version of the manuscript on virological.org in the late evening the previous day (16 Feb), possibly just after Farrar and Collins approved the latest version, and possibly not in public mode.
[84] See quote from an article in the Süddeutsche Zeitung of 9 Feb 22:
What did you think when you first heard about the experiments in Wuhan?
Drosten: It really didn’t have to be. Above all, some people in the USA knew about these experiments. Right from the start, when these public allegations came up, one should have communicated aggressively and proactively what was being done in the laboratory. At the time, many scientists, including myself, put their hands in the fire for their colleagues from Wuhan in a statement in the medical journal The Lancet, but were not informed about these projects. If I had known about it, I would have at least asked questions before signing.
[85] That paragraph argued only against deliberate engineering, based on the allegedly non-optimal binding of the RBD, and supposing that deliberate engineering would have used a binding that was expected to be optimal, in its construction plan.
That ‘known optimal binding’ argument does not apply to passaging, which simply finds its own way to what could, in retrospect, easily look like a theoretically and/or practically suboptimal RBD.
[86] We note that, in Spike, Farrar is totally silent on what happened during that key month, from the time of the upload on virological.org on 17 Feb, to the eventual publication in Nature Medicine on 17 Mar after Nature’s rejection.
[87] Both Daszak and Andersen were appointed to that Committee. The Committee does not seem to have ever discussed the origin question, even when it became a hot topic again.
[88] Donald McNeil, the reliable GVP enthusiast at the NY Times (see 28.2), would again oblige, with another article pushing for global virus hunting published on 30 Aug 2020. That article was of the same ilk as the one that had caused Andersen’s ire in Oct 2019. To be fair, McNeil has since then denounced Daszak as having manipulated him [footnote 47].
[89] As documented in another Medium piece, as soon as his NIH grant was interrupted, Daszak benefited from some ‘bridge-financing’ via private foundations with the help of well-placed people. This type of funding has the added benefit of being untraceable, with no reporting requirement beyond a mention of the beneficiary in the tax declaration of the foundation, and an occasional mention of the benefactors in EHA publications. The title of some of his grants point to continued focus in South East Asia.
[90] Mentions of Dennis Carroll joining the Biden campaign can be found in David Morens’ FOI’d emails.
[91] As an exercise, in green below are elements of vocabulary that may be worth comparing to contemporary tweets or statements of the candidate whistleblower, at around the time the letter was sent (25 Jul 2020 +/- 3 months).
[92] The usage of ‘back-channels’, private emails instead of work email, is a constant feature of the effort to obfuscate the relation between Fauci and Daszak, managed via Morens.
One email by Gerald Keusch (‘Jerry’) nevertheless gave the game away; in an email sent in Oct 2021, when FOI requests were accumulating, asked Morens to purchase a personal mobile phone and to stop using a government-issued phone for his Gmail, as that made them FOI-able. Keusch also explained that he would do the relay with Daszak, while David Morens was not to be directly in touch with Anthony Fauci (‘Tony’). All as per Fauci’s instructions.
[93] We do not know exactly when Farrar learnt of the Statement of Support before he offered to get it published in Lancet on 16 Feb. What is nevertheless sure is that:
- already on 7 Feb, Farrar was on a Daszak’s list of people to contact as a potential signatory of the Statement,
- Farrar and Daszak would have had the occasion to meet in person in Geneva on 11–12 Feb 2020 as part of the WHO Blueprint meeting [footnote 40].
Nevertheless, we also observe that Farrar did not seem to intersect with the Statement of Support until 16 Feb; he was not actually even yet listed as a signatory of the Statement Draft circulated at the time (see 19.5.c), despite mentioned as such by Daszak in his email that day. This means that Farrar most likely agreed to be a signatory on or just before 16 Feb.
We may also note that Daszak’s conflict-of-interest issue was relevant to Farrar in the scope of a side check for the MI5 — standards and confidentiality have to be abided to there. Hence, Daszak was excluded from the Farrar call. But when it comes to shaping a message for the benefit of WHO access to China, it was clearly a totally different game, with different stake-holders and objectives, and there the conflict mattered much less (especially when not revealed in the text of the Statement).
[94] Some may object to my describing Drosten as a GoF practitioner. No harm is intended; the problem seems to be that entirely different conclusions may be reached based on how one slices, dices, and then recombine the different elements of a definition, and keeping in mind what the expected audience was at the time.
For instance, Drosten himself has testified under oath that he ‘does not run any “gain-of-function experiments” in the sense of the theories about a laboratory origin of SARS-CoV-2, i.e. experiments in which viruses are created that do not exist in nature and contain a predictable increased danger to humans’. Still, the web page of the RAPID project (focused on MERS), for which Drosten is coordinator, lists ‘Identification of host factors by loss-of-function and gain-of-function experiments’ as one of its subprojects. Last, Simon Wain-Hobson (Emeritus Professor, Institut Pasteur), was asked to review the RAPID project and concluded that such work should rationally be considered as having a real potential for increased pathogenicity, which would make it fall under his own definition of GoF, and would also be in concordance with the web page mentioned above.
[95] While the NIH under Collins had suspended Daszak’s R01 grant, the NIAID under Fauci was working hard to alleviate the suspension effects [footnote 89] and to try to get that suspension revoked. In relation to Jon Cohen’s article on 25 Jul, we can see for instance how Morens was in contact with Cohen that very day, following publication, to try to further shape its narrative.
[96] A recent important FOI, NIH-55351–56077-December-Partial-Production_Redacted.pdf, details the composition of the advanced team and its objective (p. 214), the composition of the team that went to Wuhan (p. 4, 40), the extension of the WHO mission by two days for that purpose (p. 99), the presentation of the WHO report findings and conclusions to the Chinese Minister of Health during the short visit to Wuhan on 23 Feb (p. 4), the fact that the US CDC team in Beijing had a tentative visit to Wuhan on its WHO team agenda as early as 13 Feb (p. 226), the contacts between the US members of the WHO team and the US HHS/CDC personel in Beijing (p. 31–36).
Appendix A: The evolution of a scenario
Here we shall examine the progression of the treatment of the lab-product scenario over the various drafts of P.O. We will purely focus on the sections that communicate an explicit or implicit evaluation of the probability of the lab-product hypothesis. All other elements of the drafting progresses are ignored.
Sources:
A document produced by the Select Subcommittee on the Coronavirus Pandemic (SSCP-Drafts-of-Proximal-Origin.pdf), which I originally found on their website by chance, includes a history of the key versions. That document is our main source, but one still needs to use the FOId emails (farrar-fauci-comms.pdf) to confirm which versions exactly were sent by Farrar to Fauci and Collins, as he kept them informed of the drafting progresses.
On dating the drafts:
The watermarks in SSCP-Drafts-of-Proximal-Origin.pdf give us a time when the version existed. However, be aware that it may have been created at an earlier time. For instance, version 5 has a watermark date of 6 Feb 2020, 3:09 am (EST), but that exact version was shared by Farrar with Fauci and Collins on 4 Feb 3:53 pm EST, after a quick tidying up of version 4.
Resources:
I extracted the various versions of the Proximal Origin draft and of the Rebuttal letter from SSCP-Drafts-of-Proximal-Origin.pdf. The documents were further tidied up by removing the watermarks. For easy identification, the watermark preserved in a non intrusive-text box in the top right first page of each PDF (be aware that in the case of the versions distributed to Fauci and Collins, the watermarks are misleading: the so called ‘ First Distributed Version’ is really the second one, and the ‘Second Distributed Version’ is in fact the third one).
The files are there: Proximal Origin drafts and Rebuttal letter drafts. The reader is encouraged to use a PDF comparison tool, such as the very good and free www.ilovepdf.com. Here is an example of comparison (15 to 17 Feb, with the change to ‘improbable’ requested by Farrar):
Overview of versions:
See https://bit.ly/POVersions for a detailed review of the available versions of the text.
P.O. History (selected versions):
As there are at least 20 documented versions of the text up to publication in Nature Medicine, we shall focus here only on some key versions that are important to follow the major changes in the non-natural scenarios:
(0) Versions 1 to 3: 1 & 2 Feb 2020 as per SSCP-Drafts-of-Proximal-Origin.pdf:
These early versions have an identical wording in the main description of the non-natural scenarios:
Bioweapon. Highly unlikely and there is no data supporting this hypothesis.
Specific engineering. Unlikely as this would require significant amounts of work utilizing uncommon and currently unknown backbones of SARS-like bat CoVs. For this type of work, there are preexisting backbones that could have been utilized, but they clearly were not.
Tissue culture passage. The data is consistent with this scenario, although no specific hypothesis exists for how the furin site was gained, but could be due to passage in tissue culture. The virus could have been released via accidental infection of researcher(s).
(1) Version 4: 3 Feb 2020 11:59 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 8). Also, as per Farrar’s emails (p. 98), the very first draft (Summary.docx) sent to Fauci and Collins on 4 Feb at 7:01 am UK.
- bioweapon mention dropped while a mention of conspiracy theories is added
- passaging remains a perfectly valid hypothesis
- ‘specific’ engineering renamed ‘deliberate’ engineering (which is slightly confusing but means direct genetic manipulation) and remains excluded.
- a new ‘Limitations and recommendations’ section is added, with a minor note about passaging.
Overview. As rumours have been circulating about this virus being engineered or otherwise created with intent, we wish to make it clear that our analyses show that such scenarios are largely incompatible with the data.
Evolution of 2019-nCov. Three main scenarios could explain how 2019-nCoV acquired the features discussed above: (1) natural selection in an animal host, (2) selection during passage, or (3) deliberate engineering. As described in the beginning, engineering (#3) can be ruled out with a high degree of confidence as the data is inconsistent with this scenario. [..] the genetic data clearly shows that 2019-nCoV is not derived from any previously used virus backbone, including those recently posited by various conspiracy theories, based on a 2015 paper in Nature Medicine. The other two scenarios are largely indistinguishable and current data are consistent with both.
Selection during passage. It is possible that 2019-nCoV could have acquired the RBD mutations and furin cleavage site as part of passage in tissue culture, which have been observed in previous studies with e.g., SARS-CoV. However, it is less clear how the O-linked glycans — if functional — would have been acquired [..]
Limitations and recommendations. The identification of a potential intermediate host of 2019-nCoV as well as sequencing of very early cases, including those not connected to the market, could also help refute the passage scenario described above.
(2) Version 5a: 4 Feb 2020 as per Farrar’s emails (p. 88), in Summary.pdf sent to Collins and Fauci on 4 Feb 8:53 pm UK. Also attached in an email to them on 5 Feb 6.57 am UK time (p. 83), when answering some question about the glycans. That version is also in SSCP-Drafts-of-Proximal-Origin.pdf, (p.14), watermarked ‘6 Feb 2020, 3:09 am (EST)’, when it was clearly already produced by 4 Feb 3:53 pm EST from Farrar’s emails.
- some wording changes, but essentially same weighing of the 3 hypotheses.
- ‘Limitations and recommendations’ is more specific about the plausibility of all three scenarios.
Overview. [same as 3 Feb]
Evolution of 2019-nCov. As described in the beginning, we believe deliberate engineering can be ruled out with a high degree of confidence as the data is inconsistent with this scenario. [..] the genetic data clearly shows that 2019-nCoV is not derived from any previously used virus backbone, including those recently posited by various conspiracy theories, based on a 2015 paper in Nature Medicine. Three main scenarios could explain how 2019-nCoV acquired the features discussed above: (1) natural selection in humans, (2) natural selection in an animal host, or (3) selection during passage.
Selection during passage. [same as 3 Feb]
Limitations and recommendations. The evolution scenarios discussed above are largely indistinguishable and current data are consistent with all three. It is currently impossible to prove or disprove either. [..] The identification of a potential intermediate host of 2019-nCoV as well as sequencing of very early cases, including those not connected to the market, could also help refute the passage scenario described above.
(3) Version 6a: 7 Feb 2020 as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 26, labelled ‘Second Distributed Version’, but actually the third one sent to Fauci and Collins). Also, as per Farrar’s emails (p. 66), in Summary.Feb7.pdf sent on 7 Feb, 6:09 am UK.
Essentially the same one (with just a few added notes) was also sent by Farrar to the full call group on 8 Feb, 9:45 am UK (Proximal_Origin_Emails.pdf, p.57).
- ‘Selection in animal host’ introduces the Lam et al. pangolins, with the argument about the identical six key RBD residue. Previously, that section had hardly any data to show.
- mention of conspiracy theories dropped, compensated in party by bolding of text in ‘Overview’
- possible confusion introduced in Overview, as ‘deliberate engineering becomes ‘laboratory construct or experimentally manipulated virus’, since a passaged virus can be said to be experimentally manipulated. Clearly, the authors have problems finding the right words.
- ‘Evolution of 2019-nCov’ became ‘Origin of 2019-nCoV’ which is unfortunate, as the draft does not look at all possible origins: it ignores a research-related accident with a collected virus that was not altered in a lab, as there is no way to distinguish this from a zoonosis based on the sequence. Otherwise, just some stylistic changes in that section.
- unchanged wording in the ‘Selection in passage’ section, which shows that passaging is actually still considered a valid hypothesis.
Overview. Analysis of the virus genome sequences clearly demonstrates that the virus is not a laboratory construct or experimentally manipulated virus [Gilles’ note: sentence in bold in text].
Origin of 2019-nCoV. As noted at the start of this document, we believe that the origin of 2019-nCov through laboratory manipulation of an existing SARS-related coronavirus can be ruled out with a high degree of confidence. [..] the genetic data clearly shows that 2019-nCoV is not derived from any previously used virus backbone, for example those described on a 2015 paper in Nature Medicine. Instead we believe one of three main scenarios could explain how 2019-nCoV acquired the features discussed above: (1) natural selection in humans, (2) natural selection in an animal host, or (3) selection during passage.
Selection in animal host. [..] Notably, provisional analyses reveal that Malayan pangolins (Manis javanica) illegally imported into Guangdong province contain CoVs that are extremely similar to 2019-nov”. Although RaTG13 remains the closest relative to 2019-nov across the genome as a whole, the Malayan pangolin Covers are identical to 2019-nCoV at all six key RBD residues. Analyses of these pangolin viruses are ongoing, although they do not carry the furin cleavage site insertion. [..]
Selection during passage.[same as 3 Feb, just cell culture for tissue culture]
Limitations and recommendations. [same as 5 Feb]
(4) Version 6c: 8 Feb 2020 5:45 am (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 38)
- ‘Overview’ becomes ‘The proximal origin of SARS-CoV-2’, which is again very unfortunate as the document cannot address a research-related accident with a natural virus, and does not even mention that limitation.
- 2019-nCoV changed to SARS-CoV-2
The proximal origin of SARS-CoV-2. [beyond renaming of section, same as 7 Feb]
Theories of SARS-CoV-2 origins. [beyond renaming of section, same as 7 Feb with some minor stylistic changes].
Selection during passage. [same as 7 Feb]
Limitations and recommendations. [same as 5 Feb]
(5) Version 7a: 10 Feb 2020 8:54 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 44):
- SARS-CoV-2 changed back to 2019-nCoV by Lipkin.
- In ‘The proximal origin of 2019-nCoV’, the bolded section is removed, and some welcome clarification is introduced as to the limitations of the sequence analysis. Still, the possibility of a research-related accident with a natural virus not discussed.
- ‘Origin of 2019-nCoV’ becomes ‘Theories of 2019-nCoV origins’, which is still confusing.
- ‘Limitations and recommendations’ shortened and becomes ‘Conclusions’. Messaging as to the validity of the passaging scenario is weakened in the process.
The proximal origin of 2019-nCoV. Herein, we review what can be deduced about the origin of this virus from the comparative analysis of available
genome sequence data. [..]. Importantly, this analysis clearly demonstrates that 2019-nCoV is not a laboratory construct or experimentally manipulated virus.Theories of 2019-nCoV origins. [same as 7 Feb, with some minor stylistic change].
Selection during passage. [same as 7 Feb]
Conclusions. Although we can readily dismiss the idea that 2019-nCoV is an experimentally manipulated virus, it is impossible to prove or disprove other theories of its origin [..], Identifying the immediate non-human animal source and obtaining virus sequences from it would be the most definitive way of revealing [the] virus origins
(6) Version 8: 12 Feb 2020 6:38 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 56):
- 2019-nCo changed back to SARS-CoV-2 by Holmes (see Slack, p.39, 12 Feb, 6:07 EST).
- ‘Theories of SARS-CoV-2 origins’ is reworded, without change in meaning.
- In ‘Selection during passage’, the last sentence about the uncertain role of the seafood market is removed.
- 2019-nCoV changed back to SARS-CoV-2
The proximal origin of SARS-CoV-2. [same as 10 Feb, except that ‘this analysis clearly demonstrates’ becomes ‘this analysis provides evidence’]
Theories of SARS-CoV-2 origins. We believe it is unlikely that SARS-CoV-2 emerged through laboratory manipulation of an existing SARS-related coronavirus. [..] Instead, we propose three scenarios that plausibly explain the origin of SARS-CoV-2: (1) natural selection in humans, (2) selection during passage in culture, and (3) natural selection in an animal host.
Selection during passage. [‘it is also unclear how the virus would be linked to the fact that the epidemic seemed to ‘take off’ at a particular food market, although the exact role of this locality is currently uncertain.’ dropped]
Conclusions. Although genomic evidence does not support the idea that SARS-CoV-2 is a laboratory construct, it is currently impossible to prove or disprove the other theories of its origin described here, and it is unclear whether future data will help resolve this issue.
(7) Version 9: 13 Feb 2020 8:10 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 62):
- ‘Selection during passage’ gets much longer and the passaging scenario is weakened to ‘extremely unlikely’, with some apology for even considering the scenario.
- ‘Selection in an animal host’ is twice as long, and pushes more against the idea that the FCS evolved in a pangolin.
- In the ‘Conclusions’, a better wording (‘laboratory construct’) is used that tends to exclude passaging from the list of rejected scenarios, instead of the previous wording (‘experimentally manipulated virus’). that tended to include it in that list.
The proximal origin of SARS-CoV-2. [same as 12 Feb]
Theories of SARS-CoV-2 origins. [same as 12 Feb, with some cleaning up of style]
Selection during passage. Because we have taken an unbiased approach to attempt to discern the origin of SARS-CoV-2, we considered the possibility that SARS-Cov-2 originated in a laboratory and inadvertently infected a laboratory worker who transmitted the infection. We consider this scenario to be extremely unlikely. However, after the emergence of SARS-CoV-1 several instances of laboratory acquisition of this virus by laboratory personnel working under BSL-2 containment have been documented and thus we cannot eliminate this possibility beyond doubt.
Conclusions. [same as 10 Feb, except for ‘laboratory construct’ replacing ‘experimentally manipulated virus’]
(8) Version 11a: 15 Feb 2020 4:52 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 75). On 16 Feb, the next version, v11b (missing but likely close to 11a), was shared with Fauci & Collins and sent to Nature.
- ‘The proximal origin of 2019’ is reworded and longer, but without change in meaning as to plausibility of the three scenarios.
- ‘Theories of SARS-CoV-2 origins’ is reworded to de-emphasise the passaging scenario, by not including it into the two scenarios ‘proposed’.
- The ‘Adaptation to humans’ is renamed ‘Cryptic adaptation to humans’
- ‘Selection during passage’ is reworked but rather clumsily, with the previous minor concerns now elevated to major ones, especially the O-linked glycan argument against passaging (since then proven wrong). Also, somehow ‘deliberate release’ is included in the same bag, maybe as a way to discredit passaging by association. It appears that even the sink was thrown at it.
The proximal origin of SARS-CoV-2. [same as 12 Feb as to scenarios plausibility wording]
Theories of SARS-CoV-2 origins. It is unlikely that SARS-CoV-2 emerged through laboratory manipulation of an existing SARS-related coronavirus. [..] Instead, we propose two scenarios that can plausibly explain the origin of SARS-CoV-2: (1) natural selection in a non-human animal host prior to zoonotic transfer, (2) natural selection in humans following zoonotic transfer. We also discuss whether selection during passage in culture could have given rise to the same observed features.
Selection during passage. Basic research involving passage of bat SARS-like coronaviruses in cell culture and/or animal models have been ongoing in BSL-2 for many years in multiple laboratories across the world23–26. There are also documented instances of laboratory acquisition of SARS-CoV-1 by laboratory personnel working under. We must consider, therefore, the possibility of a deliberate or inadvertent release of SARS-CoV-2. In theory, it is possible that SARS-CoV-2 acquired the observed RBD mutations site during adaptation to passage in cell culture, as has been observed in studies with SARS-CoV2 as well as MERS-CoV2. However, the acquisition of the polybasic cleavage site or O-linked glycans — if functional — argues against this scenario. New polybasic cleavage sites have only been observed after prolonged passaging of low pathogenicity avian influenza virus in cell culture or animals. Furthermore, the generation of SARS-CoV-2 by cell culture or animal passage would have required prior isolation of a progenitor virus with a very high genetic similarity. Subsequent generation of a polybasic cleavage site would have then required an intense program of passage in cell culture or animals with ACE-2 receptor similar to humans (e.g. ferrets). It is also questionable whether generation of the O-linked glycans would have occurred on cell culture passage, as such mutations typically suggest the involvement of an immune system, that is not present in vitro.
Conclusions. [close to 10 Feb]
(9) Version 12a: 17 Feb 2020 2:32 pm (EST) as per SSCP-Drafts-of-Proximal-Origin.pdf (p. 85). Version published on Virological.org and shared with WHO. Further changes on virological.org that day (see v12b).
- Overall, a rebalancing towards passaging via the tweaks in the ‘Conclusion’
- In ‘The proximal origin of 2019’, ‘experimentally manipulated virus’, which sounds like it should include passaging, becomes ‘purposefully manipulated virus’, which sounds like it should NOT include passaging.
- in ‘Conclusions’, ‘genomic evidence’ correctly reframes the analysis
The proximal origin of SARS-CoV-2. [..] Importantly, this analysis provides evidence that SARS-CoV-2 is not a laboratory construct nor a purposefully manipulated virus.[..]
Theories of SARS-CoV-2 origins. [same as 15 Feb]
Selection during passage. [same as 15 Feb]
Conclusions. Although genomic evidence does not support the idea that SARS-CoV-2 is a laboratory construct, it is currently impossible to prove or disprove the other theories of its origin described here, and it is unclear whether future data will help resolve this issue.
(10) Version 17: 3 Mar 2020 6.09 pm (EST) as SSCP-Drafts-of-Proximal-Origin.pdf.
First version to include ‘The finding of SARS-like coronaviruses from pangolins with near-identical RBDs, however, provides a much stronger and parsimonious explanation for how HCoV-19 acquired these via recombination or mutation’ in the ‘Selection during passage’ section.
(11) Version 19: published in Nature Medicine as a Correspondence on 17 Mar 2020, after the latest manuscript version sent around 5 Mar. This is the final text. To allow an easy comparison with the drafts, I reformatted the text using the format of the last draft (v26).
This published version has many changes with the latest draft (v18b of 5 Mar), mostly due to the required pruning of about 600 words and of about 20 references. The tone and gist is otherwise the same.
Appendix B: The evolution of a name
The multiple changes of naming convention for the virus during the drafting of Proximal Origin can be confusing.
First. on 12 Jan 2020: WHO provisionally named the virus 2019-nCoV, after Zhang and Holmes forced the release of the sequence on 11 Jan, which confirmed to all that the pathogen was a novel coronavirus.
Then on 5 Feb 2020, the International Committee on Taxonomy of Viruses (ICTV) a paper for publication in Nature Microbiology, recommending SARS-CoV-2 as name (published 2 Mar). It also sent it to bioRxiv on 7 Feb (released 11 Feb). Baric and Drosten were both part of the 17 members of the committee.
Baric gave a good description of the ICTV decision on 5 Feb, of which he was part, with its justification and the way China reacted. His own reaction to the Chinese pushback was that he had no time for their ‘nonsense’:
Q: Do you recall receiving any pushback from the Chinese?
A: The Chinese were very unhappy about that.I think several members of the committee received a lot of pushback. I believe they ultimately wrote a paper that they published saying that — giving their reasons why they didn’t
like that name.
Q: Do you recall any of the reasons?
A: I actually didn’t read the paper, because I didn’t want to put up with the nonsense.source: Baric-TI-Transcript.pdf, extracts from p.22–23
After the decision of ICTV, the P.O. draft followed its trajectory towards SARS-CoV-2:
- 7 Feb: Holmes had learnt about the ICTV choice of SARS-CoV-2, which had not yet been officially published. Holmes was also well aware that ‘China will hate it’. The P.O. draft kept 2019-nCoV. The ICTV decision, based on clear rules, was indeed followed by some strong pushback. that was suspected to be motivated by the fear of the Chinese government, and of its scientists, of losing face for having allowed another SARS-like outbreak to happen.
- 11 Feb: ICTV preprint making public SARS-CoV-2 was released, WHO took note of the new name (news). The P.O. draft changed the name to SARS-CoV-2 that day.
- 12 Feb: Lipkin was given access to the P.O. draft and changed the name to NCoV-19. Holmes changed it back to SARS-CoV-2 the same day.
- 19 Feb: Shi Zhengli, George Gao, Shibo Jiang and others published a short piece in Lancet asking for the virus to be renamed HCoV-19. Since, in the end, the taxonomy of the virus is decided by the ICTV, to which Baric belonged, and not by considerations of face, Baric could not be bothered reading the Lancet piece and considered the whole pushback to be ‘nonsense’.
- 25 Feb: Farrar asked the manuscript to use HCoV-19, following the preference of China relayed by the WHO. The manuscript is changed accordingly.
- 4 Mar: Nature asked for the manuscript to use SARS-CoV-2, in accordance with ICTV.
It is interesting to see how the WHO was still trying to do the bidding for the Chinese on 25 Feb, despite the decision of ICTV, and how Farrar was trying to have the draft fall in line.
Appendix C: Plausible research-related accident scenarios
Too often, scientists have expressed definite certitudes about the plausibility of a research-related accident without having first delineated the various plausible scenarios. Then, based on a some reductive research-accident scenario which ignores the real-world complexity, the same scientists usually brandished some supposed necessary conditions for a research-related accident to have happened (such as the virus having been first sequenced or isolated). These conditions regularly turn out to be wrong on closer inspection.
In consequence, the author took the pain to summarise the plausible research-accident scenarios, and their evidential implications, for a letter to the WHO written with the Paris Group in April 2021. I shall reproduce my summary here:
Appendix D: Virus Genesis Scenarios
A good summary of the various virus origin scenarios was produced by the Paris Group for the WHO letter of April 2021. A slightly modified version — that highlights the difference between ‘Lab Product’ and ‘Lab Construct’ — is included below.
One important point: discovery and culturing of viruses (done at BSL-2) in China is clearly dangerous work that could lead to some infection. But, quite often, the responsible virus may not have ever been sequenced, which totally belies the oft-repeated statement that if the virus came from a lab, its sequence would be in a database.
In fact, culturing/discovery is often used to try to retrieve a proper virus sequence, while another portion of the initial sample (typically bat guano), sent for shotgun sequencing, may not deliver reads with enough depth to obtain quality viral sequences for everything that is there. To make things slightly worse, the culturing part can also cause some recombinations between viruses present in guano samples, so that one may even come back with a recombinant virus that was not in the original sample. This was well explained by Baric in his House interview (Baric-TI-Transcript.pdf, p.84).
Last, please note that scenarios VS2 to VS4 below are not exclusive, something that P.O. never looked into. In particular, VS4 (deliberate engineering) would typically be followed by VS3 (passaging) to iron out issues and retain/optimize the desired function.
Appendix E: Evolutionary vs. Outbreak Origins in P.O.
