Note: This is a freely accessible serialized version of Lab Leak Fever. Audio voiceover was AI generated for accessibility. Find an overview of all chapters here or consult the book website for further information.
Independent and interdisciplinary science is important. When evolutionary virologist Michael Worobey, a professor at the University of Arizona, got involved in the search for the origins of COVID-19, he had no idea what he got himself into. With a “soft spot for wild theories,” at least according to a former colleague, and a track record of tackling hotly debated theories around dangerous viruses, the renowned scientist is a force to be reckoned with. NPR even called him the Sherlock Holmes of origin investigations for his work on identifying the origin of HIV by hunting for chimpanzee samples in Kisangani, Eastern Congo. A dangerous trip where Mike developed a life-threatening infection after he injured himself and where his mentor, Bill Hamilton, contracted malaria. Only one of them survived. Yet it was critically important work. Mike’s field sampling and phylogenetic analysis, together with that of Prof. Beatrice Hahn, was instrumental in debunking the widely propagated notion that HIV came about from a contaminated polio vaccine trial in the 1950s. They discovered that HIV had its origin decades earlier, before the turn of the 20th century, and was spawned by at least four separate human-chimpanzee contacts that seeded the outbreak near Kinshasa (called Leopoldville in colonial times) and would spread for decades before it was recognized scientifically in the 80s by making people sick in Los Angeles in the US. Based on Worobey’s data, they reasoned that somewhere around 1910, HIV-1 emerged in humans during a period of rapid urbanization and demographic change (Leopoldville was the largest city in the region at that time) and thus was a “likely destination for a newly emerging infection.”
In a 2022 podcast conversation with Kristian Andersen, the YouTube science communicator Sam Gregson, and myself, Mike recalled had been frustrated by the inconclusiveness of the WHO mission report. The WHO mission opened more questions about the origins that it answered, or at least that had been his impression given the media environment. “I never had a moment where I thought the furin cleavage site needed a non-natural explanation… As an evolutionary biologist, evolution can certainly deal with that pretty aptly,” he explained where he came from. “What was a bit of a curveball to me is that quote here: ‘Market authorities have confirmed that no illegal trade in wildlife had been found’” he elaborated on a different occasion, explaining why he grew hesitant about the market hypothesis.
At the time, he was unaware of the struggles the WHO mission had in getting their Chinese counterparts to admit to wildlife being sold at the market. “I have been amongst the most open scientists to this idea that at least some form [of] a lab incident, maybe even with a virus that has not been characterized by the lab, could have infected someone,” Mike Worobey admitted. “So, I sort of initiated this fateful letter in Science magazine.” Mike reached out to virologist Jesse Bloom, a well-known lab leak proponent on the origin question, to organize the letter titled “Investigate the origins of COVID-19” to the journal Science (published in May 2021). They, along with 16 other authors, such as Alina Chan, Ralph Baric, and David Relman were arguing for giving the lab leak theory a “proper” investigation that should be objective, transparent, data-driven, and “subject to independent oversight.” The letter to Science made a lot of waves internationally and contributed to the vibe shift that legitimized the lab leak theory, ultimately prompting the Biden administration to start their 90-day intelligence investigation.
Despite giving the impulse, Mike was not involved in drafting the wording of this letter, which came out way more accusatory of China, specifically Shi Zhengli, than he was comfortable with and retrospectively regretted. “That letter then took on a life of its own,” he recalled. All he wanted was to give the origin investigation another look. “I was pretty naive about how that letter would land,” he said with chagrin. “I should not have been and regret the tone.” Politics aside, Mike was a man of his word and no-nonsense scientific rigor. Evidence mattered to him, not political implications. The lab leak community was energized by believing scientists of such a caliber were now on “their side.” Mike was celebrated as a hero. However, soon enough, his work would become a target of their ire, and he would be cast as the greatest traitor to their cause.
Like any good scientist worth their salt, Mike set out to poke holes into the natural origin hypothesis, starting by trying to falsify the outbreak association with the Huanan seafood market in Wuhan. Wuhan is a huge place, one-and-a-half times the size of the five boroughs of New York City. “It has a whole lot of places where you might notice the first cluster of a respiratory infection,” he said. “Think about all the places where it could have emerged if it did not start at the Huanan market. We have to think about what are the chances it would pop up there?” That would be a remarkable coincidence indeed. If one were to make the case for a lab leak, that wildlife market and the early patients associated with it had to be explained somehow. Was the market maybe just an amplifier event? Did the Chinese authorities just look there preferentially but not in other places in Wuhan? How did the doctors decide which patients to test for COVID-19? If market affiliation was a criterion for testing patients, then maybe the patient association with it would be a mere mirage, something called ascertainment bias.
These are all scenarios that could potentially explain why the case epidemiology looked like the virus came from the market when, in fact, it might have come from somewhere else entirely. After all, in the wake of SARS, China had set up an early warning and reporting system for detecting unknown viral diseases, which kicked in on January 3rd. This system might have led to an undue focus on the Huanan market. “There is, however, a way to step back to a period before any such bias could have crept in, by considering what happened in the hospitals that first pieced together that a new viral outbreak was underway,” Mike would state in his paper titled “Dissecting the early COVID-19 cases in Wuhan,” published in Science in late 2021.
“I was focused, largely by myself in my basement, for month after month after month at what is going on in Wuhan spatially,” Mike explained how he spent his summer of 2021. Not only did he analyze the WHO mission report and all the scientific papers in Western and Chinese journals, he was also “reading news reports [and] digging of the web archive [for] some of these reports that had gone out by Chinese public health officials before the national authorities even knew the pandemic.” In the quiet of his isolation, following up on every single patient’s history and how, when, and where they were diagnosed, he reconstructed what happened.
On December 27, Dr. Zhang Jixian, a clinician and respiratory specialist at the Hubei Provincial Hospital of Integrated Chinese and Western Medicine (HPHICWM) near the Huanan market, noticed characteristic lesions in CT scans of the lungs of two pneumonia patients that reminded her of something she had seen almost two decades ago. The patients were a couple brought in by their son, who looked “ostensibly healthy.” Nevertheless, she asked him to also do a CT scan, and sure enough, his lungs were full of lesions as well. Lesions she had seen before—with SARS. “At that point, she figured it was probably related to SARS and that it was transmissible to humans because it had infected all three members of the family,” Mike explained.
She also realized that patients could be potentially asymptomatic, running around and spewing the virus all over the place. From that point forward, she paid attention, and while the initial three patients did not have an association with the Hunan market, the next four patients who came to her hospital with the same symptoms all worked at the market. “At that point, on December 29th, she and the administration of her hospital got in touch with municipal and provincial health authorities.” The hospital administrators called other hospitals that were not close to the Huanan market for pneumonia of unknown etiology patients; it turned out that most of their patients were also linked with the market. The Hubei Provincial Hospital “identified both the outbreak and the Huanan Market connection and passed on these fully formed discoveries to district, municipal, and provincial public health officials by 29 December,” he concluded. They were not the only ones. Mike would write in his paper:
A notably similar situation unfolded at Wuhan Central Hospital. On 18 December, Ai Fen, director of the emergency department, encountered her first unexplained pneumonia patient, a 65-year-old man who had become ill on either 13 or 15 December. Unbeknownst to Ai at the time, the patient was a delivery man at Huanan Market. [...] By 28 December, Wuhan Central Hospital had identified seven cases, of which four turned out to be linked to Huanan Market. Notably, these seven cases, like those at HPHICWM, were ascertained before epidemiologic investigations concerning Huanan Market commenced on 29 December.
These findings are important because they highlight how the unknown pneumonia cases before the 29th of December were independently picked up by various hospitals. The market link became known only after, thus dispelling any notion of “ascertainment bias” being responsible for the diagnosis or discovery of SARS-CoV-2 patients. So, while Mike had set out to disprove the market theory, he dramatically strengthened its case. Many of the earliest patients fell sick at the Huanan market.
But maybe it was just an amplification event while COVID-19 was spreading throughout Wuhan more sporadically. Mike wanted to test the idea that the market was not the place where the human-to-human transmission chains actually started. He did this by looking closely into all of the patients who fell ill in December 2019.
Out of 164 early patients (December 2019) within the city boundaries of Wuhan that the WHO mission had identified, he was able to reconstruct precise geolocation data for 155. “About two-thirds were not epidemiologically linked to the market,” he explained, which means that when the WHO mission did the questioning, these patients did not work at the market, did not visit the market, and did not have contact with anybody who shopped or worked at the market. Some lab leak proponents jumped onto that, arguing that the market was just an amplification event but cases were already spreading in Wuhan at the time.
“But there are two types of associations: epidemiological and geographical,” Mike continued. All the unlinked patients—sick people who were retrospectively identified by doctors as having COVID-19 with no market association—were not randomly distributed over Wuhan either. After having identified their place of residence, the unlinked patients seemed to live remarkably close to the Huanan market “compared to what you would expect by chance,” Mike elaborated. This was remarkable.
Their geographic relationship to the market held even when Mike started controlling for all possible alternative explanations, such as demographic data. Was it a very populous area? No. Was it age-related? Since “COVID-19 does not affect all age groups proportionally, you are more likely to end up in the hospital” when being elderly, he explained his reasoning. So maybe there were just a lot of old people in that particular part of town? Again, demographic data said no. No matter how he sliced the geographic and demographic data, the early cases were “ridiculously centered on the market.” He tried to put some numbers to show that the low likelihood of chance caused the clustering. “We also did something called kernel density estimates”—a statistical method to build a probability density function, the equivalent of a bullseye— “and ask what is the peak area with the highest density of cases?” The kernel density method is entirely agnostic of anything but the geospatial position of those patient cases on a two-dimensional map. The absolute bullseye of those cases, when overlaid to the Wuhan city map, gives a radius of a bit more than 300 meters. What is within that radius? The only thing it really includes is the Huanan market. Remarkable again.
He tried to make the point of this result clear. “You are not looking at where the market is in this picture.” He had calculated the density over a two-dimensional space. It was only after determining the bullseye that the 2D map was overlaid onto a city map of Wuhan. Where did the bullseye sit? Right on top of the Huanan market. “If you understand that analysis, that it has nothing to do with where the Huanan market is,” yet still hitting the market exactly, that result is just uncanny. Furthermore, Mike found that this “bullseye” is not just a statistical artifact coming from weighing patients with a known epidemiological association to the market; in fact, he removed all those linked cases and found that the unlinked patients—those that had no epidemiological link with the market—lived much closer to it than the linked cases. An incredible finding again.
I think this is worth emphasizing: the only people sick in December 2019 in Wuhan were those who either worked and shopped at the market (no matter how far away they lived from it) or those epidemiologically unlinked cases who lived in direct geographic proximity to it. All these early cases were picked up and diagnosed independently by doctors in seven independent Wuhan hospitals before the various Chinese CDCs or other authorities ever heard about an outbreak. The Huanan market was the unequivocal epicenter of the outbreak in Wuhan. This is where human-to-human transmission chains started to take off.
Any other study published since Mike’s has upheld his findings. Outside of cases somehow associated with the Huanan market, there were no other clusters, no indication of the virus spreading anywhere in the world. A retrospective study of over 34,000 pre-pandemic blood donors that the WHO mission had asked for (and was performed by Chinese scientists) also showed the same thing. No positive cases in September, October, November, or December of 2019. Very few people were infected, and the virus was just not widespread in Wuhan before December 2019. The only case cluster in December was the Huanan market. These findings also indirectly disproved the Trump State Department’s fabricated claim that three WIV workers had been hospitalized with COVID-19. If, indeed, three young WIV workers had been sick with COVID-19 in October of 2019 to the point of being hospitalized, as the State Department alleged, then hundreds of other people would have had to be infected as well, many of them hospitalized with severe disease. This is what epidemiology and demographics would predict, because the hospitalization rate for young people is multiple orders of magnitude less likely than elderly. There was no way the Huanan market was not involved in the outbreak.
So, what exactly was going on inside the market? Who could help him figure this out? Mike’s investigation into the early outbreak would bring him together with a colleague who had fiercely disagreed with his Science letter. Disagreed to a point that almost tainted their professional and personal relationship. For months, the two of them had worked together on a different project related to how the pandemic spread in America, but the origin question still loomed in the back of their minds. Both had pursued investigations into it after the WHO origin report came out. While Mike was focusing to understand the epidemiology and what happened to cases outside the market, Prof. Kristian Andersen had dived into a different facet of the outbreak, looking at what was going on inside the Huanan market. “The genomic data has always been inconsistent with the idea that it [the virus] was widespread by the time it was detected,” Kristian clarified, explaining why he did not think that the outbreak was going around in stealth mode for a long time. When Xiao Xiao and Zhao-Min Zhou from the Southwest China Wildlife Resources Conservation Laboratory published their paper on wildlife sales from 2017-2019 in the summer of 2021, Kristian was inspired to drill down through what information could be gathered and confirmed on activities within the market. How was the layout of the market? Where did the sick vendors sell their merchandise? What merchandise was sold? Where were live mammals held and butchered? What samples did Chinese scientists take?
One of the key pieces of evidence was a Chinese CDC report from January 22, 2020, which protocolled the collection of environmental samples from the Huanan market—this was George Gao’s team. That report, which included a table with what environmental samples had tested positive, had been made public in the summer of 2020 by Chinese newspapers, namely the South China Morning Post and, funnily enough, The Epoch Times. Early Chinese CDC updates, and Chinese newspaper articles had reported about the market sampling at the time before geopolitics and blame games made it all controversial. Kristian knew that some of these data existed throughout 2020, but they were all in Chinese, so he put them to the side. The pandemic demanded all of his attention, especially when the alpha variant exploded in the traumatic winter of 2020/2021, and he, along with his colleagues, became so occupied with the pandemic that he forgot. Months later, in the annexes of the WHO report, some of this data was published in a low-key fashion. While Kristian had seen the data in the WHO mission report, it wasn’t until a Twitter user running by the name of “Babar” (@babarlelephant) put up a website that they created with pictures of the Huanan market in a type of virtual visit click-through tour that Kristian was jolted into remembering that there were some data to be followed up on.
“Folks, just going back to the amazing resource from babar - can we get this table fully translated and overlaid with his map? It’s from that old Epoch Times article”
Kristian would write to his colleagues at the time. (They offered Babar authorship on the scientific paper they were working on, but Babar declined. Later, the paper would thank Babar for their contributions). With all the ingredients in place, excitement quickly took over.
“How could I have missed this?” he asked, raising his hands in response to the forgotten data. “All the details are in there. What stalls were positive, what stalls were negative… so I started looking at that.” He looped in Eddie Holmes again, who was still interested in looking deeper into the emergence and origins of the virus. Babar also put the photos Eddie Holmes had taken of the Huanan Market in 2014 online and geolocated them to a specific stall on the western side of the market.
In the meantime, Kristian had gotten serious about having these early Chinese media and CDC reports translated professionally. Finally, with the help of contemporary translations of the market sampling efforts from Chinese news reports, the WHO mission data annexes, and the visual support from photos and videos “Babar” had collected from social media and all over the internet to help them orientate virtually around the market, Kristian and Eddie made some important discoveries.
One of them was that the environmental samples within the market were not evenly distributed. There seemed to be a cluster in the corner of the western side of the market where multiple environmental samples tested positive for the virus. The corner happened to be where Chinese sources and contemporary evidence had identified wildlife stalls and merchants. Interestingly, no human case was associated with that particular stall. “I was like holy smokes! This cluster where we have these animals being sold, where they had reported on in 2020, right? This was the shop that Eddie visited,” Kristian recalled how suddenly the pieces fell together for him. A lot of virus positivity seemed to aggregate in this one small corner on the west side of the market—the same shop that Eddie Holmes visited the Huanan market in 2014 with his collaborator Zhang. Eddie’s first-hand experience and documentary evidence proved invaluable. “All this [virus] positivity stuff is like right on top of this shop that Eddie visited in 2014, and guess what—took photos of raccoon dogs.” The raccoon dogs, a species involved in the first SARS outbreak, were held in very specific cages. The early reports from the CDC, almost forgotten, indicated that environmental swabs of these very cages were positive, as well as two carts in the same stall. And then Eddie was like, “Yeah, if you look at the back of my photo, you can see two carts.” Kristian recalled their excitement. “And then there was this feather remover” that was found positive as well from environmental swaps. Why a feather remover? “Eddie was like, ‘Zoom into my photo and look at those raccoon dogs… what are they sitting on?’” They were sitting on top of cages that held some unspecified bird species. “You got to be kidding me,” Kristian exclaimed, his excitement still vivid.
Within that tiny corner of the market, where no human case was reported, not only were machines, carts, and cages testing positive for SARS-CoV-2, but samples taken from the open sewer under and the sewage downstream of that stall had environmental swaps that had tested SARS-CoV-2 positive. It all fits with the idea that not only had SARS-CoV-2-susceptible wildlife been held in this corner, but they might have contaminated their environment and what they came in contact with. Isn’t that highly relevant to follow up on?
I remembered what Peter Daszak and Marion Koopmans had told me about their WHO visit and how they came to a similar conclusion. These wild animals provided a direct link back to the bats. That they were never tested, nor even acknowledged as being there by Chinese authorities, was a direly missed opportunity. Chinese authorities aside, most Chinese scientists were not oblivious to what had likely happened here. George Gao’s team returned multiple times to take more samples from the Western side of the market, specifically that corner. In my interview with Shi Zhengli, I learned that she had also visited the Western side of the market in early February to take around 30 environmental swabs herself. At this time, it had been over a month since the outbreak and subsequent decontamination, so one would not expect to find any viral material anymore. Yet, surprisingly, she confirmed to me that five out of her thirty samples were still positive by PCR, albeit at such a low abundance and quality that sequencing those samples would have been impossible. These events certainly enforced the impression that the market, and that particular corner, was awash with viruses. This was not true for other places in the market.
The big irony is that the WHO mission had meticulously collected much of the sampling location data, which Kristian and Mike now carefully re-analyzed. However, because of pushback from Chinese authorities, these loaded inferences about wildlife trade were not allowed to be made and spelled out explicitly in the WHO report. Nor was the clustering around the market discussed, although the WHO members had created one figure. But most of the data was there, just waiting to be taken seriously by independent scientists if one cared to look into the evidence and not just the politically negotiated conclusions of the report. That is why Kristian had been utterly frustrated by Mike’s and Jesse Bloom’s letter to Science, to put it mildly. “The message around ‘we should keep an open mind, we should keep investigating,’ nobody disagrees with this,” Kristian explained. If that had been the only argument, he would have also signed the letter if given the opportunity. But that was not the real message, spirit, or impact of the letter. He objected to its grandstanding tone and many open insinuations, which were already prevalent in the media at the time. It implied that we know nothing about the origins, that we have no data to inform our opinion, that the WHO mission was a failure and tainted, all while grandstanding Western scientists were explicitly accusing Chinese scientists of lying and a lab-origin cover-up. All without evidence. He saw the Bloom et al. letter, instigated by Mike, as enforcing a false narrative that was not only unjust but was also putting scientists in danger, including himself.
Ever since his proximal origin paper, Kristian Andersen has been under severe harassment from the lab leak conspiratorial community. An escalation happened in 2021, when FOIA requests of Anthony Fauci showed how Kristian had first raised the alarm about the possibility of an unnatural virus, only to then change his mind with the emerging evidence. The conspiratorial fever pitch of 2021 media narratives contorted these events beyond recognition, leading to allegations of Kristian being paid off by Fauci to cover up a lab leak. Kristian understood sooner than most what it entails to become a target of conspiracy theories. The last thing he needed was his scientific colleagues and friends, who should know better, to pour gasoline on the fire with a poorly thought-out letter and bestow purely conspiratorial notions with a veneer of scientific legitimacy. Which, of course, happened in the most dramatic fashion. The fact that the contrarian non-expert Alina Chan was on that letter and indicative for much of its tone should have been a red flag in the first place. Mike stood behind what he saw as the letter’s main thrust—a call to keep an open mind on the origins until more data came in. A personal conflict.
It got worse when Jesse Bloom published a preprint under great media fanfare, alleging he had recovered “deleted sequences” of the early Wuhan outbreak that Chinese researchers had uploaded to the SRA sequence database but ominously requested to have deleted after. Kristian, who collaborated with Jesse and still does on other matters, had a run-in with Jesse about the preprint, which he saw as unsubstantiated, riddled with scientific errors to create a self-serving story, and unjustly accusatory towards innocent Chinese scientists. Chinese scientists are not a faceless prop of the state; they are human beings that Jesse was accusing based on nothing but his own faulty reasoning and biases. Nevertheless, Jesse pressed ahead with the allegations that would catapult him to fame and sharpen public criticism of China. Ultimately, his insinuations proved to be baseless and highly misleading. On top of that, Jesse deliberately took data that was published in a different format and used it for his own purposes. It polarized the debate even further, making the conflict between Kristian and Bloom et al. authors like Mike even bigger. It didn’t help that conspiracy theorists would insert themselves into the conversation, elevate the renegades as heroes, and use their words and actions to attack and discredit the embattled Kristian as dishonest.
Yet despite the polarized environment and personal obstacles, both Mike and Kristian had found something relevant, one outside the market, the other inside of it. Both scientists had a separate set of skills and knew that they could challenge each other critically if only they could overcome their ill will about the Science letter and subsequent media frenzy, which came with the predictable death threats towards Kristian personally. Not that any of this was Mike’s fault. Some lab leak activists love to decontextualize the words of one scientist to attack another, to instigate coordinated harassment and hate campaigns against their enemy, and this distorting effect is often hard to shake off.
“Professionally and personally, this was not a high point,” Mike said, admitting to how the Bloom letter he instigated had strained their working relationship. Scientists argue all the time about evidence, sometimes bitterly, so it is nothing new. “It’s always good to have that heterodoxy up front rather than just having the same voices in the room,” Kristian acknowledged. But the personal dimensions and media frenzy were harder to ignore. Scientists are also humans. “Kristian sent me an email, and the subject heading was ‘Compartmentalization’,” Mike explained, meaning to just decouple from all the personal feelings and focus on the science. Today, both of them were able to laugh about the pragmatic email. But that is exactly what they did at the time—put feelings aside and focus on facts. Kristian explained how they bridged their divide:
So, I think that’s where the good science comes in, where we disagree on some of these initial points, but we don’t disagree on the fact that this should be science-based. It should be focused on evidence, and we should just look at that.
“To come together and do this work” was also something Mike was proud of. “Whereas in a lot of other cases, it would have been like a flame war and the end of a relationship, and then this scientific work that we did would never have happened.”
Once the two scientists teamed up, with a promise to scientifically challenge the market origin hypothesis any way they knew how, they started recruiting from a wide array of talented scientists with varied expertise. Their goal was to systematically collect, analyze, discuss, model, and interpret all data that somehow stood in relation to the Huanan seafood market, as well as what epidemiology, phylogenetics, demography, geography, and statistics could tell them about the early outbreak. What does the totality of verified evidence tell us about what happened here?
This team of independent scientists was a force to be reckoned with. Especially the work of a talented graduate student named Jonathan Pekar, a computational biologist, and his supervisor Professor Joel Wertheim at the University of San Diego. They would offer crucial insights into how the early Wuhan outbreak likely unfolded. Jonathan specialized in creating epidemic simulations on his computer, essentially rerunning the outbreak virtually tens of thousands of times to look for repeating patterns.
The goal of these simulations is to test and learn crucial parameters about the outbreak dynamics, such as the rate of spread or the timing of the first infection that started the transmission chain. By starting with a wide array of possible parameters, the simulations produce a wide array of virtual epidemics that all unfold a bit differently from one another and are then compared to the observed real-patient data that was collected and verified by Chinese scientists, the WHO mission, and Mike Worobey. Those epidemic parameters that gave rise to simulations that matched the real data would give a good estimate of the conditions of the initial outbreak in Wuhan.
Doing this, Jonathan and his supervisor Joel found something quite surprising. Many of his epidemic simulations would burn out after infecting a few people, rather than take off and start a sustained outbreak. It turns out that the original SARS-CoV-2 virus that spilled over in Wuhan actually was not extraordinarily infectious (certainly compared to later variants such as alpha, delta, or omicron), meaning there was a decent chance it would have burned out by chance, unable to sustain human-to-human transmission. This is because pandemics are not just about the virus but also about the hosts. They also have a social dimension. One of the most crucial parameters in Jonathan’s simulation was the likelihood that an infected person would come into contact with other people, as well as how many people. In a remote village, where population density is sparse and a single infected person can only meet a handful of other people, a virus like SARS-CoV-2 would have burned out with an over 99% likelihood. But in a crowded city like Wuhan, where there is a high population density, the odds of the virus causing an outbreak increased dramatically, to around 30% if started by a single infected person. No matter if a lab leak or a spillover event caused the first infection, these were the rough odds of SARS-CoV-2 causing an outbreak in Wuhan, given the virus properties and environment. However, did that not imply that humanity had a 70% chance to dodge that particular bullet? Did we just get massively unlucky?
Unfortunately, it isn’t that simple. This is what the careful work of Mike, Kristian, Jonathan, Joel and around twenty other talented collaborators would make clear. I apologize to the coauthors and readers for not being able to go into the details of their individual contributions, which were often highly significant and can be found in the primary literature. Science is a highly collaborative endeavor, and no single book can do that reality proper justice.
There was one more fact that puzzled the researchers since the early days that needed to be cleared up for the outbreak data to make sense. From the very start of the first few hundred cases in Wuhan, there had been two separate virus lineages, named pragmatically lineage A and lineage B. They differed by two mutations from each other but seem to have spawned independent transmission chains, with lineage A accounting for around one-third of all early cases and lineage B for two-thirds. Lineage B would be the one that exploded into the world and give rise to subsequent variants such as alpha, delta, and omicron.
What was so odd about two lineages? Usually, whenever a new mutation happens, the phylogenetic tree of the virus gets an extra branch—a bifurcation on the tree that can potentially spawn a new lineage. Most of these branches are limited to a few people and die out. Others are successful and propagate forward, acquire more mutations, branch out again, and so on, making the family tree grow. While the genetic diversity between branches can be extremely high, the genetic diversity at the root of the tree, from which all subsequent branches spawn, is, by definition, zero. It is the starting point before the diversification of that particular tree. The most dramatic illustrative case example would be a super-spreading event, such as what happened in the Shincheonji Church of Jesus, Daegu, South Korea, in 2020. In that superspreading event, 5,200 cases would be traced back to a single infected woman. By analyzing the viral genomes of hundreds of infected patients at the time, researchers could reconstruct that all genetic diversification they observed started from one particular starting point: the woman in the church. She was the root of that outbreak, the center from which all other branches arose (in technical terms, this would be known as a “polytomy” or multifurcation of the genetic tree). The sum of genome sequencing of the genetic diversity of patients involved in an outbreak can usually be used to identify the root or starting point of an outbreak; this is what phylogenetics is often about.
Surprisingly to Jonathan, Joel, Mike, and Kristian, doing the same exercise for the early Wuhan patients showed a different picture. Phylogenetic analysis could not identify a single clear genomic root from where the diversity started. At first, because lineages A and B were just two mutations apart, researchers just presumed that there was an intermediate genome between the two or that one lineage branched out from the other early before being noticed. However, the particular branching pattern (polytomies) of lineage A and lineage B told a different story: It appeared that one genome was at the root of all diversity from lineage A cases, and another genome was at the root of all lineage B cases, while these root genomes were not directly connected but rather separate from each other. Somehow, the early patient cases in Wuhan were not branches from a shared root but belonged to two independent trees. How was this possible? Did scientists miss swaths of early patients that held an intermediate genome connecting the trees? That seemed unlikely, given the complementary data that Mike unearthed. So, where or how did this early lineage split happen? And why did these two large polytomies suddenly explode almost simultaneously?
Jonathan Pekar, Joel Wertheim, Kristian Andersen, and Mike Worobey explored plausible epidemic scenarios with computational methods that would recreate this particular early diversity pattern, and they ended up with a surprising discovery: SARS-CoV-2 most likely spilled over multiple times into humans.
The reason why there were two early lineages has to do with the introduction of SARS-CoV-2 into the human population not once, but at least twice, from an animal reservoir that had diversified the virus by these two lineage-defining mutations before it spilled over. The moment an animal infected a human, the virus took off and started the human-to-human transmission chain, diversifying from the genome that spilled over, which was lineage A root or lineage B root, respectively. That’s why the lineages were associated with two separate large polytomies. Their calculations based on lineage-specific epidemic dynamics also showed that the timing of these spillovers happened close to each other but not at the same time. They were at least a week apart from each other, with lineage B spilling over first sometime in late November. This scenario best explains the totality of phylogenetic data, the progression of cases, and the family trees of the viruses we observe.
The multiple spillover explanation also solved another important riddle that had tripped up scientists previously: how come lineage B is the predominant version when lineage A is ancestral to it? The epidemiology showed that lineage B cases came first, but genetics indicates that lineage A was closer to SARS-CoV-2 relatives found in bats. If it was a single spillover event or lab infection, and lineage B simply arose from lineage A, how come lineage B got a head start and seemed to have caused sickness earlier, further, and wider than lineage A?
The observed genetic, timing, and patient case data are very hard to explain by proposing a single introduction event. However, the multiple spillover scenario neatly solves this conundrum. All it required was a pool of sick and infected intermediate animal hosts at the market, with a certain amount of viral diversity in circulation. If that condition was given, then lineage B just happened to make the successful jump into humans first, and lineage A spilled over a week or so later.
I could not help but be reminded of what Linfa Wang and Peter Daszak told me about the Nipah virus outbreak in Malaysia. They also first believed that it was a single freak spillover event from bats to pigs, but later (once genome sequencing technologies were invented and became much cheaper), they found out that the genetic diversity of the viral genomes also suggested multiple independent spillovers from bats to pigs. It makes sense; with a single introduction, the virus would have run through the farm and likely burned out. Instead, the immunity from the mother sows, in combination with the rapid piglet breeding and weaning practices, meant that little piglets would become susceptible to new spillovers at this risky bat-pig interface the moment they were about to be sold. That is how a few pigs ended up propagating Nipah forward to cause the sustained outbreak. Retrospectively, outbreak investigations always only get to observe the survivors, those viral lineages and transmission chains that cause outbreaks, never those spillover infections that burn out by themselves after a few days in a new host and do not successfully transmit to others.
Back to the Huanan market. By necessity, the observation of two distinct lineages—two survivors—provides a lower range for how many spillovers happened at the market. At least two spillovers were required, but it’s likely there were more that were not observed because they burned out. Given how quickly the cases exploded from the market, epidemic simulations are consistent with up to almost two dozen independent spillover events shared between A and B. Around three-quarters of them were always expected to die out within the infected human hosts, as the characteristics of the virus and population dynamics predicted. Over 99% die out in a village; over 70% die out in Wuhan. Yet we observed that not only one but two different lineages successfully established themselves in the human population. Either we were dramatically unlucky with two freak events, or there were just a lot of spillover events we did not observe because they burned out. In their totality, these multiple spillovers likely made the outbreak of SARS-CoV-2 in the market pretty inevitable. Once infected animals shedding the virus were brought into proximity with many, many immune-naive human hosts, once this very risky animal-human interface was put in place, we had sealed our fate.
Jonathan, Mike, and Kristian’s computational modeling work would come to support the latter notion. Here is the relevant statement from the scientific paper published in Science (2022):
The extinction rate of our simulated epidemics (simulations that did not produce self-sustaining transmission chains) indicate that there were likely multiple failed introductions of SARS-CoV-2. Similar to our previous findings, 77.8% of simulated epidemics went extinct. These failed introductions produced a mean of 2.06 infections and 0.10 hospitalizations; hence, failed introductions could easily go unnoticed. If we treat each SARS-CoV-2 introduction, failed or successful, as a Bernoulli trial and simulate introductions until we see two successful introductions, we estimate that eight (95% HPD, 2 to 23) introductions led to the establishment of both lineage A and B in humans.
In other words, we were not unlucky victims of one or two single freak spillover events; rather, we ignored a very risky animal-human interphase at that particular wet market in Wuhan, which allowed for SARS-CoV-2, already circulating in animals, to spill over repeatedly until two of those human infections took off to take humanity by storm.
The multiple spillover theory explained the available data pretty perfectly. But could it do more than that? What implications does that theory entail? With this multiple spillover model in mind, the team around Mike and Kristian could ask some more interesting questions about the Huanan market. For example, if multiple spillovers from infected animals were true, then by necessity, some other testable data would need to line up. For example, one would have to show that sick patients from both lineages A and B independently centered around the market. While the earliest patients might have been retrospectively identified and tested for PCR, full genome sequencing was scarce. Only two lineage A genomes had been sequenced, and neither patient reported any association with the market. But when Mike looked at their geographical location, both were significantly closer to the market than expected by chance, with one spending five days in a hotel next to the market before symptom onset. Certainly supportive of the multiple spillover theory.
Then came another breakthrough from a surprising source. George Gao’s team from the Chinese CDC finally published a preprint about their analysis of environmental samples from the Huanan market. They found both lineage B cases and a single lineage A genome at the market, which contradicted the common notion that the market was a mere amplifier or super-spreader event. It confirmed Mike and Kristian’s suspicions. “Our analysis predicted that both lineages would be at the market,” Kristian explained. And behold, their prediction was correct. With the identification of the third lineage A genome, both lineages were now spatially centered around the Huanan market and radiated outwards into Wuhan from there. Both lineages were now found and confirmed inside the market by environmental swaps. As Mike explained:
It’s kind of like me saying I’m going to shuffle this deck twice, and each time I’m going to pull out an ace of spades… It’s really unlikely to do it once; it’s ridiculously unlikely to do it twice, and so that's where we were with the lineage A. That’s important because it destroys the idea that the Huanan market was just simply a super spreader event.
These results also contradict one of the last naive cop-outs from lab leak advocates, who have long been on the defensive about the market: the idea that an infected lab worker might have brought the virus into the market and caused an outbreak there, but not anywhere else in the city. This was a magical assumption, given how far the lab was away from the market and about 10,000 other more likely places in Wuhan where an outbreak cluster would happen than the semi-open Huanan market. Yet it was challenging to disprove because it is theoretically possible that an infected lab worker somehow went to that market by chance and started an outbreak there. But if the virus came from a lab infection, why would it show up at the Huanan market first and nowhere else? One of only four places in Wuhan that actually sold wild animals known to be highly susceptible to SARS-related viruses? Lab leak advocates had no scientific explanation for the Huanan market association even before lineage A was confirmed to be at the market.
With both lineages confirmed inside the market, any idea that SARS-CoV-2 could somehow have been carried in twice, causing a low-probability sustained outbreak twice, in the same market but nowhere else, was rather absurd. “You don’t get a super-spreader event and have two lineages associated with that,” Mike concluded. On top of that, it would require that both lineages come from two different lab workers who were independently infected by slightly different viruses a week apart from each other, and both decided to go to the Huanan market far away from their work or place of residence on the other side of the Yangtze River. All while infecting absolutely nobody else at work or in their personal life yet still being so highly infectious as to set a wildlife market ablaze twice in a single visit.
“You look at all the evidence; it points straight to a market in Wuhan,” Kristian Andersen was very clear about this. “It is not a single line of evidence we are using here,” he elaborated on their work. Indeed, taking the multiple overlapping lines of epidemiological, phylogenetic, geographic, forensic, and statistical evidence together, a single picture emerges: Infected animals were brought to the market, likely kept there for weeks in unsanitary conditions, giving opportunity after opportunity for circulating SARS-CoV-2 lineages to spill over into immune-naive human hosts visiting the market. While many of these spillovers likely burned out, at least two of them started sustained transmission chains about a week apart in late November 2019 that would radiate outward from the Huanan market into the city and from the city into the world. “I think it is important when we talk about what are the most likely scenarios; you gotta take all the evidence,” Kristian emphasized again. The power of the scientific method comes from triangulation—finding that multiple different experiments, methods, and scientists all converge on a single hypothesis, theory, or conclusion that can explain all different types of evidence parsimoniously and coherently without internal contradictions. Ideally, a scientific theory also has predictive power that can be experimentally tested, such as the supposition that if those environmental samples taken for the market were ever to be tested, lineage A would most likely be found.
The market theory would also predict that SARS-CoV-2-susceptible animals must have been present in the western corner of the Huanan market in November 2019. Something Chinese authorities vehemently denied and that neither photographic evidence from random social media accounts in 2019 (or outdated photos from Eddie Holmes in 2014) nor the Xiao Xiao general survey could conclusively prove for that particular shop. Yet, in due time, unequivocal genetic evidence would be found to substantiate that prediction.
No matter how one looked at it, all the scientific evidence to date told a very one-sided story about where SARS-CoV-2 came from. It pointed away from the lab and toward the Huanan market.
For Mike and Kristian, their two dozen international coauthors, and large swaths of the scientific community, the picture arising from the totality of the evidence on key issues surrounding the origins of SC2 became pretty clear. The lab leak idea was a mirage, wholly unsupported speculation contradicted by evidence, running up against a compelling zoonotic spillover explanation with strong, albeit not perfect, evidentiary support. A naturally evolved virus caused the first known outbreak at a wet market. However, this recognition does not mean that there are no more questions to be asked or no more mysteries to be solved about where the virus came from. “This is not the end of understanding the origins of COVID-19. This is the beginning; it gives us the necessary focus that we need, which is that market in Wuhan,” Kristian added. What animal species were physically present at the market in late November? Were they really infected? Who brought the animals into the market, and from where? Where were the Wuhan animals exposed to a bat virus that could infect them? Did the wildlife traders have antibodies against COVID-19? Where’s the original bat reservoir? How long did the virus circulate in intermediate animal populations before it emerged in humans? We need to follow up on these questions if we want to prevent SARS-CoV-3.
Mike, as best I can tell, hoped that their scientific breakthrough would infuse the public discourse around the topic with some much-needed reality check and lead to follow-up investigations into the wildlife trade, both in China and abroad. While no amount of scientific evidence was likely to change the minds of entrenched lab leak believers, there was a reasonable expectation that journalists, elites, and mainstream media would do their part in correcting the record. Indeed, the few remaining science journalists who had not yet lost their jobs in the new information ecosystem (which increasingly values commentary and op-ed desks over science desks) dutifully reported about the scientific pre-prints and later papers that would be published in the renowned journal Science after peer review. CNN, the BBC, NPR, the Los Angeles Times, The Washington Post, and The New York Times all had a story running. So did National Geographic and Scientific American. The message about a scientific breakthrough in the origin case was out.
In ordinary times, this should have been enough to bring lab leak speculations back on solid evidentiary ground. But I am afraid we do not live in ordinary times anymore.
§
The market origin papers were followed by an explosion of activity from lab leak proponents to denounce the independent researchers, to deny any type of breakthrough happened, and to deceive citizens about the new evidence. Many revenge plans were set into action, not dissimilar to what happened in the Trump orbit after “proximal origin” was published in 2020. The only difference was that this time, loud and powerful voices from both the left and right—crowds, journalists, and politicians—all seemed to do their damnedest to smear the work, ethics, and character of independent scientists in the process. A relentless campaign against the mainstream outlets that reported about the papers would also ensue, advocating to retract or correct their reporting or give lab leak innuendo equal space to peer-reviewed science. Many such efforts would eventually succeed. As always, it is impossible to map out all the plots, ploys, and players, but if I had to put my finger on one pulse-giving narrative, it would be a brilliantly crafted but extremely misleading article by Katherine Eban. A special correspondent for Vanity Fair, a magazine covering fashion, popular culture, and current affairs, she became in my opinion one of the most fervent crusaders for the lab leak narrative, providing motivated rationalizations to justify the feelings of elites in both media and politics alike.
Katherine is an investigative journalist with a very specific beat and style. Years prior to the pandemic, she had written a book to supposedly highlight bad practices and dirty secrets in the pharmaceutical manufacturing industry, relying heavily on selective quoting from leaked documents and utilizing quotes from real and self-proclaimed whistleblowers to tell the – in my opinion – most sensationalist and damning story possible. If one thing can be said about Katherine, it is that she is a truly gifted storyteller who always knows how to create heroes and villains from ordinary people and extrapolate their conflicts into world-sweeping battles of global relevance and moral saliency. In her corporate crime thriller Bottle of Lies, Katherine tells the story of Ranbaxy, an Indian generic drug manufacturing company that was found liable by the FDA for producing subpar medicines and fined $500 million USD in 2013. Her book, leaning heavily on what appears to be real fraud on the part of Ranbaxy and some oversight failure by the FDA, did however not stop at one company but framed the whole overseas generic drug manufacturing industry, from India to China to African Nations, as fundamentally corrupt and untrustworthy. In her book-selling pitch, her investigation supposedly “reveals how the world’s greatest public health innovation [generic drug manufacturing] has become one of its most astonishing swindles.”
I have my fair share of criticisms of pharmaceutical companies and capitalistic business practices, but painting a whole life-saving overseas industry with an overall too broad and sensationalist brush as a mere “swindle” is quite dishonest for my taste. There is no proportionality. The reality is that generic drug manufacturing serves hundreds of millions of patients all around the world, multiple layers of regulation are in place, and the vast majority of generic drugs are entirely safe and effective no matter where they are produced. That is the big picture. So, while alleged or real cases of negligence or misconduct unquestionably happen, they make for rather thrilling books but are hardly representative anecdotes. Kathrine’s broad, sensationalist framing, however, hit a cultural moment. “A new book argues that generics are poisoning us,” The New York Times would pick up her tale uncritically. Others would follow. The fearmongering about “overseas” drug manufacturing was convenient for elite newspapers at a time when geopolitical tensions increased. Such narratives about “foreign-manufactured medicines are poisoning us” are bound to make waves in popular discourse, especially with Trump’s trade war against China and public demands for companies to bring back American manufacturing jobs through onshoring supply chains and related industries. As an author, I tend to see that mainstream success in books is often reliant on hitting the sweet spot between popular appeal and satisfying the agenda of the powerful.
Kathrine’s book became a New York Times bestseller and was quite impactful because it convincingly portrayed an overseas manufacturing industry as “a Wild West environment, where being first mattered more than getting it right.” Who doesn’t like to read a well-crafted thriller about the brave underdogs unearthing corporate corruption and governmental failings? I certainly do, and I am rooting for the underdogs every time. Those are the stories and tropes we readers want to engage with; nuance and context are secondary at best. Especially emotive narratives about foreigners striving to surpass us technologically with their recklessness (or lawlessness) leading towards mortal peril for citizens hit a deep tribal fear in us. It would not be a surprise to me that when Katherine Eban started hearing the same tales about allegedly reckless and unregulated “gain-of-function research” overseas, she potentially saw another opportunity to write a sweeping thriller.
Correspondingly, she made her first foray into the origins topic in 2021 by mainstreaming various conspiracy theorists and political actors deeply involved in lab leak mythology. She portrayed shady Trump operatives pushing the bioweapon myth within the State Department as underdogs against the National Security establishment and conspiratorial activists on social media as either neglected whistleblowers or brave scientific renegades. These were her heroes, lavished with generous embellishments to enhance their credibility or abilities while downplaying or outright withholding their flaws, political entanglements, and activist nature from readers.
In contrast, she was decidedly less generous about describing her villains, deliberately constructing dubious story arcs from cherry-picked documents, out-of-context quotes, and poignant insinuations from their adversaries. For example, she painted the warm and candid Shi Zhengli as compromised and untrustworthy for the pure fact that she was a Chinese scientist in China. Katherine would also rehash just about every major story trope for the lab leak myth: the alleged biosafety issues, the supposedly sick workers at the institute; the imagined hiding of the Mojiang miner connection when Zhengli published RaTG13; Alina Chan’s flawed preprint about “pre-adaptation” that never made it past peer review; the “risky gain-of-function research” Zhengli allegedly conducted with Ralph Baric; USRTK’s FOIA’d private messages; the Lancet letter that Peter supposedly crafted to suppress any inquiry into the lab leak theory;… on and on her 12,000-word article goes. Her story needed Zhengli, as well as people in contact with her, to be the villains—researchers who lie about many things either because they were coerced by or involved with foreign governments. The more mendacious her villains were, the brighter Katherine’s heroes would shine. In my opinion, she would utilize any and all misleading interpretations and one-sided comments from conspiracy theorists and motivated actors to craft the most sensationalist and thrilling story about how the lab leak underdogs had triumphed over the suppression of the lab leak hypothesis. In her retelling, it was allegedly suppressed by conflicted scientists and shadowy government bureaucrats. That was in 2021, and she was just getting started.
When the market papers driven by Mike and Kristian received their first media coverage at the end of February 2022, her “sources,” that seemingly consisted largely of conspiracy theorists, activists, and power holders, were outraged about the results. Science had just destroyed their favorite narrative. They were incapable of challenging the meticulous papers on scientific merit, and the wide media coverage made it so they could not deny these papers existed. In this conflict with scientific reality, many lab leak believers would transition from being advocates for investigating a potential lab leak origin into outright science denial. Rather than simply changing their mind with the evidence, they opted instead to attack, smear, and discredit the independent scientists in the public’s eye. All they needed was a good counter-narrative to discredit the market papers. Something that would allow the world to ignore facts and focus on the messengers instead.
“When I go out to report, I am not looking for expert voices; I am looking for sources,” Katherine Eban would explain in a panel discussion about her investigative approach to scientific controversies. It seems, from her own description, that she did not pay attention to scientific evidence or bother to understand contested science to differentiate between competing expert opinions. In my subjective experience, she also appeared to be very proud of that, as if scientific literacy on the issue would bestow a conflict of interest on science reporters that would color their judgment. In the panel discussion, she would allege that science journalists were conflicted by relying on experts as sources, and that would disqualify them from being critical in their reporting about scientific controversies. In contrast, her investigative approach focused on sniffing out human intrigues and interpersonal conflicts through a mix of contradictory testimonies, speculations about motivations, and human idiosyncrasies. The human angle. On that part, there was no shortage of conspiracy theorists and motivated actors who could provide ample ammunition to Katherine Eban in the form of insinuations, innuendos, and opinions against the authors of the market origin papers, even some of the science journalists who covered them.
From these building blocks of human intrigue, the gifted storyteller could, in my opinion, fabricate a larger narrative that would not only implicate Peter Daszak and Shi Zhengli in a supposed origin cover-up, but basically all of virology, the NIH, and a very prominent public health scientist, somebody who US Republicans have long sought to hang in the court of public opinion. About a month after the market papers made news, Katherine had her next thriller ready: “This Shouldn’t Happen”: Inside the Virus-Hunting Nonprofit at the Center of the Lab-Leak Controversy.
The counteroffensive began.
Her piece started with what I would classify as a maliciously misrepresented anecdote about a conflict between Kristian Andersen and Jesse Bloom. Since 2021, Jesse has gained a large media presence based on his contrarian stance and what I consider a willful falsehood about Chinese researchers having deleted pertinent sequences that he recovered under much media fanfare. In reality, the Chinese authors did nothing wrong. It was later found that Jesse manually deleted the sampling date of “30th January” and replaced it with “early in epidemic” to create the appearance of relevance to the origin question when there is none. Kristian had challenged Jesse, with whom he collaborated on other projects, and said that Jesse’s accusations against Chinese authors were baseless. This was the root of their conflict.
Enter Katherine Eban, who crafted the “bespectacled, boyish-looking 43-year-old” lab leak proponent, whom her article described as “the most ethical scientist I know,” into the underdog trying to get the word about Chinese duplicity and cover-ups out into the world. In contrast, Kristian was painted as his irascible detractor, attempting to silence Jesse, all while also in cahoots with Anthony Fauci somehow. Why this odd construction? She would let the reader find out soon enough.
“Fauci and a small group of scientists, including Andersen and Garry, worked to enshrine the natural-origin theory during confidential discussions in early February 2020,” she would state, referring to the February 1, 2020, conference call that Jeremy Farrar arranged after Kristian and Eddie Holmes began raising suspicions about the viral genome. First, notice her use of the evocative word “enshrine,” ascribing a malicious agenda to the meeting. Second, none of that happened; it was a completely fluid situation, as we have read about before. After Katherine set the tone, she pressed on to describe that these events were mere examples of a “wagon-circling” that “reflected a siege mentality at the NIH” when it came to the origin topic. Again, this was baseless best I could tell. But in my opinion, Katherine did what she knew best, creating heroes and villains through subtle insinuations, word choices, and framing of events, breaking context and truth when necessary to advance her story.
Her construction worked well. Readers were wondering why there was a supposed “siege mentality at the NIH?” What were they defending? Katherine was ready to let them find out. Because in her thrillers, the cover-up conspiracy seemingly must go all the way to the top.
“Of all those high-level people, almost no one ranked as high as Fauci, a scientific kingmaker who dispensed billions in grant money each year,” Kathrine would write, completely mischaracterizing the role of the former agency head as if he personally dispensed funds rather than the scientific committees at NIAID assessing the grants researchers submit to them. But she needed this inappropriate “kingmaker” framing to advance her thriller’s main villain plot: How Peter Daszak was supposedly circling Dr. Fauci to get grant money for his nonprofit that would ultimately be instrumental in giving the Chinese money and knowledge to create SARS-CoV-2 and cause the pandemic. By itself, I found this insinuation not only naive but also pretty paternalistic, if not outright Sinophobic, as if Chinese researchers were incompetent and only able to do great or terrible things with an American mastermind behind them. Then, Kathrine laid out her case for blame. Dr. Fauci, as head of the NIAID, was supposedly not upholding his oversight responsibilities. Peter Daszak was cast as a ruthless mercantile researcher for whom being first mattered more than doing it right. Shi Zhengli continued to be the dishonest foreigner operating in a regulatory Wild West, possibly in cahoots with the nefarious Chinese state and military. How could the reader take anything else away from this forced constellation, but that disaster was about to strike? Readers inclined to believe that a lab leak happened were fully on her side, but Kathrine found a way to tune the emotional force of her narrative to up the stakes even further.
Because this alleged grant-outsourcing setup to fund risky research between Fauci, Daszak, and Shi that Katherine Eban constructed was supposedly not just a single incident, she assured her readers. No. As her article’s skillful framing alleged, these types of setups around grant money are characteristic of wider problems in virology and reckless gain-of-function research. It’s about power and corruption, her article alleged not so subtly in my opinion, and the whole scientific industry was supposedly the problem here, according to its tenor. That is why seemingly independent scientists are all “circling the wagon” around the NIH and its “kingmaker,” Dr. Fauci (who only headed the NIAID, just one of the 27 institutes under the NIH).
Now Katherine could take us back to Mike, Kristian, and their market origin papers that made the news recently. Katherine would write, “Worobey, Andersen, Garry, and their 15 coauthors rushed their preprints into the public domain”, ostensibly implying in my opinion that this was to protect the “king” who dispenses grant money to them or to obstruct the public from finding out how the scientific gain-of-function industry has worked unregulated for years. Notice the framing again by using the phrase “rushed… into the public domain,” implying a hidden agenda or agency. Without ever arguing over the scientific analysis or evidence contained in the papers, Kathrine had created a compelling story framing that implied the market papers were a product of biased authors who had something to gain by pointing the finger away from the lab. Again, a casual reader might be excused to take from this a conclusion that either implied conflicts of interest or, worse, scientific fraud on behalf of the market authors. Too deepen this conclusion, Kathrine’s article also highlighted Kristian’s involvement in “proximal origin” using a “just asking questions” framing technique:
Why top scientists linked arms to tamp down public speculation about a lab leak—even when their emails, revealed via FOIA requests and congressional review, suggest they held similar concerns—remains unclear. Was it simply because their views shifted in favor of a natural origin? Could it have been to protect science from the ravings of conspiracy theorists? Or to protect against a revelation that could prove fatal to certain risky research that they deem indispensable? Or to protect vast streams of grant money from political interference or government regulation?
For a storyteller, the purpose of these types of questions is to frame emotions, not give answers. Eban’s thriller tale ends predictably, with a salient quote from a biased source that expresses a strong opinion she wants the reader to adopt as the take-away message: “The group of scientists pushing the claim of natural origin, he says, ‘want to show that virology is not responsible [for causing the pandemic]. That is their agenda.’”
The moral implications of Katherine Eban’s thriller were sweepingly broad and on the nose. Just as with “generic drugs were poisoning us,” her—in my opinion —carefully constructed narrative hit a checkbox with the elites in media and politics. It gave permission to discredit anyone and everyone—not just Mike, Kristian, and colleagues—whose research would end up supporting a zoonotic origin. Katherine’s narrative, supported by accolades from conspiratorial influencers and crowds, made those in power believe that all virologists, in general, must have an agenda and be conflicted about the origin topic. Certainly, the market paper authors around Mike and Kristian were not to be trusted. This is, of course, extremely dishonest and ignores the reality of how Mike Worobey fell into the investigation and how, by trying to disprove the market origin hypothesis, he and two dozen independent coauthors uncovered evidence that would end up supporting a zoonotic spillover much more than a research-related accident.
“If it had been otherwise, I would have published a different perspective in Science, and we would have published a couple of papers showing … that it was a lab leak,” Mike stated matter-of-factly. If that had been the outcome of their research, he believed he would have gotten away looking much better. He had been the animating force behind the Science letter that contributed to the vibe shift. He could have been the renegade, the maverick underdog, sitting in his basement for months tracing all those early cases to prove the WHO mission assessment wrong. He certainly would have saved himself a lot of harassment and trouble afterward. Alas, the evidence fell the other way, and conspiracy theorists have had a special ire for him ever since—a betrayer. Staying true to science can be inconvenient like that.
Unfortunately, Katherine Eban’s article was not an outlier; rather, it served as the opening shot of the hunting season on the market paper authors. Alina Chan and Matt Ridley were not far behind, penning multiple opinion pieces for various mainstream outlets to discredit them. The cottage industry of conspiracy theorists around DRASTIC would post a barrage of pseudoscientific criticisms, none of which held any water, but collectively, they polluted the clarity of the scientific findings in the public domain; as if Twitter threads and blog posts could debunk peer-reviewed scientific articles. With Dr. Fauci as a potential trophy at the end of their crusade, soon, Kristian Andersen would be subpoenaed by Republicans, asked to come in front of US Congress together with Robert Garry (which they did voluntarily), and later smeared by Republican representatives and their client propagandists. The political witch hunts against public health scientists, virologists, and vaccine researchers had started, and the market papers put a target on scientists back.
Quality newspapers that had previously covered the market papers often buckled under mounting political and popular pressure too. Fear of losing popularity with readers, perceived legitimacy, and access to elites prompted many editorial teams in big newspapers to run op-eds to counter the reporting of their own science journalists. They platformed contrarian scientists and fringe figures to give their opinions versus emerging facts, creating the impression that there is a vivid dispute, “alternative science” and high uncertainty among experts. They framed the origin question as a conflict between two warring factions, at best equally likely to be right on the issue.
But at the same time, discarding the lab leak myth as baseless became a sin, punishable by character assassination and potential political persecution. An asymmetry where every garbage argument for the lab leak narrative was treated as newsworthy, and any counterpoint treated as possible conflict of interest dictated media coverage. The reality was that an overwhelming majority of scientists felt the body of evidence for zoonosis was strong, and any evidence of a lab leak was lacking. Yet in the press, remaining uncertainties not addressed or addressable by the market origin papers were played up, amplified, and repeated ceaselessly. Both-sidesism, false equivalency reporting, and “teaching the controversy” took up the majority of news coverage.
The evidence is amongst the best we have for any emerging virus... What other part of science do you know where we say there’s some uncertainty therefore both hypotheses or all hypotheses are 50/50? Do we do it with creationism versus evolution? Do we do it with climate science? … That is the most unscientific thing you could communicate to the public.
Mike was outraged and disillusioned. He was not alone; most scientists with a front-row seat to the spectacle became overwhelmed and increasingly lost trust in government, the free press, and the chattering classes. Many retreated from social media and mainstream media alike, unable to shout against the tornado yet in constant danger of being its next victim for speaking up. The rift between scientific knowledge and societal knowledge on this topic seemingly became insurmountable. Those scientists who were most knowledgeable about the topic had been tainted and could not get the evidence out to the public anymore. Their scientific breakthrough had been contained. Most citizens will never get to hear their story, the conflicts and obstacles that stood between them, and the excitement when the clues they unearthed started forming a coherent picture.
Which is tragically unjust for the scientists who just did their jobs, but even worse for society. Independent science is important because the true power of the scientific method is unleashed when multiple different lines of evidence, created by different methods and by different people in different nations, all converge on one hypothesis. By 2022, various viral discovery programs, bat researchers like Alice Hughes, and the scientists around Shi Zhengli, Linfa Wang, and Peter Daszak had collected enough data to show that only nature’s neglected gain-of-function laboratory, not genetic engineers, could bring about SARS-CoV-2’s intricate mosaic genome. This has been the work of hundreds of scientists for more than a decade.
On the other side of the world, dozens of independent scientists, spearheaded by the efforts of Mike and Kristian and their international collaborators, established the Huanan market as the unequivocal epicenter of the first outbreak in Wuhan. Chinese scientists created a lot of good data despite the obstacles, often working behind the scenes to get the word out to their international collaborators at great personal risk. The WHO mission experts also added crucial insights about what happened in Wuhan, as did the rest of the scientific community that worked on molecular virology, epidemiology, or viral evolution to contribute knowledge about the nature of this virus. Any lab or research involvement, especially various ideas that the virus was created or heated up at WIV, was now easily refuted, contradicted, or made highly implausible by the available scientific evidence in 2022. Like with SARS-CoV-1, all carefully acquired and analyzed evidence pointed to the biodiverse Karst region of Southern China and Southeast Asia, with the wildlife industry possibly serving as a conduit to create the right circumstances for a natural virus to cause a sustained outbreak in a Chinese megacity.
Scientifically, the situation was much clearer than for most outbreak investigations. This virus had nothing to do with gain-of-function research or any research labs. It took more than two years for the evidentiary basis to be established and for science and scientists to converge on actionable certainty around the topic.
So why wouldn’t the world take this as an answer? Why would establishment media continue with false equivalency reporting, platforming media-hungry contrarians, and ignoring scientific evidence while running character assassination campaigns against independent researchers? Why do false and mutually contradictory myths about a man-made virus, a gain-of-function flask monster, still persist today? Why are baseless speculations and fact-free innuendo still allowed to distort our politics, sabotage our pandemic prevention efforts, and put biomedical science as a whole under pressure?
At this point, I became intrigued as a science communicator. A chasm was opening in front of me in real-time. Where do our modern rifts between science and society come from in the first place? Were we always so divided, or did something change?
My suspicion was that our current conundrums and crisis of trust in science emerged from another neglected biodiverse ecosystem. One that increasingly exerted a malicious force on our public perception of the pandemic that matched, if not surpassed, the biological virus itself in emotional energy and societal friction.
During the pandemic, this neglected but dangerous ecosystem grew in power and now aims to rewrite the history and reality of how the virus came to be and who society ought to punish for creating it. Using a more intriguing narrative full of heroes and villains, outrage, and hope, the myth of a man-made virus that leaked from a lab became much grander than the mundane reality of multiple zoonotic spillover events at a wildlife market. And crowds were all too eager to buy into it.
Unfortunately, as we should have learned from history, such periods of myth-making and grand narratives always bring forth leaders, sages, and prophets who seek to shape, subjugate, and weaponize them to impress their political worldview onto others. Their victory lap over science and society would have its roots in 2021, with the release of a rejected grant proposal Peter Daszak had sent to the US Department of Defense’s Advanced Research Projects Agency, DARPA for short.
His life would never be the same.
Adapted from Lab Leak Fever: The COVID-19 Origin Theory that Sabotaged Science and Society by Philipp Markolin.
Copyright © 2025 by Philipp Markolin. All rights reserved.
Continue reading chapter 9 here.
Note: If you want to download, print-out, share or otherwise collect this chapter, run it through an LLM or just store for record keeping; here is a high-quality pdf version as well:
Find more background info, chapter footnotes and video interviews at www.lab-leak-fever.com. If you want a physical copy of the book, kindle ebook, or support my work, you can buy directly on Amazon. (for non-US readers, please check your regional Amazon such as amazon.br or amazon.in as ebook prices may differ dramatically)
How did you feel about this chapter? Please let me know any feedback or comment below. Make sure to share.
