WHO’s report is it, anyway? The murky origins of SARS-CoV-2.

A long-anticipated report from the World Health Organization leaves more questions than answers. The Chinese government’s lack of transparency isn’t helping.

Today, the WHO published a long-awaited report about the origins of COVID-19. It’s the culmination of a months-long investigation. 

In 123 pages, the report outlines four hypothetical origin stories investigators rate from “likely to very likely” to “extremely unlikely.” I’ll discuss each of them in turn, and offer my own rating scale from “plausible” to “extremely implausible”—as well as reflect on what the report tells us about how global geopolitics shapes the interpretation of science around public health. 

The “bat” hypothesis: direct transfer from a zoonotic reservoir host.

Early on in the pandemic, the narrative about SARS-CoV-2, the novel coronavirus, was that it had emerged out of the Huanan seafood market in Wuhan—from a bat that passed it directly on to a human in the stifling confines of the market. At least that was the Chinese government’s official narrative. 

The explanation does have its basis in some plausible observations. First, there’s precedent for zoonotic diseases to emerge in wet markets—where people are routinely exposed to non-domesticated wildlife. Second, the virus with the closest known genetic sequence to SARS-CoV-2—RaTG13 with 96.2% homology—is found in bats native to China.

Early in January of 2020, the Chinese government shut down the market and had it sanitized before any proper samples could be collected. And by February of 2020, it was clear that the first cases of COVID-19 likely had no connection to the Huanan market. The Chinese government itself gave up on this hypothesis by May. The market itself didn’t sell bats. And though the RaTg13 bat coronavirus shares 96% homology, it’s still decades apart in evolutionary time—with no phylogenetic “smoking gun” that bridges the distance between them.

The WHO report puts the probability of direct transfer from a zoonotic host such as a bat or pangolin as “possible-to-likely.” 

Though the evidence regarding how this might have happened remains thin, the hypothesis remains plausible. 


Hypothesis: Indirect transmission through an “intermediate host.”

Along with the “bat” hypothesis, the report advances a related hypothesis—that rather than direct transmission from one of these animals, the virus jumped into humans through an “intermediate host,” a densely-farmed domesticated animal, like mink, in which the virus didn’t originate. The distinction here is the necessary role of the intermediate host rather than a direct transfer from the reservoir animal in which the virus originated.

The report rates this hypothesis as “likely to very likely.” That’s a simple function of exposure: Humans spend a lot more time with domesticated animals than they do with wild ones. Domesticated animals, in turn, are far more likely to be exposed to wild animals. 

Investigators point to the documented infections of plausible intermediate hosts as evidence of this potential route of emergence. Indeed, 17 million mink were culled in Denmark after the emergence of a new variant among them (Which led to the emergence of “zombie mink!”). The problem with this hypothesis, though, is that there’s only evidence of significant transmission going one way—from humans to other animals, rather than from animals back to humans. Further, limited screening of domesticated animals across China shows no evidence of SARS-CoV-2 circulating there. 

This is a plausible hypothesis. 

Hypothesis: Laboratory accident at the Wuhan Institute of Virology.

The Wuhan Institute of Virology is a top-level biosafety (BSL-4) laboratory. While researching the coronavirus that had caused the original 2002-2004 SARS outbreak, scientists at the WIV had identified a colony of bats in Yunnan province where they believed the original SARS-CoV-1 to have originated. In their search, they identified three new viruses with spike proteins that could bind the human ACE2 receptor in the lung. Unclassified State Department cables warned of the Institute's potential risk as early as 2018 (for more on this, I encourage you to read Josh Rogin’s excellent reporting):

Most importantly, the researchers also showed that various SARS-like coronaviruses can interact with ACE2, the human receptor identified for SARS-coronavirus. This finding strongly suggests that SARS-like coronaviruses from bats can be transmitted to humans to cause SARS-like diseases. From a public health perspective, this makes the continued surveillance of SARS-like coronaviruses in bats and study of the animal-human interface critical to future emerging coronavirus outbreak prediction and prevention.

Beyond simply identifying these viruses, the lab may have been engaged in “gain-of-function” research, deliberately manipulating viruses to enhance their transmissibility or pathogenesis beyond what occurs naturally to better understand how they might evolve in the wild. These experiments were deemed so risky by the Obama administration that it put a moratorium on them in 2014.

The accident hypothesis postulates that the virus may have escaped the WIV inadvertently. Laboratory accidents are rare, but not as rare as you think (more on this from Alison Young). Heavily-guarded pathogens can escape even if a technician is not inadvertently infected: they can escape on clothing or equipment, they can accidentally be flushed in liquid waste, and so on. All of this is particularly true if technicians are poorly trained. On this, another 2018 State Department cable reads:

During interactions with scientists at the WIV laboratory, they noted the new lab has a serious shortage of appropriately trained technicians and investigators needed to safely operate this high-containment laboratory.

Indeed, even experts at the WIV expressed initial concern that SARS-CoV-2 originated in their facility. 

The investigators argue that such an accidental release is unlikely based on three pieces of evidence they were offered by the Chinese government. First, they report that there were no viruses similar to SARS-CoV-2 at the Institute prior to December of 2019. Second, they report no evidence of employees with COVID-19-like illnesses prior to that point, nor do they have serological evidence of infection with SARS-CoV-2 among employees at the Institute. And third, they report no unusual safety incidents prior to December of 2019. Yet because investigators were denied direct access to the data, all of this is based on second-hand reports from the Chinese government—a glaring flaw I’ll discuss below.

The WHO investigators rate the possibility of a laboratory accident as “highly unlikely”—so unlikely, in fact, that to them, it doesn’t warrant further study. But it doesn’t seem highly unlikely to me—particularly if its based on second-hand reports from an interested party. Indeed, based on the evidence, it seems entirely plausible. 

Hypothesis: The virus originated in frozen food.

This hypothesis builds on three pieces of limited evidence. First, SARS-CoV-2 has been isolated from frozen food packaging. Second, after China had largely mitigated viral spread in 2020, there were a few isolated outbreaks linked to frozen foods. Third, humans shed SARS-CoV-2 virus in feces. Though it doesn’t go as far as stating this outright, this hypothesis insinuates that the original 2019 outbreak in Wuhan resulted from a virus that had been shipped into the country via frozen foods. 

This hypothesis is so outlandish as to strain credulity. It’s theoretically possible that the virus could be transmitted in frozen foods--but we know definitively that foodborne exposure is not the dominant form of spread. Indeed, the critical feature of SARS-CoV-2 is its capacity to bind to the ACE2 receptor in the lung. The vast majority of cases are transmitted through the air. It’s why masking has been the single most important public health intervention to mitigate spread. The notion that the pandemic originated from frozen foodborne virus is farfetched.

If the virus were to have come from somewhere else, why didn’t we see the initial outbreak there? Though the virus can survive freezing—it does not replicate while frozen. Therefore, to have been transported in high-enough doses to be transmissible through frozen food, the virus would have had to be present in at least as high a dose at the origin—where it should have also sickened people. 

The report finds this hypothesis “possible.” I find it extremely implausible.


The implications of non-transparency.

The nature of the WHO's investigative process was wide open to political influence. The Chinese government initially balked at what it characterized as a blame game, leveraging its power to delay the investigation, claim veto power over participants, and pass on only the results of analyses, rather than the raw data itself. The on-the-ground investigation was highly choreographed by the Chinese government, under whose authority 17 of the report’s authors live. 

But transparency is the cornerstone of science. We ask a question. We posit an answer to that question in the form of a hypothesis. And then we do everything we can to disprove that hypothesis—abstracting away our priors and biases through transparency about what our evidence actually tells us about the validity of our hypothesis. 

As a regular consumer of science, one is used to a level of linearity. The arguments build on themselves. There’s a certain cadence for which scientists strive: premise, premise, premiseconclusion. Premise, premise, premiseconclusion. It’s like being driven down a familiar road by an excellent driver. No strange turns. No jerky motions. No surprises.

This report doesn’t read that way. Particularly as we approach the arguments for and against both the frozen food and WIV hypotheses, the report strains. It signals right, then turns left. It reaches feebly for far-off conclusions at times, ignoring the obvious at others. Premise, premise, premisestop. Premise, premise, premiseturn

In science, there’s a principle called Occam’s razor: the simplest answer is the likeliest answer. Following Occam’s razor: this virus most likely emerged from a reservoir host, jumping into humans either directly or indirectly through an intermediate host. An accident at the WIV is less likely, but remains plausible—while spread through frozen food is as unlikely as it is implausible. 

And yet the report takes great pains to establish the credibility of the least plausible hypothesis, and to play down a plausible one. Are we supposed to assume that it’s simply a matter of coincidence that the report struggles to validate an implausible hypothesis that would absolve the Chinese government of blame, while it downplays a plausible hypothesis that would lay the blame at its feet? Premise, premise, premiseconclusion.

Read in the light of the process that created it, it’s clear why this report leaves more questions than it answers. Given the geopolitical realities of this moment and the Chinese government’s omnipresent role in shaping it, the obfuscation seems intended—which calls into question the validity of the entire investigation. 

That does all of humanity a disservice. Indeed, only in understanding the process that created this pandemic that took millions of lives—destroyed billions of livelihoods—can we start the process of protecting ourselves from the next one.

In the end, though, science always wins (I believe it so deeply, I even made shirts!). It’s simply a matter of time. Though the WHO doesn’t answer the basic question of how SARS-CoV-2 emerged, it nevertheless does offer us some valuable insight: someone would like to delay that answer from coming out as long as possible.