How did the COVID-19 pandemic begin?

How did the COVID-19 pandemic begin?
© Getty Images

Over the last several months the question of where SARS-CoV-2 originated, exactly how the COVID-19 pandemic began, has become an increasing focus of public discussion. The issue has become fraught and politicized, debated in the halls of Congress, on the front page of leading publications and perhaps most contentiously on cable news.

The two dominant theories are that the virus was introduced into humans from an animal reservoir, or that a lab accident seeded the virus into the population of Wuhan. Recently, I joined in writing a critical review of this question with 20 other virologists, epidemiologists and evolutionary biologists from around the world. We examined all the scientific evidence available, and we came to the conclusion that SARS-CoV-2most likely originated in animals — although the possibility of a lab accident cannot be entirely dismissed.

The evidence for an animal origin of SARS-CoV-2 is strong and has steadily built as we’ve learned more about the early epidemic in Wuhan, China. A review of early cases shows a stark clustering in the neighborhood surrounding the Huanan Seafood Market. This we’ve known for a while, but information on excess deaths from the World Health Organization (WHO) Phase 1 origins report strengthens the association. There was some concern the link of a large percentage of early cases to this market might be due to a testing bias. Reporting of excess deaths, however, is testing neutral and that data too shows the first spike occurred in the part of the city centered around this market.


Another powerful piece of circumstantial evidence only recently came into view. The WHO origins team was told by Chinese authorities and carefully selected local residents that the Huanan market didn’t sell live mammals, including animals with the potential to carry SARS-CoV-2. If true, then the pandemic couldn’t have started at the market. No animals, no virus.

In June 2021, however, a bombshell paper was published reporting the opposite. Thousands of live mammals were sold illegally in Wuhan markets through November 2019, the month when the first likely first emerged, including potential reservoir species like the aptly named and strange-looking raccoon dog.

The genomic evidence is similarly consistent with a natural origin. Very early in the epidemic, as far back as December 2019, two distinct lineages of SARS-CoV-2 were already circulating simultaneously in Wuhan. It’s possible one evolved from the other, but given how early this occurred, it’s also possible slightly different versions of SARS-CoV-2 spilled over not just once, but twice. Is it far-fetched? According that June 2021 paper Wuhan markets shared supply chains, and in the 2002-2003 SARS-CoV-1 epidemic in China, infected animals were found in multiple markets across Guangdong province.

If a virus escapes from a lab not only would it be unlikely to happen twice in quick succession, but only one version, the lab version, should escape. If SARS-CoV-2 jumped twice, it must have done so from animals.

Despite all of this evidence, there still remains much suspicion that not only might SARS-CoV-2 have escaped from a lab, but that the virus was engineered or manipulated in the lab, explaining its high transmissibility and lethality. None of these arguments hold up. It’s often claimed SARS-CoV-2 is unusually good at attaching to human cells, that it must have been designed or adapted to do so. But other viruses are just as good or even better. There’s simply nothing special about SARS-CoV-2 in this regard.


The other feature of the virus that’s often pointed to as suspicious is the “furin cleavage site.” which is where the spike protein is cut while the virus is exiting one cell to pre-activate it for entering the next. Furin cleavage sites come in different flavors though, and the SARS-CoV-2 site is a particularly poor one, the spike protein isn’t cut very effectively so it’s decidedly not how a virologist would go about engineering an improved spike protein.

Out in the “wild” the furin cleavage site is even evolving to get better. Both the Alpha and Delta variants have mutations that improve the efficiency of the spike protein being cut — further bolstering the idea this feature arose via sloppy natural evolution, not by a skilled hand. 

So, what’s left in the possibility of a lab escape? Well, we can’t prove a negative, but there’s simply no evidence that SARS-CoV-2 was known to science, in or out outside of a lab, before it was discovered in patient samples in late December 2019 or early January 2020.

Of course, labs are always working on things they haven’t published yet, but in October 2019, before the pandemic, the Wuhan Institute of Virology submitted a paper containing over 600 new SARS-like coronavirus partial genome sequences, seemingly consistent with the lead researcher from the coronavirus lab telling Science magazine that they’d tested every single sample in their possession and, even before the pandemic, submitted a paper with no trace of the pandemic virus. If SARS-CoV-2 wasn’t in a lab, it can’t have escaped from one.

However, one ultimately views the balance of evidence (and myself and colleagues do clearly evaluate the evidence one particular way) we can all agree that the Chinese government has a responsibility to collaborate with international investigations on the origin of SARS-CoV-2.

Phase 1 of the WHO origins mission provided a wealth of useful information, even if Chinese officials weren’t forthcoming about the sale of live mammals in Wuhan markets. However, the prospects of further productive investigations look bleak. WHO Director-General Tedros Adhanom laid out a five point proposal on July 16 for Phase 2 with the first four “following the animals,” while calling for a more thorough review of relevant laboratories than was included in Phase 1. Given the relative priorities suggested by Adhanom, this is a good roadmap, and an important one since understanding the origin of this pandemic will help muster the political will to prevent the next one.

China, however, has categorically rejected the entire plan and tried to focus origins investigations on frozen food supply chains originating in other countries. This origin hypothesis is plainly absurd, and China’s seeming refusal to cooperate with any investigation — into markets or labs — will stymie any serious efforts to understand exactly how this virus came to infect us. Absent that understanding we await the next collision between humans and novel viruses.

Stephen Goldstein is an evolutionary virologist at the University of Utah. His most recent research examines how coronavirus evolve through genomic recombination. Follow him on Twitter: @stgoldst