How Can We Know What’s a Conspiracy Theory and What’s Not?


Last week, US intelligence agencies completed a 90-day investigation into the origins of SARS-CoV-2, concluding that a lab leak was as plausible as any explanation. And yet, for over a year, we were told by authorities that the virus had entered society naturally from bats, and those who suggested otherwise were regarded as conspiracy theorists and ostracized by their peers and censored by tech platforms.


The story of how this woefully premature consensus came to dominate discourse on COVID should be examined, as it reveals flaws in the way society decides what’s true, and offers lessons to prevent us being fooled again.


Most information about the world is transmitted along a chain of institutions before it reaches the average person. It’s first handled by researchers, before passing to the press, and then to tech platforms. As each of these institutions processes the information, it skews it according to its biases. The result is that people acquire not the truth but a distortion of a distortion of a distortion, and if the interests of all three institutions happen to align on a single story, then the cumulative effect of these distortions, all pushing in the same direction, will yield a single compelling mirage. Let’s consider how this happened with beliefs about COVID’s origins, beginning with the researchers.

Most information about the world is transmitted along a chain of institutions before it reaches the average person.

On February 19, 2020, a letter signed by 27 public health scientists was published in the esteemed medical journal, The Lancet. It described the lab leak hypothesis as a conspiracy theory, claiming “scientists from multiple countries have published and analyzed genomes of the causative agent,” and that such analyses led to only one conclusion: a natural origin for the virus. This was grossly misleading; among other things it didn’t account for “serial passaging,” a common method of modifying viruses by controlled natural selection, which yields viruses that are genomically indistinguishable from those that evolved naturally.


How the scientists overlooked such an obvious counterexample to their claim makes more sense when one considers the letter’s architect. The letter was secretly organized by Peter Daszak, a zoologist with strong motivations to want the lab leak hypothesis to be false. He is president of EcoHealth Alliance, a not-for-profit that had funded studies at the Wuhan Institute of Virology, which is only eight miles from where SARS-CoV-2 was first identified, and which was involved with “gain-of-function” research: modifying viruses to make them more infectious to humans so that potentially dangerous mutations can be studied, predicted, and preempted. Clearly, if such research was responsible for the pandemic, Daszak would be implicated, so it was in his interests for the virus to have emerged naturally.

Picture Credit: The National Academies

Fortunately for him, his plight was shared by many others who’d also been involved in funding, advocating, or conducting gain-of-function research. These included Shi Zhengli, director of coronavirus research at the Wuhan lab, who told lab leak theorists to “shut your dirty mouths,” and Anthony Fauci, White House health advisor and director of the National Institute of Allergy and Infectious Diseases (NIAID), who gave a White House press briefing in which he publicly dismissed the lab leak hypothesis (for which he was privately thanked by Daszak). Further support came from figures in the US and Chinese governments, the chief funders of gain-of-function research, who apparently dissuaded colleagues from investigating the true origin of COVID-19.


While such people may not have been intentionally trying to deceive, their personal interests plausibly skewed their judgement. Wanting to believe something is often as convincing as proof itself. And yet, the world didn’t necessarily want to believe it. So how did it become consensus? The problem began when it passed to the next link in the information chain: the media.


Right-wing outlets like Fox News were quick to examine theories that the virus had come from a lab. But such organizations are not taken seriously by most cultural elites, who get their news from liberal media like the New York Times, The Washington Post, CNN, and the BBC.


The liberal media had plenty of leads to investigate. There was the fact that Wuhan researchers had been studying RaTG13, a coronavirus that is the closest known relative of SARS-CoV-2, or the experiments modifying and testing various coronaviruses on mice genetically engineered to have human lung cells, or the US intel that in November 2019, before the first cases at the Huanan wet market, three workers at the Wuhan lab fell sick with COVID-like symptoms.


A few in the liberal media did follow some of these leads. For instance, Washington Post writer Josh Rogin wrote in April 2020 that the US State Department had previously found the Wuhan lab to be dangerously lacking in safety standards. But such points were ignored by most mainstream journalists, who, rather than question Daszak or Fauci, amplified or idealized them as pandemic heroes, while parroting their opinions about the origins of the virus.

Important points were ignored by most mainstream journalists, who, rather than question Daszak or Fauci, amplified or idealized them as pandemic heroes.

The mundane explanation for such behavior is that journalists are not scientists; they tend to learn of new science not from reading the actual papers but by reading the press releases, which are mainly just hype, glossing over the caveats and limitations of the studies. In many cases, science journalists don’t know enough to make sense of the data they’re presented with, so, rather than question the findings, as, say, a political journalist may question a politician’s economic proposals, they’re forced into the role of simply explaining the research.


But ignorance wasn’t the only force at work. The liberal media’s “woke” preoccupation with identity politics caused it to reflexively view the lab leak hypothesis as bigotry, with, for example, one Slate writer claiming “The rumors of a lab escape or a bio-weapon stem from historical amnesia, a caricatured villain, and good old-fashioned racism.” This idea was reinforced in liberal journalists’ minds by the fact that the lab leak hypothesis’ most famous advocate was the “racist-in-chief,” Donald Trump. The media had good reason to doubt him; he’d routinely pushed baseless conspiracy theories like Obamagate and Birtherism. The lab leak hypothesis was far more substantial, but unfortunately ideas, like people, are vulnerable to guilt by association, and the liberal media dismissed the hypothesis because of who’d shouted it loudest.

The lab leak hypothesis was initially widely discredited, in large part because its chief proponent, Trump, is notoriously dishonest and fond of conspiracy theories

We’ve seen that the two institutions people rely on for truth – the scientific establishment and the liberal media – are compromised by an array of personal, professional, and political motives. But there is a third institution in society’s consensus-manufacturing system, the last link in the chain, with the greatest power of all: big tech.


The very structure of tech platforms makes them natural amplifiers of the scientific establishment and the liberal media. Firstly, government bodies like the National Institute of Allergy and Infectious Diseases (NIAID), and liberal news outlets like the New York Times, are generally assigned higher credibility scores by search and recommendation algorithms, and even by such things as the list of acceptable sources for Wikipedia, meaning their messages will always be more visible than those outside the mainstream.


Secondly, the fact-checking organizations that social media giants rely on to tell them what’s true and what’s misinformation are not composed of experts, but of laypeople who reach their conclusions either from what they’re told is true by organizations like the NIAID or the liberal media, or from conducting online research using Google and Wikipedia, which usually leads them to organizations like the NIAID or the liberal media.


Thirdly, these same organizations can pressure tech platforms to silence dissent. The liberal media pressures the tech giants with headlines like “Far-Right Misinformation Is Thriving On Facebook” and “How Facebook Fuels Far-Right Profit,” which can scare advertisers away, hurting business. Further pressure comes from politicians, who regularly threaten tech companies with regulation if they don’t do more to combat misinformation. The demands became particularly vociferous after the storming of the Capitol in January, when Twitter, Facebook, and Google were accused of having “played a role in the fomenting of insurrection,” and were ordered to clamp down on conspiracy theories. Congressman Bill Bilirakis warned the tech giants, “We can do this with you or without you.”


Unsurprisingly, big tech platforms were quick to shut down the lab leak hypothesis. For instance, in February 2020 Facebook censored a New York Post article by the social scientist Steven Mosher arguing that the virus may have leaked from a lab, and a few months later Twitter suspended the account of the virologist Li-Meng Yan for claiming the virus was made in a lab. Meanwhile, Google ensured that typing terms like “coronavirus lab leak” into its search bar didn’t autocomplete (so users wouldn’t be led “down pathways”), and that searches for the terms would always yield statements by the WHO and other consensus-upholders as the top results. However, these measures were not enough for many liberal journalists and politicians, who demanded more be done, and so, in February 2021 Facebook issued a blanket ban on all posts suggesting the virus was “man-made or manufactured,” a policy it awkwardly reversed four months later when the consensus began to change.


As a result of pressure from officials, politicians, and journalists, the tech giants, who control the world’s information traffic, ended up not just promoting the idea that the lab leak hypothesis was just a conspiracy theory, but also actively silencing it.

As a result of pressure from officials, politicians, and journalists, the tech giants ended up censoring the lab leak hypothesis.

Once the lab leak hypothesis had been deemed a conspiracy theory and outlawed on social media, scientists who may have disagreed with the new consensus became afraid to oppose it, for fear of being denied funding by the scientific establishment, condemned by the liberal media, or silenced by big tech. Their reticence in turn reinforced the illusion that a consensus had been reached, beginning the cycle of lies again. In essence, the chain of information dissemination, from scientists, through the liberal media, to big tech, ate its own tail, like an ouroboros, creating a feedback loop in which the same idea was recirculated until it became familiar enough to sound like the consensus. The mainstream basically became an echo-chamber, because each of the three institutions depends on the others for information, and the incentives – professional for the scientists, political for the journalists, and financial for the tech giants – happened to align with a single premature conclusion.

The worst part is that this wasn’t an isolated incident. We saw something similar early in the pandemic, when mainstream institutions manufactured a false consensus that masks were ineffective at reducing infection. More egregious examples can be found throughout history. Ignaz Semmelweis was a physician who advised handwashing as a way to reduce patient deaths after surgery. This when the consensus was that illness was caused not by germs but by imbalances in the “four humors,” and so his life-saving advice was rejected and he was ridiculed, his life ending in a mental asylum. The Semmelweis reflex, named after him, refers to the tendency for people to dismiss new explanations in favor of established ones.


So what can be done to prevent future false consensuses? If we can’t trust our institutions to get things right, how do we distinguish truth from conspiracy theory


Truth is often measured in probabilities, not absolutes; you can only know what’s likely, not what is. And the probability of something being true increases in one direction: toward the root of the information chain. The fewer links in the chain, the fewer opportunities for information distortion. The further from the root you are, the more your beliefs rest on faith, and the more tentative they should be. One should therefore try to follow claims to their roots, and whenever one cannot, one should refrain from definite conclusions.


While it’s difficult to ascertain what’s true and false, it’s relatively simple to determine what is and isn’t a conspiracy theory. Conspiracy theories possess certain characteristics, none of which is in itself a perfect tell, but all of which, taken together, are reliable indicators.

Conspiracy theories possess certain characteristics, none of which is in itself a perfect tell, but all of which, taken together, are reliable indicators.

The first common characteristic of conspiracy theories is that they are expressed as certainties, leaving no room for doubt. The second is that they are based on unverifiable evidence; for instance, some claim by an anonymous source. The third characteristic is that, in lieu of evidence, they sustain themselves through circular reasoning, which typically manifests in conspiracy theorists as a habit of regarding attacks on the conspiracy theory as evidence of the conspiracy. The final major component of conspiracy theories is a villainous scapegoat, which can be an individual or institution. The scapegoat is usually regarded as shrewd and calculating, and evils are attributed to its design rather than its incompetence.


When one can identify these four components in a claim – certainty, unverifiable evidence, circular reasoning, and a scapegoat – it’s safe to conclude that one is looking at a conspiracy theory. Consider, for example, the claim by Chinese state media that SARS-CoV-2 leaked from Fort Detrick in the US. It’s a bold claim that is never itself questioned by the outlets that push it. But follow the chain of sources for the claim, and you find that everything it rests on either cannot be independently verified or is not compelling evidence of a coronavirus lab leak (e.g. the fact that Fort Detrick has conducted bioweapons research). Lacking any substantial evidence, this claim sustains itself primarily through circular reasoning, with, for example, one Chinese state media outlet suggesting that the unwillingness of US officials to investigate their theory is evidence for it. Finally, the claim has a clear scapegoat – the US government – which is accused not just of incompetence, but of trying to secretly engineer bioweapons. This conspiracy theory has been resurrected by Chinese state media in response to calls from the US and others to allow a full inquiry into COVID’s origins, and it doesn’t take a conspiracy theorist to see why the CCP would want to avoid an audit of its activities by redirecting blame for the pandemic onto a geopolitical foe.


Now let’s consider the belief that the lab leak hypothesis is just a conspiracy theory. This belief was held with enough certainty to become an enforced consensus. Yet it is based on flimsy evidence (mostly centered around Daszak and the genomic analysis he cited). It also involves circular reasoning: not only did the information chain become a feedback loop (a kind of institutional circular reasoning), but the hypothesis became a conspiracy theory only because those who asserted it were declared conspiracy theorists. Finally, the hypothesis became associated with a villainous scapegoat: conspiracy theorists, particularly Trump and his followers, who were accused by many in the media of purposefully trying to demonize Chinese people.


As such, the idea that the lab-leak hypothesis was a conspiracy theory was more of a conspiracy theory than the lab leak hypothesis.


Of course, some claims of a lab leak were indeed conspiracy theories, such as the one pushed by Trump, because it expressed certainty (“I think you can take the word ‘potential’ out that it came from the lab,” said Trump) and it did so at a time when no evidence was even available. As a result, it rested largely on circular reasoning (Trumpists believed it because Trump believed it). It even had a scapegoat – the Chinese – whose culpability was crudely reinforced with terms like “The China Virus” and “Kung Flu.” This conspiracy theory was the worst caricature of the lab leak hypothesis, and it was used as a straw man to attack and invalidate all forms of the hypothesis, including those that had none of the qualities of a conspiracy theory.


The most credible variants of the lab leak hypothesis were asserted early on by researchers like author and commentator Jamie Metzl, the biologist David Baltimore, and DRASTIC, an informal group of researchers collaborating via Twitter. These people held to a weak (and therefore strong) form of the hypothesis: not that the virus escaped from a lab, but that there was a strong enough possibility that it did to warrant further investigation.


Secondly, the researchers didn’t rely on unverifiable evidence; they presented all their sources for everyone to see. Nor did the researchers rely on circular reasoning; attacks on their hypotheses were not taken as evidence for the hypotheses, often the contrary was true. For instance, after David Baltimore was criticized for too hastily calling a part of the virus’ structure, the furin cleavage site, a “smoking gun” for a lab leak, he was quick to walk back the claim.


Finally, the researchers, unlike the conspiracy theorists, were careful not to scapegoat anyone; they didn’t assume the leak was malice rather than incompetence, and they focused on cause rather than blame. As if to draw a line between himself and the Trumpists, Metzl wrote on his blog, “I in no way seek to support or align myself with any activities that may be considered unfair, dishonest, nationalistic, racist, bigoted, or biased in any way.”


Of course, none of this did anything to prevent the genuine researchers from being lumped in with the Alex Joneses of the world, and denied a platform in the mainstream. Fortunately, the investigators were able to find a platform on alternative media. Metzl’s ideas reached their greatest audience not through CNN or MSNBC, but through the Joe Rogan Experience podcast, while Baltimore’s view entered public consciousness largely through a Medium blogpost by the independent journalist Nicholas Wade. Such forms of alternative media proved themselves to be as powerful disseminators of ideas as the establishment itself. The institutions are losing their monopoly on information; the chain is becoming a tree, with many different branches. Whether or not this is a good thing will depend on how well we can distinguish a researcher from a conspiracy theorist.


We may never know for sure whether the pandemic came from a lab or the wild, but the story of how we decided what was true has lessons for us all. The institutions that were supposed to protect us from conspiracy theories themselves fell victim to a conspiracy theory. They may not learn from their mistake, but we can learn from the people who corrected this mistake by refusing to stop investigating when everyone else had. Truth is not always found at the top of Google search results, or in the New York Times, or in what the “experts” say. Truth is a process of constantly asking questions and refining hypotheses, and consensus, well, that’s a mutual agreement to stop.