Our long, pandemic-inspired videoconference moment comes with several benefits, including the comfort of having to live and dress the part from the waist-up only (in my case, wearing basketball shorts and house shoes below the reach of the webcam), to how it has encouraged us to get creative in our approaches to sharing our work.
In March 2021, I was able to deliver a research seminar at the University of Chicago—to an audience full of frighteningly smart people with large reputations—without the risk of being screamed at or having a tomato thrown at me.
The freedom of the videoconference emboldened me to try different things. For this seminar, I dedicated precious time to telling the audience about predictions and ideas that I was wrong about. Not about my broken NCAA bracket, but about the many ways that my early assumptions and predictions about the Covid-19 pandemic were incorrect. By doing this, I was hoping to give myself an intellectual challenge (to say something smart about being wrong), as well as mask my insecurity, impostor syndrome, and fear of talking to an audience of extremely smart people. This strategy is more than a little bit pretentious: By dissecting a wrong idea in front of everyone, I would signal how awesome I truly was.
The self-serving aspects of the approach were not, however, the only motivations for admitting I was wrong. Over the last year I’ve been frustrated with the scientific community’s general reluctance to openly discuss when and why we’re wrong, and specifically, in our study and prognostications of the pandemic. Our unwillingness to highlight what we were wrong about was a missed opportunity to teach the public about the scientific process, to put its necessary ups and downs on fuller display.
Our aversion to discussing our wrongness has had dire consequences: We (perhaps unintentionally) oversold our confidence in concepts that were still underdeveloped, alienated many who had legitimate questions, and (ironically) fanned the flames of misinformation and disinformation. For example, quacks have generated mashup-edits of prominent scientists saying one thing about Covid-19 in June 2020, a different thing in August, and something else in November. In response, we mostly offered the same flabbergasted response: “C’mon. That is wrong, and that isn’t how science works.” But our responses are missing something: We might be part of the problem.
What underlies scientists’ inability to cop to mistakes, flubs, or poor predictions?
It would be easy to pin it on the notoriously large egos of scientists. And while egos fuel many problems in science, I suspect that the real reasons for our Covid-19 stubbornness are more complicated.
From the beginning of the pandemic, misinformation and disinformation were not mere nuisances, but defining forces in the global response. And their most influential authors were not only renegade “doctors” with YouTube channels, but government officials directly responsible for the pandemic policy.
At the very least, bad information stymied or derailed public conversation about the science of Covid. The truth is more grim: The doubt that was inspired by bad faith actors drove formal public health policies (or non-policies). Skepticism and science denial had stakes far greater than the winner of a Twitter spat. Simple unknowns were weaponized, and many Covid lies were actively orchestrated and propagated in order to sow doubt about the way that science works, sometimes for political gain.
In the face of this, the scientific community’s reluctance to come clean about uncertainties and missteps are not only understandable, but even appropriate: There is a time and place to have abstract debates about the true meaning of “efficacy,” and a time to act on the information that we have in service of the public good. The pandemic, and the millions of lives (globally) that we lost in its wake, qualify as a large enough emergency that one can forgive a little chest-thumping bravado: We’re scientists, we’ve spent decades studying this stuff, and your bullshit is harming people. We, experts and the informed citizen-science public, might know that science is a process that cannot exist without accumulating new data and discarding old ideas. But much of the public is unaware of how this process actually works. Our “trust me, I’m a scientist” appeals can be misguided.
That said, sometimes our confidence and disinterest in engaging alternative takes is justified. For example, the evidence supporting the effectiveness of the Pfizer-BioNTech, Moderna, and Johnson & Johnson vaccines are overwhelming. Their clinical trials were rigorous, well-organized, and produced results that supported our insistence that they were a critical intervention. And today, most vaccine skepticism is truly magical or nefarious in nature, not based on legitimate critiques. The facts are very clearly on our side.
But not all of our Covid-19 opinions were driven by such strong evidence. During the spring of 2020, for example, I was a member of a small but mighty chorus of experts who were very concerned about the possibility that SARS-CoV-2 was transmitted via physical surfaces (indirect or surface transmission via “fomites”). To our credit, this chorus did little more than suggest that the public health sector should seriously consider this in their recommendations, or suggest that this route of transmission may have helped define the shape of the early outbreak.
Importantly, this sentiment was based on peer-reviewed studies that suggested that detectable virus did exist on physical surfaces of various kinds, and on our knowledge of other viruses that are transmitted this way. The possibility for surface transmission was also a part of the early justification for the “washing hands” public campaign. It’s not a crazy idea.
As the science behind the transmission of SARS-CoV-2 became more sophisticated (with evidence from engineers and physicists chiming in), the debate shifted to the nature of aerosol transmission, whether the virus was truly “airborne,” and whether long range transmission was possible. As the science of transmission progressed through the summer of 2020, the surface-transmission story fell further and further down our lists of things we were worried about.
As of April 2021, very few people discuss surface transmission at all, as it’s now understood to be (at most) a minor source of infection. When I became aware of these new developments, I did what a responsible scientist must do: I admitted I was wrong, and in my public discussions of Covid-19 transmission, I have since talked about why I was wrong, and what I learned.
While the debate about whether the virus was surviving on our mail, packages, and frozen food boxes came and went with a relative wimper, a different debate—whether the virus was “evolving” or not—has lingered, and with far greater consequences. This is an arena where I and many others who studied virus evolution offered early opinions. Most fell along the lines of: Mutations accumulate in virus lineages all the time, and these mutations often have geographical signatures that are reflective of their ancestry. Viruses circulating in Denmark can have different genetic signatures from those circulating in Chicago. The important question, and one where strong opinions (like my own) manifested was in expressing doubt that these mutations changed anything fundamental about how the virus was infecting and causing disease. We were skeptical about whether different geographical populations of SARS-CoV-2 constituted truly different strains. Even further, I thought that, given the relatively low mutation rate of coronaviruses (as compared to influenza, for example), and other attributes, that future evolution of new strains that caused essentially different diseases was unlikely.
Less than a year later, I must admit, at the expense of my ego, I was wrong. A major part of the pandemic narrative today is defined by the emergence of several global strains of SARS-CoV-2 that evidence suggests are more transmissible and possibly more virulent. While I stand behind my original objection to hysteria surrounding the existence of SARS-CoV-2 mutants, my grander claim that we shouldn’t be especially concerned about the evolution of new strains was incorrect. Not only do the set of newly evolved strains pose an imminent threat to billions of people, the possibility of future virus evolution now looms over vaccination efforts, and developers are preparing for the possibility that the variants may emerge that render our vaccines ineffective.
Again, I changed my tune as the data began to mount against my original stance. And the reason this is not embarrassing is because my wrongness is not a referendum on my expertise, but instead, is the normal price to pay for making predictions in the process of science. Or rather, there is no better way to learn and improve than for us to accumulate better data that we incorporate into our picture of the word, while letting go of our prior notions. This only works if the scientific community lays bare those instances when we make an honest mistake or bad prediction, and most importantly, explain why we’ve shifted our perspective.
As we reflect on ways to prevent the next pandemic, most conversations have appropriately focused on improving the sciences of prediction—data science, genomics, virus ecology, computational modeling, and others. But the science of communicating with the public is just as essential. We must build effective ways to explain more than the facts, but also the process through which ideas are born, live, and die.
Walking the public through this process will involve a willingness to take the world inside the capricious, sometimes chaotic world of science, where even the best of us are often wrong. And wrongness is not the sign of a flaw, but is a feature of a healthy, strong scientific instrument that may one day take us to the stars, prevent a climate crisis, and conquer the plagues of today and tomorrow.