A year into the pandemic, much more is known about the coronavirus. In any study, especially scientific ones, it takes time to collect data, it takes time to collect samples, it takes time to run experiments, and it takes time to find the right and diverse sample size to have meaningful and representative data. However, subsequent changes in public health policies and guidelines make a population weary, and many didn’t and couldn’t trust the science, especially when they were already hesitant to believe the institutions making the suggestions in the first place.
With the recent NIH disclosure that funds indeed were shuttled to Wuhan for research on coronaviruses preceding 2020, it validates concerns of many once seen as conspiracy theorists. The combination of the slowness of scientific research that guides critical public health recommendations and mandates in addition to the fraught political polarization that wants a solution now clash and results in further confusion, mistrust, and even betrayal.
At the beginning of the pandemic, I remember thinking, naively and hopefully, that maybe this could be the external threat that could bring even just the United States together as a country, if not the world. Unsurprisingly however, statements and actions quickly became political and divisive. If you didn’t wear a mask or refused to get vaccinated, you were probably conservative. If you were pro-lock down of schools and businesses, you were probably liberal. Once you were in a camp, you had a set of heuristics used to gauge whether or not a source was credible, whether or not a person should be listened to or not, whether or not a person was dangerous or a hero.
Heuristics aren’t inherently bad; indeed, such social epistemological tools are critical to not be overwhelmed by the sheer amount of information and chatter that surrounds us. However, they can quickly become a vice when bought into uncritically and when opposing opinions are shut down immediately rather than understood and engaged with in some capacity. It’s tricky though because no one wants to admit that they were wrong, least of all very public officials, with the repercussions being seen as stupid or in some cases a liar.
With so much focus on emerging scientific research, in many ways it made sense to “follow the science.” The problem became that if there was any skepticism — a very scientific disposition, ironically — about recommendations, then skeptics were seen as anti-science, as risks to public health, as Other. Perhaps this is why recommendations, like the Great Barrington Declaration, did not get much coverage because, instead of promoting shut downs and vaccines, it suggested “Focused Protection,” that would lead to herd immunity by “allow[ing] those who are at minimal risk of death to live their lives normally to build up immunity to the virus through natural infection, while better protecting those who are at highest risk.” Not a terribly popular opinion, but one signed by physicians, epidemiologists, and immunologists of prestigious universities and hospitals around the world. What does it mean, then, to “follow the science” when consensus conflicts?
Reporters, journalists, and editorial staff of news outlets have their sources that they trust. This trust comes from both the source’s knowledge and credibility, but also their values and arguably their values play a bigger role. Humans align themselves together based on ideologies rather than data, and people trust stories not facts. Any one watching Rand Paul interrogate Anthony Fauci about government funding coronavirus research will likely be rooting for one or the other, based on a set and history of identity politics. Subsequently, respective news outlets will spin the coverage in their favor.
When information comes out later that discredits the hero of the story, outlets and their audiences have a critical choice to make: double down on their beliefs or let go and admit they were wrong. Now that it’s been made publicly known that the NIH did give funds and resources to a group called EcoHealth Alliance to study coronaviruses in Wuhan in 2018-2019, it raises doubts, or at least questions, about who can actually be trusted. When Fauci vehemently says that the NIH did not fund gain-of-function research, and Peter Daszak of EcoHealth Alliance and prominent epidemiologist in the field vehemently claims that a lab leak is a conspiracy theory, can or should we actually believe them?
No one wants to admit that they were wrong or a liar, and they want to remain aligned with their camp to not be cast out as Other. This psychology is actually dangerous when enacted at the level of politics and journalism, especially consciously. It’s one thing to be unable to stay afloat in a rapidly changing environment, but it’s another to be willfully ignorant, and the space in-between is very much in the fog. From a journalist’s or reporter’s standpoint, it begs the questions of who should sources be? How much of the opposing side should be reported? What amount of uncertainty should be in the story? It seems to me that, intentionally or otherwise, the mainstream eyes and mouth that critiqued and judged the minority side were not used in favor of portraying a bullet proof story. Perhaps this was in favor of not feeding scientific doubt, but it still ended up hiding information.
Many of these problems come down to needing answers and guidelines right away, not to mention the 24-hour news cycle demanding our (emotional) attention. With so many people getting sick and dying around the world, something needed to be done and I don’t blame the extreme measures that governments and individuals took to ensure the safety of themselves and their loved ones. So many things were not known, but something needed to happen so actions were taken based on history, scientific studies, belief, and anticipated outcomes. However, the extremity of the situation played no small part in polarizing perspectives, all while in an uncertain and incomplete information ecosystem.
In many cases, while studies were just getting off the ground and not yet yielding results, the public had to trust the word of experts in the meantime instead of data itself. At first, there was no data that indicated that wearing a mask was critical, no data showing what type of mask would work, no data of long-term effects of Covid-19. Even now, there’s no data showing that vaccines impact human fertility, no data indicating kids wearing masks impacts their development. The danger in the headline “There’s No Evidence to Support This,” suggests that “This” has been studied and the results show a non-correlation. Instead, it could mean that there is no evidence at all, that studies haven’t started yet or are still ongoing, that there’s literally not enough data to even be making the statement in the first place. We want answers, we want certainty, but headlines like that are misleading (and whether or not that’s intentional is unclear).
Given that in many cases there’s little or no evidence to support one way or the other, I cannot blame those who want to be mothers for not getting the vaccine or parents who don’t mask their small ones. Their decision is just as scientifically informed as vaccinated mothers and parents of masked kids without comprehensive long term studies. It comes down to values anyway more than scientific evidence: is it better for kids to be vaccinated and be able to attend school or to be unvaccinated and wear a mask which could impair their learning? Even with studies about the impacts of Covid-19 on health and education completed, peer reviewed, published, and reported on, I doubt that anyone would change their mind or feel regret for their past decisions anyway. By then, most will be so ingrained in their opinions and polarized claims that they’ll double down rather than let go, or they’ll be accustomed to that way of life and feel no need to make a change. Cannot reverse getting vaccinated anyway.
So much of this goes beyond media literacy and scientific understanding, and gets right at the heart of group identity and the myth of evil that leads us to find shelter in a siloed heuristic model that pins us against Other. It’s uncomfortable to see that the structures we’ve bought into aren’t as solid as we hoped, that maybe the other side we’ve been busy vilifying saw something in the fray that we didn’t, but it’s better to face the realities of the situation however complex they are and to face the mistakes to learn from them.
This idea came from reading and discussing some Undark articles (this one and this one) with some science writing friends. So thanks to ST, KB, AT, & LW for these meetings and great conversations.
Leave a Reply