Power, Knowledge, and COVID-19

Transcript

[Sandro Galea] Welcome to Ideas Matter, a podcast hosted by WashU. I am Sandro Galea, vice provost of interdisciplinary initiatives and dean of Public Health at WashU. Today, we are going to discuss something many would say should not exist, scientific orthodoxy. The dictionary defines orthodox as conforming to established doctrine, especially in religion. So officially, science has no doctrine. Science at its best operates through the reasoned pursuit of truth. That’s the ideal.

But sometimes science falls short of the ideal. The COVID-19 pandemic was arguably such a moment. During the pandemic, voices within the scientific community sometimes neglected reason and evidence in favor of dogma, certainty, and groupthink. This hurt our capacity to respond to the pandemic and contributed to a collapse in public trust in scientific institutions. Rebuilding this trust means reckoning with the COVID moment to understand what went wrong, how scientific consensus hardened into orthodoxy, and how we might avoid that trap in the future. That is a tall order. Fortunately, today’s guests have written a book on how we can do just that. Alex Broadbent is professor of philosophy of science at Durham University and deputy executive dean for international partnerships in the Faculty of Arts and Humanities. He’s also a visiting professor at the University of Johannesburg and director of the Durham Johannesburg Centre for Philosophy of Epidemiology, Medicine and Public Health, based in Durham’s Institute for Medical Humanities, where he is a fellow.

Piet Streicher is a research associate at the University of Johannesburg Centre for Philosophy of Epidemiology, Medicine and Public Health. He’s also an honorary fellow in the Institute for Medical Humanities at Durham University. They have a new book out called Power, Knowledge and COVID-19: The Making of a Scientific Orthodoxy. The book explores how science and public health fell short during the COVID pandemic and how we might do better next time. I’m delighted to be speaking with them today. Alex, Piet, welcome.

[Alex Broadbent] Thank you so much, it’s great to be talking to you.

[Pieter Streicher] Thank you.

[Sandro Galea] So lead me through the book. So why did you come to write this book? Why this topic and why now?

[Alex Broadbent] Well, I think we both came at initially slightly different angles and then we kind of came and found we had some complementary expertise. Suddenly for me it was very much a sense-making exercise. When the pandemic kicked off I was actually in America and I was just flying back to South Africa where I then lived and I looked at the World Health Organization’s advice and it seemed to be the same for Africa as it was for Boston where I was leaving, which was just, hey, keep two meters distance from And I just didn’t understand how the same advice could be given everywhere. And then when I got back, of course, then lockdowns began and I became involved writing sort of opinion pieces saying, how can we do this in the African context? How are you going to lock down in a township? How are you going to separate if you have 15 people living in a shack and so forth, those were the early days. But then as the dust settled later, the urge came to try and actually understand how some of these things that had puzzled me in the first had come to pass in that way. That was certainly how I initially came to it.

[Sandro Galea] So Pieter lead us through the book’s central thesis.

[Pieter Streicher] So we noticed that science appeared to be a lot more dogmatic during the pandemic. There were black and white statements made, lack of nuance, and a one size fits all policy recommended. I came to this project from a slightly different angle.

My original background is engineering. So I got involved in modeling the pandemic. And the first thing that struck me was some of the hospital bed estimates, which were way over inflated. That’s when I wrote my first article, specifically on the South African modeling teams where I simply asked the question, why do they expect hospital loads that exceeded Italy and New York by 10 times? Yeah, so in that process both Alex and I, when we tried to engage with authorities, there was just a level of hostility, so we were not used to that in science. We see science information as open to correction. And that was certainly not the attitude that we experienced.

[Sandro Galea] So Alex, lead us through how did this happen? I think you and I agree that a scientific orthodoxy emerged during COVID. So why did that happen and what were the features of this orthodoxy?

[Alex Broadbent] Well, we’re in the slightly embarrassing position of having a whole book and still having to say that we don’t quite know exactly why it all happened. But I do think what we’ve done is offered a framework for understanding. So as we see it, the sense we were able to make of it, is that a certain view emerged due to I think a very strange constellation of factors, which meant that you you suddenly had a collapse of policy, apparent policy choices, into a binary where you really just had seemingly two options. One was to do a lockdown and one was to do nothing, where doing nothing has somehow expanded to include all kinds of mitigation measures and all kinds of things that would normally have been layered on top of each other.

The construction of that moment, I think, was a contribution of science and of specifically of actually of it was an artifact of models that there was this binary choice really where the sort of mitigation scenarios all got collapsed into one. And in fact, trying to distinguish between mitigation scenarios was almost at some point actively discouraged. it was seen as, dilly dallying around when really what you had to do was go in hard with a lockdown. And I think what then happened was that then tried to assess the effects of what had been done. And in doing that, were very resistant to evidence suggested anything other than what they’d originally proposed. So there was this, what we call a dogma, scientific dogma, which we define as resistance to recalcitrant evidence. And I think people, they looked at the lockdowns that happened in Europe and in America and so forth, and they said, well, they saved millions but the assessments weren’t actually that, well, weren’t that secure, really. They weren’t done in the spirit of, let’s really subject this to a rigorous test, they were done with a different spirit, let’s say.

[Sandro Galea] Well, Piet, let me push on early modeling for a second. So lead us through how the early modeling distinctions and assumptions, particularly around suppression and mitigation, how did that contribute to the emergence of lockdown as the default policy position? So how did that happen?

[Pieter Streicher] Yeah, so, I think some of the earlier reports we go through, Imperial College Report 9, they set up a dichotomy between mitigation and suppression and to a large extent, it was manufactured. So on the one side, you have the counterfactual scenario where perhaps you do nothing or even with optimal mitigation, they had extremely high attack rates, 40% for optimal mitigation and 80% attack rates. for unmitigated scenaria. Now the models are based on certain assumptions which really do not hold in the real world. The models assume that everyone’s equally susceptible, equally contagious. And interestingly, the secondary household attack rate of the first variant was only 15%. So what that means is, typically if you have a household of ten, or let’s say five people, perhaps one other person would be infected if an index case is infected. So those 40% and 80% levels were extremely high, but they were difficult to falsify because they were not playing out in reality. Most countries decided to lock down. So the cracks started to show when countries that did not lock down also had relatively low attack rates. So I would say attack rates varied from below 1% to 15% in certain cities. And Sweden ended with a 6% attack rate for the country and 14% for Stockholm without locking down.

So there was not such a big difference between locking down and not locking down. And the way that Imperial College Report 9 set it up was, it’s almost like a threshold effect. You have high attack rates and you might increase restrictions with no effect until you exceed a certain threshold and then you will achieve suppression, you will achieve low attack rates. When in reality we believe things lie on a continuum. As you ramp up restrictions, you will reduce attack rates, but there’s no magical threshold. We almost see it as a situation of diminishing returns where if you increase restrictions above a certain level, the benefits you start to get in terms of reducing attack rates is very little. But at the same time, you see escalating costs. So the first model, an all or nothing scenario with a threshold, sort of drives you towards maximal suppression. And that’s why we see all countries opted for in lockdown. While the way we see it is more like an 80-20 scenario where you can significantly reduce attack rates with very mild restrictions. As you ramp up those restrictions, you start to see diminishing returns and escalating costs. And with that framework, it’s a level of judgment that you need to exercise to decide on what level of restrictions you need.

[Sandro Galea] I think it’s interesting that you attribute some of these orthodoxies to the of binary thinking, which of course bedevils a lot of science and it’s interesting that in a fraught moment it became something larger than itself. Let me shift focus for a second. Let me talk about one of the other flashpoints of the COVID moment, one which I think certainly took me and I think many of us by surprise, which is the intensity of debates about masking and vaccination. So Alex, maybe lead us through how did these debates become so fraught? Why did the side of science and public health at times struggle to engage with good faith concerns about vaccine and masking, and to my mind, essentially then becoming dogmatic on these issues in the face of that struggle. What happened?

[Alex Broadbent] Yeah, I think it’s very interesting how that played out in both cases, both masks and vaccines. Masks are very interesting because their public health significance is arguably not nearly as great as that of either vaccines nor lockdowns, nor is their significance as an impingement on freedom. So they took on a kind of symbolism, possibly just because they were so visible. You can tell if somebody’s wearing a mask just by looking at them. I think that had a lot to do with it. With masks, the interesting thing is that the evidence around masks is complicated and not that conclusive. There is evidence, but it reflects a kind of an old sort of split between people who think you need randomized control trials and people who think you don’t need randomized control trials. You can act with enough observational evidence.

And in a way the situation with masks plays into that existing, pre-existing methodological thinking about public health. And people tended, I think, to have their prior buttons pushed a bit around that. Around the vaccines, the interesting thing, again we think so we think there was a dogma in relation to both masks and to vaccines, that’s not the same as saying that the claims made were false, it’s not the same as saying that masks had no effect, certainly not the same as saying that vaccines were ineffective. We do think vaccines were effective but again there, that the claims that were made about vaccines were far far stronger than the evidence necessarily fully supported. So you see well-known effects, the healthy vaccinee effect, for example, which is a well-known statistical bias that can affect observation studies, sort of reproduced in leading journals. And the result was a significant overstatement and push back against perfectly reasonable expressions of doubt. It’s reasonable to have some elements of doubt. This might be a good moment to explain a little bit the religious overtone as well, of the term orthodoxy that we use. So, science is supposed to distinguish itself by having a different epistemology, an epistemology that doesn’t depend on authority.

And it’s supposed to be different from, hopes to be different from its predecessors’ knowledge systems in that regard. And what was interesting here was that in our analysis anyway, you see pronouncements and you see in particular attitudes to evidence that are actually much more rigid and much more reminiscent of older, you know, medieval religious attitudes to dissent and to recalcitrant evidence. And I think that that happened around masks and around vaccines, both of them. So I mean, to reiterate, we certainly don’t think that, particularly in the case of vaccines, we certainly are not saying that they were not effective, quite the opposite. We think they had significant effects. There were, however, strong statements, categorical statements, about their being safe and effective, which just were not, it was sort of not permissible to ask if there was evidence telling against them. And of course there’s always evidence on both sides.

[Sandro Galea] Yeah, let me, Alex, let me just push on something you alluded to in your last responsible speech. So I think, as any observer would have concluded during COVID, that the scientific community was challenged in its dealing with speech. There were times during the pandemic when some views were not just rebutted through presentation of data, but they were essentially rejected, pushed aside. And we saw this treatment applied to ideas and to people.

And it was a curious turn of events. Why do you think this happened? What were the means by which views were sidelined and some elevated? And perhaps central to the thesis of this conversation, how much did these speech wars, to call them that, contribute to this notion of a scientific orthodoxy?

[Alex Broadbent] I think they were a major component. Suppression of dissent is one of the elements in scientific orthodoxy in our analysis. So we looked at, in particular, the Great Barrington Declaration here. And then the Great Barrington Declaration is not something, you know, we don’t necessarily agree with every element of it at all. But what we thought was striking was that it was handled in a way that wasn’t, well, it wasn’t, I suppose for want of a better word, scientific. But we also make the point in the book that it was itself was not a scientific intervention. So the Great Barrington Declaration, was issued publicly. It was an attempt to sort of get government to take it on. And in a similar vein, the response was to sort of take it down and to attack it. So the whole thing was happening.

It was like scientists sort of fighting in the street situation. It wasn’t a debate in the journals. I think is what’s key to understanding what’s going on here. So this wasn’t, a whole thing that was happening in the pandemic, it wasn’t just scientific debate. It was something happening at the interface of science and policy. It had to do with power. It was scientists along with other people trying to decide whose voice was going to actually influence what was done. And I think that explains a lot of what happened. People aren’t doing this, weren’t doing this for the sake of personal gain or something. They were doing it because they really thought they were right and that they were going to be saving lives. But the reason that these things happened, I think, that there was this attitude to dissent, is because they were engaged in essentially political battle. They were they were trying to control what was done. So it’s about the interaction between science and policy, science and authority, and that’s why this wasn’t a sort of peer-reviewed debate in a series of journals I think

[Pieter Streicher] Yeah, we definitely see that.

[Sandro Galea] Yeah, so Piet, let me come to you.

[Pieter Streicher] Yeah the way I see it is, science is suddenly thrust into a position of significant power. And that’s part of the reason why these dynamics play out. You know, it’s a matter that is serious, affects people’s lives, and is playing out in the public sphere. And that’s why we see it something new. It’s something that perhaps affected the medieval church at a time where they were in, or had significant power. And I see it as a part of human nature. You can see the way the scientists discuss their messaging to the public because there are minutes of the meetings that they held. You can see that they are clearly thinking of saying things to make people believe certain things, so they simplify the messaging. The messaging becomes more black and white, becomes more dogmatic. It makes it easy for the public to distinguish between misinformation and the correct information. But that in effect puts people in the firing line. If they make a very simplistic black and white statement, for instance, vaccines work and vaccines are safe, I mean, there are scientists that work on the safety of vaccines. Their job is to point out if there is a signal, if there is a problem, and then even they come under attack because misinformation is then simply seen as something negative about vaccines or something negative about lockdowns.

[Sandro Galea] So Pieter let me build on it a little bit and let’s accept for a second that this orthodoxy emerged during COVID, which I think is a relatively easy argument to accept today in 2026. Now, some might read your book and think, okay, so an orthodoxy emerged during COVID, but that was a unique historical moment. We shouldn’t worry too much about the excesses of the period. They were a product of an unusual time and it’s unlikely to happen again in our lifetime. So from your perspective, having been immersed in this and writing this book, to what extent were these circumstances that produced orthodoxy unique to the COVID moment? And to what extent are they generalizable at other times when science risks becoming dogmatic? Should we be worried about science falling into these traps again?

[Pieter Streicher] I think we need to be worried.

It was definitely a unique episode, but we do foresee this situation happening in the future whenever it’s a serious matter and a matter that plays out in the public domain and where scientific institutions see it as important to control the narrative. Then you will always have that risk to be more dogmatic and you will have the related problems where people aren’t just seen as being wrong, but where people are seen as heretics. And it’s something that we need to look out for because we believe science cannot function properly in such an environment.

There must be level of dissent and discussion. I mean, with COVID-19, decisions had to be made very quickly when perhaps all the information was not available. But what we saw is then it was just assumed that whatever they decided was correct when perhaps a month later with new information, things could have been re-looked at and then reinterpreted in a different way. Instead, the initial understanding hardened into dogma and almost became canonical. And it only, as I said, only became obvious later on during the pandemic, especially in the UK, where much lighter restrictions were selected by the government and then those counterfactual scenarios or intended counterfactual scenarios became the actual scenarios. And then the predictions were way over. We pointed out one example during the Omicron variant deaths were overestimated by three to 30 times for a scenario of light restrictions, which at that time the government chose to persist with.

[Sandro Galea] Let me shift, although this is really related to power. Alex, let’s talk about power for a second. So on the one hand, power is necessary for the application of data towards a range of society improving ends from preserving health during a pandemic to shaping better policies in different contexts. On the other hand, as philosophers have long written about, power can have a corrupting influence. It can make us partisan. It can turn us away from reason towards orthodoxy. So how can we engage with power constructively without being corrupted? What are the practices that a field, but let’s here focus on public health, can adopt to help with this?

[Alex Broadbent] Yeah, I think this is a major challenge of our times. Science is the knowledge producer of our contemporary world. And as such, it rightly should wield a certain kind of political power. It has a certain right to a seat at the table that other
practices and institutions don’t necessarily, at least in our view. In virtue of that it has the great potential to improve the quality of our decisions. On the other hand becoming a scientist doesn’t involve any process that normally you’d go through if you were going to hold political power. You’re not elected, you’re not appointed through some process. Science itself doesn’t have an internal governance structure which is suitable for bearing the weight of the political authority that contemporary society needs to place on science. That’s our analysis anyway. So what I think really needs to happen is that that needs to change. There needs to be a lot more thought given to governance structures within science that make it able to bear the weight that contemporary society needs to place on it. If science is indeed going to wield the kind of power that I think it probably ought to wield, given that it’s the major knowledge producer in contemporary society. So, I mean, our suggestion is that there are certain situations where scientists can wield extremely strong power, if you like, extremely direct power. We talk about a situation where they can have indirect political authority, which is where you have exclusive epistemic access, only they know something. And then there’s an agreed set of background conditions that everybody agrees to. And then there’s some fact that they discover, which when you slot it into those background decisions, just greatly restricts the range of policy options. And we discuss this situation in the book.

And it’s an attempt to be really charitable because we think a lot of scientists in the pandemic felt they were in this situation. They felt they were in the situation of the marine biologist in Jaws who knows there’s a shock there and tells the mayor to close the beach. They feel like they’re the relevant experts. They know something. They’re the only people who can know it and people really should listen to them. And we have some sympathy with that. We think that there are situations where that is the case. We happen, however, not to think that those conditions were satisfied in the pandemic. And that’s supposed to illustrate how if you have a sort of a governance structure, you have a set of tests you can apply, then you can actually tell whether certain claims that scientists are making, that they really have a claim to urgency and to be listened to. You can tell whether those claims are correct. And you can do so in a way that can in principle be sort of, you know, bureaucratized and run through committees and so forth.

[Sandro Galea] Pieter, I think we are still living with the fallout of the COVID moment. Science has lost the public’s trust. We face political headwinds all over the world, particularly in the United States, where I’m sitting. And there are moments when it’s hard to see a way back to a kind of prelapsarian public health. So how do we move forward in the moment? How can we reclaim trust lost, rebuild from COVID, and avoid mistakes going forward?

[Pieter Streicher] I see trust as something that is installed over time. You can’t tell people to trust you. You have to earn trust. One of my concerns during the pandemic is a lot of the restrictions were sort of assuming that people were not adults. And for public health to be trusted, public health also has to trust the public. I do believe if people get the right information, they will do the right thing. And it’s something that also shows up in the data, which I found extremely interesting. In the UK, when they lifted all restrictions in July, or most of the restrictions in July 2021, that was called Freedom Day. It turned out that the elderly continued taking special care for another six months. So they didn’t go back to normal behavior immediately when the restrictions were lifted, when they were allowed to do anything they liked. The younger people did in fact go back to normal behavior very quickly. And we see immediately infections rise amongst the younger generation, but amongst the older generation during the Delta wave in the UK, even though there were no restrictions, the attack rates remained low. And even with the first Omicron wave, attack rates remained low. And that just shows you that if people are of their risks and they are informed how to protect themselves, they will do so. We think it’s an extremely blunt instrument to force everyone to stay at home. The economic costs such policies, social and economic costs, are just enormous. And you benefit where it counts the least. The people that you succeed in keeping at home will be the youngsters, the elderly people that were at risk would have stayed at home in any event.

[Sandro Galea] I feel like these are arguments and discussions and debates we’re going to be having for a few more years. And it’s probably a good thing for the science, that we continue to actually have this conversation. I think your book is a nice addition to the thinking behind this. So thank you to both of you. Last question, Piet and Alex. In this moment, and particularly in a moment science is under challenge in some respects, like never in recent memory. When I see the work that you’re doing as engaging seriously with the shortcomings of science as a way of strengthening it and preserving it, which is a tricky thing to do. It’s tricky to defend something while also critiquing it. I respect the effort to do that. So in this moment, what gives you hope? Piet.

[Pieter Streicher] I think there is a positive tendency in the sense that it’s easier now to reflect critically on the pandemic and to publish something on the pandemic compared to what it was in 2020 or 2021. I think as the dust settles, people are more open to critical views, critical reflection.

And that’s why we engaged in this project and we believe it’s extremely important. Once the dust has settled, stand back, think what just happened, and see if we can make a contribution.

[Sandro Galea] Alex.

[Alex Broadbent] Yeah, yeah, I would also say, if I think about how it feels to say the things we want to say now compared to how it felt four years ago, or even two years ago, it’s considerably different. I notice if I talk to people and tell them I’m maybe writing a book on the pandemic, and it’s perhaps a bit critical, people’s reactions now, they’re like, really, that’s interesting. They want to know. Go back a couple of years and you’d be very careful about saying, even just a couple of years ago, about saying that sort of thing. So that’s hopeful. Because it does mean that people haven’t just memory holed it completely, which I think is the risk. I’m also sort of hopeful just because I do think what happened was the product of a certain moment. That’s not say other such moments can’t arise. This is not an anti-science book. This is a book about, as you said, to try and defend science in a way from how it can go wrong. And I do think that the core idea of, as I see it, of science, of not basing things on authority but of basing things on a rational response to evidence, I think one has to hope that that has the power to win out.

[Sandro Galea] I’m Sandro Galea. I have been talking with Alex Broadbent and Piet Streicher about their book, Power, Knowledge, and COVID-19. Thank you, Alex and Piet for joining.

[Pieter Streicher] Thank you, Sandro.

[Sandro Galea] And thank you to everybody who has joined us for this Ideas Matter. I look forward to continuing the conversation.