[MUSIC PLAYING]
I’m Ezra Klein, and this is “The Ezra Klein Show.” [MUSIC PLAYING]
So as a million media theorists before me have argued, in a few short decades, or depending on how you want to rate it, a few centuries, we’ve moved from the defining problem of human civilization being information scarcity — we didn’t know enough — to the defining problem being information abundance — we know too much, and it’s paralyzing.
The people we’re following right now are the ones who don’t just seem to have a lot of knowledge, but who seem to actually know what to do with all that knowledge, who seem able to find the signal in the noise. And few have a better track record on this in the last couple of years than Zeynep Tufekci. As my colleague, Ben Smith, wrote in an August profile of Tufekci, “Tufekci has made a habit of being right on the big things.” She saw the threat of coronavirus early, and she raised the alarm. She saw that the public health community was ignoring the evidence on masking, and she wrote about that persuasively enough that she actually tipped the Centers for Disease Control and Prevention towards a new life-saving guidance. But — and I think this is really important — some people are just right in a pessimistic direction. They always think things are going to go wrong. And so, they’re right whenever they do. But that’s not true for Tufekci. Recently, she’s been pushing the media on the fact that it’s become too pessimistic in everything it reports on coronavirus. And in particular, it is overreporting what we don’t know about vaccines and underreporting how good they actually are. But before Tufekci was being prescient about coronavirus, I knew her because she was prescient about disinformation online, about the way social media was changing political organizing, about the way the YouTube algorithm was radicalizing, about the rising threat of authoritarianism in America. She is unusual in being able to say important, ahead-of-the-curve things in a lot of different areas and spheres. So I asked Tufekci, who’s a sociologist at the University of North Carolina, and a columnist at “The Atlantic,” and a contributor at The New York Times Opinion Section to come on the show for a conversation about metacognition. How does she think about what she knows, and how can the rest of us learn to think a little bit more like her? As always, my email, if you want to tell me what you think, is ezrakleinshow@nytimes.com. Here’s Zeynep Tufekci.
So I want to begin where you begin, which is in the systems thinking. What does it mean to think in systems? What’s even the alternative?
When I say systems thinking, I’m saying looking at the whole and its interactions as much as possible to understand both each part of it, but also how it all comes together. For example, you have something called the case fatality rate in epidemiology, right? Like this virus. What percent of the people who are infected go on to have severe disease, and what percent of them have terrible outcomes, like death? Now, that’s not like a fixed number that occurs the same way everywhere. Because how’s your hospital capacity? How’s your I.C.U. capacity? How many nurses do you have for people? How is your clinical understanding? I mean, even if you have the same set of medical tools, which in the beginning, that’s what we had. We had the same set of medical tools, which was not a lot. If you have an overloaded hospital, that means that people are getting oxygen too late because they’re not admitted, because they’re just overloaded, that they’re not being checked up as often — all these other things that happen that increase the fatality rate.
The difficulty of thinking in systems is that you need to learn about systems. And in particular, you need to learn about many different systems. So how do you do that? You’re a sociologist. I follow your work on politics. It’s very good. That’s my system that I know pretty well. You’ve been way ahead on coronavirus. You’re very good at moving into new disciplines and understanding how those systems work. And I’m curious what your approach to that is. How do you learn about new systems when you identify one you need to understand?
So I try not to move into completely new stuff, of course, because that’s how you get into epistemological trouble, where you try to think about things you don’t really understand well. And I did kind of move into pandemic writing, partly because there was an emergency. There needed to be more writing on certain aspects. And I was in a position with a platform to do so. So I ended up doing that. And I don’t really have a formula, but one of the things I do, do is, I read a lot of things directly. I mean, I don’t just read newspaper articles or press releases about a paper. I go read the paper. And I have enough of a background to at least understand some of the statistics or methods, especially if it’s a field like epidemiology, which has a lot of relationship to sociology. And plus, it was something I taught a lot as part of teaching people sociology. I used to teach pandemics. I wasn’t completely unfamiliar with it. And then I go out of my way to try to find experts in the field to keep asking questions, too. To give you an example, before January 20, 2020, the news coming out of China, which was repeated to us by the World Health Organization, was that there is no human-to-human transmission, or that there’s no evidence for it, which I thought was not true at the time. Because we were seeing all these news coming out of Wuhan. It was clearly spreading. Taiwan, which took action very early, had some experts on the ground. They were saying that it probably had human-to-human transmission. We had a case in Thailand, I think, on January 14th. And she had never been to the Wuhan seafood market. So it was clear she had gotten some place. So till then, I didn’t really believe the pronouncements. When China announced that they were closing Wuhan, my conclusion was that this was very real. It was a big deal because I understand authoritarian systems. And one rule for what they do is, you want to look at what they do, and not what they say. So if a country like China is closing down a city of 11 million, this is a big deal. It is spreading. It is deadly. And we’re going to get hit. So from that moment on, we started getting these papers coming out of China, saying — and Chinese officials telling us and Chinese scientists telling us that not only was there human-to-human transmission, that they were seeing asymptomatic transmission, that they were seeing incubation periods be infectious. And that was met in the U.S. with great skepticism. I have quotes from some of the top public health officials in this country, who say, that doesn’t sound right. We don’t think China can back this up. Now, I mean, obviously, they are medical people. Obviously, they know the medical part better. But at that point, I thought there’s a principle called the “principle of embarrassment” when trying to understand the historical accuracy of stories, is that if a story is really embarrassing to the teller, you kind of think they might be telling the truth. Because otherwise, it’s the kind of thing that people don’t usually admit about themselves, or institutions. So, when China was telling us after January 20th that it was spreading during the incubation period from people that didn’t have symptoms, that was actually making it look very bad because they had told us until then it wasn’t happening at all. And all of a sudden, they’re telling us something. And I thought, you know what? They’re telling us the truth. Because right now, they just really want to prevent the pandemic because they covered it up for too long. They kind of got caught. Now it’s going to spread to the world. And they’re going to get blamed for it. And now, they’re telling the truth. So I had a completely different sense of what they said before January 20th when they lied and covered it up. And it was kind of not treated with the correct suspicion compared to what they said afterwards. Now the reason I’m telling you all this is, there’s these ways in which even if you don’t necessarily have direct evidence on the medical side, if you kind of understand how institutions and authoritarians work, there’s a way in which you get more information about their claims.
This is a very useful framework, and it helps me contextualize something you’ve been arguing recently. So, the media is an institution. Public health in America is an institution. And both institutions regret not raising the alarm on this and being more pessimistic or realistic in their outlook right at the outset. But something you’ve been arguing recently is that these institutions have become too pessimistic now. They’re downplaying optimistic findings. They’re downplaying things that are actually getting better. Can you talk a bit about that, what you see that is getting better, and then how you think of that as interacting with some of the incentives for these players and institutions now?
So I think there’s a lot of what we would call “herding effects” or versions of groupthink effects. These are like sociological concepts where the players in the institution kind of look at each other to decide what the norm is. And then they kind of reflect that. And by the norm, I mean, just, there’s the mood, how you feel about it. They have a framework through which they interpret stuff. And very often, they don’t move fast enough because such frameworks, such heuristics, such ways of looking at the world have a lot of inertia to them. When the sort of news out of Wuhan came out on January 20th and I thought, that’s it, if they’re closing this, and we had millions of people travel around the world since the beginning of January. And then on January — the next week or so, I saw a paper. There was a paper in Lancet, and there was a New England Journal of Medicine report that were saying, yeah, transmission without symptoms. We started seeing cases pop up. And I thought, OK, we’re going to get hit with a pandemic. This is serious. It’s a novel coronavirus. And then I spent a lot of February kind of living in an out of body experience because a lot of journalists, friends of mine, were talking about the harms of the panic, rather than talking about the pandemic, the thing that was about to hit us. So they were conflating alarmism with misinformation. And sometimes, alarmism is misinformation, right? If it’s unwarranted, but accurate alarm isn’t misinformation. If there’s a tsunami about to hit you, then it’s not misinformation or alarmism to get worried and say, we’ve got to get to high water. We got to get to high water. It hasn’t hit you, but you know it’s going to hit you. So I was seeing articles that were saying things like travel was fine. I saw an op-ed by travel industry person in The New York Times saying, don’t try to create fear around travel, when in a pandemic, it’s not fear. It’s completely realistic to say, all right, how are we going to adjust to this? Because travel is going to carry the virus. There was a lot of conflating of the fact that Trump’s racism against China meant that we couldn’t discuss anything about this virus as a threat originating in Wuhan. It was really weird. It was sort of seen as a racist thing sometimes because Trump is racist, right? There’s just no doubt the way he’s talked about the virus. And all the ways he’s gone about it isn’t healthy at all. But the opposite of that isn’t acting like there is no pandemic coming our way that originated in one part of the world, which is Wuhan, China. And that’s where, initially, at least, where a lot of sort of the cases were coming out from. And then, of course, since then, it’s just caught in lots of places. I even got told this was just an elite panic. There were articles in The New York Times saying it’s the pandemic panic that’s the real problem. We’re going to overreact and do more harm. And I’m like, no, we’re underreacting. It’s a pandemic. Pandemics are a regular feature of human societies. And we’re about to get hit with one. So I spent February completely separated mentally and psychologically from the people that I am normally friends with and normally the people I read and I hang out with because they were just acting like this was just a panicky thing, and what about the flu? Now, if you want to fast forward to January of 2021, it’s a complicated, different thing. Because on the one hand, it’s certainly true the winter is grim. The case numbers are finally going down a little, but they’re still very high. The United States has very high levels of community transmission, which changes the calculus because the odds that you’re going to interact with someone who’s infected is directly related to how many people are there getting infected. And we also have these new variants that are more transmissible. And they’re not exactly sure why they’re more transmissible. But basically, it means that they’re evading something better, right? Whatever that is, until we figure out, it just kind of means you have to be even more careful because you’re not going to get away with things you got away with before the same way. So we have a grim winter coming ahead of us and all the things. But on the vaccine front, we have hit home runs that were so unlikely if you look at the history of this, that once the history of this is written, they are going to be referred to as some of the greatest achievements of science. To sort of give a context to this, in May of 2020, when people like Dr. Fauci were saying, maybe we’ll have vaccines in about 18 months, meaning September or something, like eight months into the future, people thought, no, that’s too optimistic because usually, it takes longer. And when the Food and Drug Administration was deciding the threshold for sort of clearing vaccines for emergency use authorization, they thought 50 percent efficacy would be good. Now what did we get with Pfizer, BioNTech, and Moderna? We have 95 percent efficacy against any symptomatic disease, meaning 95 percent of the people develop nothing if they’re vaccinated, no symptoms. And I went and read — this is the thing that I really think helps inform how I think, is that I go read all the documents, which is pretty time consuming. But it’s really amazing how much more you learn if you go read the whole thing, appendixes and all. What I found was that even in the 5 percent — so if you combine — on the Moderna trial, there’s not a single severe case. On the Pfizer trial, there’s one. That is one out of 30,000, 30,000 something. And that one, the details in the appendix says, is a woman who developed a 93 percent oxygen saturation. So if that had been 94 percent, it wouldn’t even count as severe. She was right at the cutoff, and she needed no medical care. She needed no hospitalization. So when you hear severe, you’re thinking I.C.U., ventilator. None of that was necessary. So a doctor described it as a clinical nothing burger. She just had a low reading on our [INAUDIBLE], and then it passed. She got better. So that’s just amazing. By vaccine standards, to have something that amazing is wonderful. And it’s the kind of thing you would have national celebration and fireworks and church bells ringing and all of that. And then I spent the last couple of months reading a lot of doom and gloom about the vaccines, that I’m just like, I wake up every day, and I think this is amazing. This is so exciting. And instead, we’re flooded with articles about, you can’t do this, you can’t do that. You can’t take off your mask. We don’t know if they prevent transmission. We don’t know this. We don’t know that. Which they’re not technically wrong, but to put the emphasis on some temporary limits and some temporary limits to our knowledge, rather than emphasizing how amazing they are, and then when the variants came, of course, the obvious question was, how will this affect vaccine efficacy? And people just went to looked at it clearly because it’s an important thing, and the news was great. There was some drop in neutralizing antibody titers, but vaccines aren’t drugs, right? So they don’t work the same way. But they still cleared the bar required for them to be effective, so to the degree their efficacy might be affected, like people think it’s not going to be a lot. But I saw these scary, doomy, grim headlines talking about neutralizing antibody titers dropping as if that’s what the public should care about, rather than saying, good news, they’re expected to be effective. And it’s great, and they’re already working on the booster of full — so this is kind of like, I’m now having this other outer body experience.
And I think that’s one of those herd dynamics. Nobody wants to be in the position of underselling how wrong you can go again. Right? That’s something that the media, say, learned, right? And now it becomes hard to say, well, maybe we could actually be surprised to the upside. But one way I’ve been thinking about this period — I just wrote an op-ed about this — is, the picture for six months from now, because of how good the vaccines are, is significantly brighter than we had any reason to expect. But the picture for six weeks or three months from now could be much grimmer than we expected. And that gets to another way that is hard to think in this, which is trying to think in terms of exponential time. The virus grows exponentially. We know these new strains are here. It seems reasonable to say that the U.K. strain, the B.1.1.7, is something around 50 percent more transmissible. And if that’s true, then for a little while, it doesn’t seem like anything is happening. And then, all of a sudden, everything is happening. And I think we’ve had a lot of trouble, the media, policymakers, human beings, operating in that way, right? Feeling that the times when things are quiet does not necessarily mean they’re quiet. Can you talk a little bit about that, the question of how the world changes when you begin looking at things through exponential growth and decay?
That’s a very good point. Just flattening the curve is just trying to explain exponential growth, basically. And so, I’ve been writing about the power of exponential growth for a long time now on the pandemic side. But of course, you know, my original area of study, social media, public sphere, there are other exponential growth phenomena, like Facebook network effects, all of those things. So it’s concepts I was familiar with. But it’s very hard to think like that. Because in February, when I was really worried, I think the U.S. had a single death in all of February from the coronavirus. But again, that was misleading because that’s kind of like the waters that are receding because we’re going to get hit with a tsunami. So that’s not really your indicator, but it’s very hard for normal people to think about exponential growth. Because it’s ordinarily not a part of our everyday experience. I wrote about the transmissible variant for an article for “The Atlantic” on December 31st. I spent New Year’s Eve writing it out because I was worried. I was seeing health departments tweet things like, there’s a new variant, but it’s not more deadly. It’s just more transmissible. And I was like, wait, that’s worse, right? More deadly only affects the people infected. And that’s a tragedy for them. But more transmissible is an exponential process. And it’s going to end up with a lot more people dead. If you just do the math, you would prefer more deadly for the infected, which, again, is an individual tragedy. But at the population level, more transmissible is going to result in much more deaths. And interestingly, some of the early examples in warning came from the tech industry. They think in exponentials because network effects, like Facebook’s growth and things that are exponential processes. And a lot of venture capitalists in sort of the VC world and the software people, they’re looking for that next exponential effect. So I think they might have had a natural affinity to thinking in exponentials because it’s a very lucrative way to get paid. So they were in a position almost they had some intuition because of the field they were in. The Western pandemic playbook was very flu-based because we are used to flu epidemics, and we were expecting a flu pandemic. Whereas Southeast Asia and East Asia, their pandemic playbook was more SARS, MERS kind of playbook, the overdispersed pathogen and all of that. Because that’s what they lived through in 2003 in Hong Kong and Japan. So they had a much more SARS-y playbook. And I kind of say this because sometimes there’s a tendency to think everything that went wrong in the United States is the fault of Donald Trump. And clearly, the administration has been terrible in managing this pandemic. But if you look at Western Europe, they’re not doing that great either. Their vaccine performance is terrible. I don’t understand what’s going on with Europe at the moment. And U.K.‘s vaccine performance is not terrible at all, but in summer, not only did they lift the restrictions, they gave everybody coupons. Like the government encouraged everybody coupons so they could go eat in restaurants, take-out not allowed. That is not what you do if you understand how exponentials work. It’s just insane. Because we know this transmits indoors. You don’t sort of lift the lid early. So a lot of Europe and Western Europe, in different ways, kind of failed. And I think there’s a bunch of things going on here. And one reason is, really, the flu pandemic playbook in our head, rather than the SARS pandemic playbook, like understanding the virology and epidemiology of this pathogen. You fight the war you got, not the one that you have the playbook for. And we weren’t, I think, good at that. And partly, you get what you get. So everybody gets surprised, but we didn’t adjust fast enough. So here we are. [MUSIC PLAYING]
These next six weeks, 10 weeks, with these variants circulating will become really, really important. And there are things that we could do that we haven’t done yet. You made a really important point, which is, I think Donald Trump’s failures, they were so visible and so profound and, in many ways, so offensive, that they can really distract from other things other institutions, other players, that are failing. And I want to talk about the F.D.A.. And I don’t want to be too hard on them. But there are a bunch of things where it seems to me that given the costs, they are being way too cautious and conservative. And one that Dr. Mina talks about a lot is rapid at-home antigen testing, where we could have these little strips. They don’t cost very much. We could constantly test ourselves. Then if we got a positive, we could double check it. And if we were positive, then begin to self-isolate. Can you talk a bit about that and why, in your view, the F.D.A. has been so recalcitrant on approving those?
I’ll say a yes and no. I think, for example, the vaccine approval process in the United States has been pretty healthy. So F.D.A. did a lot of things right, moved fast, and intervened early with making sure underrepresented groups were in the trial. The hearings were really transparent and great. Right now, there’s all these questions about AstraZeneca approval in Germany and U.K. And I can’t really find detailed data, and I don’t know what on Earth is going on. And the U.S. F.D.A. was like, no, we don’t really do it that way, which is great for that, because vaccines are serious things. And you do not want to sort of appear to have rushed or not have full data. Now, on the rapid tests, I think that’s an example of not being able to shift thinking properly. Because as far as I can tell, they keep comparing it to PCR tests, which are the ones where you look at an individual to try to find if they have any of the virus, right? Let me put it this way. You’re trying to find if you can find the fingerprints of the virus. Now, if I am about to operate on someone, right, and I need to know, is that person infected with it, do they have Covid, that test is really appropriate. I want that kind of accuracy. But at the population level, if you are using it as a population screening tool, speed is much more important than, does it miss a few cases? Because if you don’t have speed, you’re missing all the cases, right? If you don’t have rapid return to your test, if it’s like what Dr. Mina’s thinking, is like, you have almost pregnancy tests, right? Immediately tells you something. And the reason F.D.A. doesn’t seem to be approving it is that it’s not going to catch every last case like that, which is fine. Whatever it catches is something that it’s catching. And the fact that it’s not as accurate as the individual test is not a problem because we can then tell people, well, do it again. There’s all these ways of combining it with other versions of these tests to get better and better in the accuracy. Because it’s something if you do it every day at the population level, then you can kind of say, all right, let me reduce my risk. So if, let’s say, you have a scenario in which these rapid tests allow us to warn 50 percent of the people when they’re at their most infectious stage, which is another important thing, because the PCR test will catch that fingerprint long after you’re infectious, right? Because you just have some remnants of the virus that’s still kind of working its way out of your body. Whereas these tests are better at catching you exactly when you’re symptomatic and infectious. But even if you caught half the time, that’s more than zero. That’s a lot more than zero. And you could rapidly — and I think Dr. Mina’s right about this. You could rapidly get this under control. But the way we had it has just been wrong because there’s a couple of them that are approved. But they’re approved in the worst possible way. They require prescription, and they come in this expensive $25 setup. I mean, the real thing you want is the strip, right? You don’t really need the fancy casing for it. And you’re paying for it. Plus needing a prescription means you have to go to a doctor. And that’s exactly what we don’t want. We want people to be able to buy 100 of these strips and use one every day. And if it’s giving you a warning signal, depending on the type of test, you can tell people, hmm, looks like there’s something here. And now the next step is, you do another one, and you do this and you do that. So I think it’s kind of this way of thinking that we saw this with masks, too. There was all this discussion about, should we recommend it or not? And I was like, what’s the downside? Even if you’re not convinced they’re going to be perfect, now everybody’s pretty convinced that they’re going to be somewhat helpful. There’s no big downside, especially early on, when we didn’t understand things very well, to start sort of thinking about this. So the way you think about drugs is, you want to have them be accurate, and you want them to be safe. You want them to be effective because there’s a lot of harm you can do. With something like rapid tests, you want them to work for what you need them to do, which is help us at the population level, even if it’s not as accurate as a PCR at the individual level. So it’s kind of the sort of mixing and matching of tools that I think F.D.A. has been good on one kind, and not so good on being more flexible on another kind of tool, which is the population health approach, which kind of goes back to thinking about maybe it’s the sort of individualistic mindset in clinical medicine. Clinical medicine, especially in Western countries, we tend to really think about individual outcomes, rather than public health, and what we need at the population level, something that helps society in general, versus what you need to do when you have a patient in front of you, are not always identical ways of thinking. And the trick is sort of matching them at each level. And I don’t think we’ve done a really good job of it in the United States.
One of the things that the Trump administration doesn’t get enough credit for here is Warp Speed, which, really, compared to what we’re seeing in the E.U. right now, did accelerate vaccines. It did get us a bunch of doses that we needed. They did a bad job of distribution, but I do want to call out that Warp Speed seems like it was actually an important advance on what other countries did and how this has played out in the past. But given that and given what you said about the F.D.A. being pretty good on vaccines, you’ve been pushing the F.D.A. and public health authorities to make some different decisions there to go with first doses very, very rapidly and a lot of second doses to push farther back on the calendar in order to increase supply. Others or maybe you as well have talked about half dosing Moderna for young people in particular. And what you hear is pushback from regulators and others is, well, we don’t have the evidence on that. Can you talk a little bit about that distinction?
Sure, let me clarify first. I did write an op-ed with Dr. Mina, and we called for the evidence. We called for the trial. So I’m not in a position to advocate for the timing of the second dose, but when we saw the data that came out of Pfizer and Moderna, it was pretty clear that there was protection from clinical disease after the first shot. And the question was, how long does this last? How far can we push the booster, right? Because if you look at history of vaccination, delaying the booster or fractional dosing in the face of shortages is commonly done. So we weren’t really proposing something really unusual, right? Because this is what we have done with yellow fever vaccines, facing shortages. And lots of vaccines, the four weeks is a minimum, right? It’s not necessarily the maximum. So we wrote an op-ed saying, you know what? Let’s go get this data as soon as possible. Because if we get the data, and if we find that, we can push the booster to eight weeks or 12 weeks or six months, whatever we kind of find, especially for younger populations. Because from the phase one and phase two trials, the elderly, who are more susceptible to severe disease, weren’t having as good a response to the vaccine. So you might want to not do that on them because they need the maximal protection. But for young people, that might be a valuable thing. So we wrote a thing saying, this is really important to get data for so we can have the grounding to make the decision. And you would have thought we had proposed something outrageous because then we’re just sort of calling for a proper randomized trial of a practice that’s not even uncommon in the world of vaccines. And there was all this, you can’t even propose this, kind of pushback because it’s dangerous, or it’s this or that. And we were really scratching our heads, thinking, how is it dangerous to propose a trial for a practice that is not uncommon? And then, after sort of it was met with, this is outrageous, how can you think about it, shortly afterwards, what happened was a lot of leading immunologists and sort of pretty important people in the field started proposing, forget the trial. Let’s just go ahead and do it. And we had proposed a trial and get the data. But there are a lot of experts who thought we had enough data to proceed, given what we know about how the immune system works. And there were others who disagreed. So it was this contentious thing in that there were reasons to think and complicated things. But it was the kind of thing that clearly, there was a merit to all the arguments, right? Because you have an unhappy situation, which is shortages. So you’re doing something less than ideal, and there’s merit to both of the ideas. Because if you cover more people quickly, you can get the deaths down really quickly, as opposed to, like, if you had two kids, would you give a prime and a booster to one and leave the second one unvaccinated, right? That’s kind of the question that we’re facing.
Whew, that is a really clarifying way to put that.
Well, that is exactly the question, though. Like, if you have two kids, are you going to leave one completely unvaccinated if you can give one of them 95 percent efficacy? Or if you think the vaccine efficacy goes down to 80 percent, are you going to vaccinate both? That’s the question we’re facing. Are there other dangers like [INAUDIBLE] and other things? So I mean, it’s simplifying somewhat, but that is essentially the question. And there were arguments on both sides. So I’m not going to sort of pretend to have the answer myself, especially since this is not my field. But we had proposed the trial. So after we proposed the trial, I saw lots of leading people kind of say, yeah, you don’t even need the trial. We have efficacy data from the existing trials. And you’re facing that question of two kids, and you’re going to leave one unvaccinated. We wouldn’t do that for our kids. Why are we doing this for the population? So people were saying things like that. The metaphor is mine, but they were essentially saying, yeah, go ahead. And others were disagreeing strongly. There was a lot of back and forth. And then, lo and behold, the United Kingdom said, all right, we’re going to allow 12 weeks. And then the C.D.C., after the new administration came, said, OK, delay up to six weeks is fine. So, here, another interesting thing happened. All these people that were so mad at us for suggesting even a trial kind of stopped discussing some of this after the World Health Organization said six weeks OK, too. And the C.D.C. said OK. So it wasn’t actually that controversial. And right after, just last week, the person from Oxford who runs the U.K. recovery trial, he was saying, yeah, let’s randomize this sort of delaying booster thing to see what happens, which is exactly our original position. So, to me, that was kind of an interesting turn of events in the epistemology sense in that things that sometimes appear really controversial at the time actually aren’t.
Ooh, this is, I think, such a good and important space and also so tricky to talk about well, so I’ll try to do it carefully. One of, I think, the more poisonous lines in this whole conversation is, we need to listen to the science. It is almost always said on things where the science cannot give you a full answer, where there are values that play differing equities, things that we don’t fully know. But I think the idea of science operates on an undue level. There are things where the science really can tell us things, right? Do these vaccines work? The science has an answer. The science cannot tell us exactly how to structure who gets them first and who gets them next and which direction we go in. I have to think, though, there’s a difference between this idea of listen to the science, and then listen to the scientists. And particularly on Twitter, I think that public health authorities who have been, in many ways, phenomenal and are doing incredibly, incredibly difficult work, but you can sometimes see a group dynamic of, it’s important that people listen to us. So if you question us, you’re creating a bigger problem, and you need to stop. And I got the sense that sometimes that op-ed and some other things are taken as questioning, right? Like, there’s this bigger issue of keeping the science credible. But in the long run, I don’t think it’s good for Science, capital S, to react too badly when challenged. And then, on the other hand, of course, in a world of anti-vaxxers and rampant disinformation, you can understand — and for that matter, by the way, harassment — you can understand why people get defensive.
Yeah, it’s very challenging. So let me put it this way. I am completely 100 percent wholly body and soul on the side of science, right? That’s how you understand these things. That’s how you think about it. That’s how we get out of this. But that’s not really saying much. That doesn’t tell you how to go about it. Because if it’s listen to the scientists, I’ll find you a scientist saying anything. In this topic, there are scientists with the credentials making any claim you want, including anti-vaxxers. There are medical doctors who are anti-vaxxers. It’s terrible, but they exist. So when you say, listen to the science and listen to the scientists, you have to say, well, what does it mean? Ideally, it means that we have a process in which we earn that trust, where the science is, as you say, for example, it can tell us the efficacy of these vaccines. Epidemiology can tell us sort of the distribution of risk. After that, when you have a shortage, you have these really tough questions on prioritization, and what are the goals you’re looking at, and how do you deal with the shortage and all of that. And those things absolutely require a scientifically informed debate. And the science itself should explain why and how we think something. Because just say listen to the scientist doesn’t work because, as I said, there are scientists who hold really terrible and wrong views. And the fact that they have the medical degree or the PhD doesn’t exempt them from basically promoting snake oil. It happens. So you have to say, what is the process through which we earn that trust and say, this thing we’re injecting into your body, it is going to help protect you from this disease. It’s going to help protect you from the severe disease. We’ve gone to great amount of effort to make sure it’s safe. And we’re doing it in a medical clinic or setting with nurses standing by because there are these very, very rare side effects with allergies. But we know how to deal with them, too, and here’s what happens so far — that kind of sort of earning the trust in the institutions that I think is the key process here. And I understand the defensiveness. The public health people, they’re getting death threats sometimes, and they’re dealing with the quackery and things like that. But how dare you question us is not a way to earn trust.
Yeah, and I would say — and this is in politics, too, particularly in progressive politics. Because sometimes I think there is an aesthetic of scientism that is using the performance of science as a way of developing authority that you can use to end the debate. And that almost never works, in fact.
The thing is that is not even necessary. Science is amazing. And while medicine is not perfect by any means — it suffers from not listening to the people occasionally. There have been missteps. A lot of especially underrepresented minority groups have been mistrusted and not listened to. And you go back in, and it’s experimented on. So there’s all these things. But on the other hand, it is also a spectacularly successful endeavor, right? We don’t need to play performing games here. We have the real thing. We just have to kind of, I think, defend it the right way. The way I try to explain it is, if my son gets the strep throat, I mean, I just go get antibiotics from the pharmacy. And before the pandemic, it would be like, oh, OK, he can’t go to school for a day. If you go back to 70, 80 years ago, I’d be thinking about his funeral. And that’s the reality. That’s where science has taken us. Things that used to be devastating are now minor inconveniences. If you read any history from the 19th century or early 20th century, people lost their kids all the time. It was a routine thing. It was terrible, and it was routine. If you walk in a cemetery, which I like doing things like that, you will see a lot of children and babies who died very young. And if you look at sort of what we have achieved with vaccination, with antibiotics, it is spectacular. And it doesn’t require that we call blind obedience. It requires that we kind of explain it and we fix what’s wrong with it. We fix the parts that are wrong. The statistics about just something like the amount of infant mortality among Black people is the shame of this country. And it is clearly just the structural inequality and the racism and the lack of health services, and plus, the way health providers treat Black people, right? That’s what’s going on. That is terrible, and we should call it out. But on the other hand, we can defend the good things, the antibiotics, the vaccinations, and all the things, and just work to earn that trust. And so, that’s kind of what I mean. What you call scientism, I completely agree. I listen to science as a way to shut people up. It doesn’t work and isn’t necessary and doesn’t make science better, right? We need to make the practice of science and its benefits be more equitable. So that doesn’t help with that. Just telling people, just shut up, never works. I mean, it’s not even effective. If it was effective, maybe I would say, OK, this is terrible messaging, but it doesn’t even work. [MUSIC PLAYING]
You’ve done a lot of work on social media, on social media algorithms. How in the end did you feel about Twitter and Facebook’s decision to ban Donald Trump?
Well, let me say that to give an answer would be starting the story very late. So that’s the problem, is, by the time we got to the point of needing to deplatform the President of the United States, it’s almost too late to be talking about it. So whether or not one thinks it’s justified or not, the real question is, how on Earth did we get here? And what role did our information ecology from Facebook to Fox News play in this to the past decades of everything from the financial crisis to the Iraq War? So I almost feel like we’re focusing on the period at the end of a sentence, rather than trying to understand how we got to that point. Because I mean, this is a man that travels with the nuclear football, the power to end life on Earth, basically. And by the time you’re discussing, should he be allowed to tweet, I’m kind of like, I think we lost the plot here. How did we get here, is the more important question. It’s almost more important than what did he do in the last two weeks, can he tweet or not. But if I were running that place, as an emergency measure, I would, too, right? I would have said, I am pulling this megaphone. But that is not how it should be, obviously, right? Because it is an emergency that is partly of those companies’ making. It is a failing of our own society that we got to that point. And as I keep writing about, there is 66 percent of the House G.O.P. voted to overturn the election right after the storming of the mob happened. We have these problems with minority rule, the structural minority rule in the Senate, the opportunistic minority rule through gerrymandering, all those things, kind of Democratic backsliding combining. So even though I studied the social media part of it, the question of what do you do on that day in an emergency, every answer is wrong at that point, right? Because deplatforming the President of the United States is clearly wrong. But at that point, not deplatforming is also wrong, right? There is no good answer once you get to that point. So the question should be, how on Earth did we let things get to this point? Which is important because that’s how you can start thinking how on Earth do we get out of that point? And that’s kind of where — that’s why I’m kind of hesitant to say it was a good choice or a bad choice because at that point, every choice is a terrible choice.
I very much agree on this. It’s a very hard cases make bad law kind of moment. But Trump is interesting from the worldviews perspective we’ve been talking about here in that he is a place where systems thinking and individualistic analysis collide. So I wrote a book about polarization and radicalization in politics. And it’s framed entirely at the beginning around systems thinking and seeing American politics as interlocking systems. And at the same time that I think Donald Trump is a product of those, certainly a product of, say, the electoral college and its preference for Republican-leaning areas — without that, he never wins the election — he’s also distinctive. You’ve written some really, really great stuff that I agree with on the ways in which Trump could be succeeded by a much more effective kind of authoritarian. But the pushback I always get on that idea, which I hold as well, is that Trump’s celebrity, his particular personality made him capable of doing this in a way that just somebody else who’s a more normal, capable politician would not be able to, that you can’t separate what made him win from what made him, in certain ways, ineffective. And I’m curious how you think about that, how you think about balancing, clearly, individual qualities of leaders from the systemic things that create them.
I love this question because this is exactly the kind of thing I like thinking about. And what I’d like to say is that history creates its personalities in that there is a tendency to highlight heroic efforts to do something good, especially in poorer countries — the teacher who’s going out of their way to sort of walk for miles to go to that village school and do their best and the sort of above and beyond the call of duty kind of thing. And I’m from a third world country myself and have spent a lot of time here and there. And usually, what I think is, in a lot of those places, you have a lot of heroes, but what you don’t have is sort of the opportunities, like the routine opportunities that people in better off countries have. And that’s what needs fixing. You don’t want to send more heroes to that place. You just want to make them unnecessary. There’s a Bertolt Brecht play, The Life of Galileo, where one of the sort of secondary characters talks about Galileo’s heroism and his response from Brecht’s phrasing of it is that pity the one that needs the heroes because it means something’s wrong, and you need the heroism, right? Ideally, things work. So flip that thinking around. And if you have what we had, which is a bunch of things — the elite disconnect, the divergence of trajectories of this country, where the stock market can be doing great, even as the bottom 20 percent, 40 percent of the country is not increasing their wealth, the way that the Republican Party can keep sort of losing the popular vote and have a good shot at the presidency nonetheless, the Iraq War that we just don’t ever talk about anymore as if it never happened. And there’s all these things that have led to the moment in which you got Donald Trump. If you don’t have all of those things, you don’t have Donald Trump having a shot at this. So, instead of trying to figure out who the next Donald Trump might be, I’m kind of like, I don’t care who it might be. I just want to make sure there’s no circumstances under which that person, he or she — and if I had to bet, I would bet it would be a woman. I don’t want them to have a shot. It’s not some sort of casting call where we’re trying to find the next star, or anti-star in this case. It’s you’re kind of looking at the world and thinking, how do we make sure nobody has a chance at this? Because if the ground is amenable to it, somebody will rise to the occasion, unfortunately, or go sink to the occasion, whatever you want to call it. And that, I thought, was more important. And when people point out to me all the things that are unique about Donald Trump, I point to them all the things that are extremely common in history about elite failure, right? Donald Trump might have been unique. But having elite failure have catastrophic consequences for your country or your civilization is very common in history. And so that’s the part you need to worry about, rather than thinking, will the next one you have a show in NBC or not? Fine, that’s specific. But the pattern, it’s kind of like pandemics. You don’t know what the next pandemic will necessarily be. But you know you’re going to have pandemics. If you look at human history, they’re there all the time. They happen all the time. So while you might not be able to exactly predict what the next pathogen that’s going to threaten us will be, the flu or a coronavirus or something else, you know that you need the infrastructure to whack it as fast as it tries to appear. Otherwise, you’re going to get hit. It’s the same thing. So the place we were at in 2016 was conducive to this specific threat. If that hadn’t happened, I think we would have had a version of it in 2020.
I think that’s right. And one thing that strikes me as very underplayed in the analysis here is the way in which the systems here were changing. And we knew some of them, but I don’t think people have taken them seriously enough. And in particular, what was changing was that for a variety of different reasons, the base of the two parties, but in this case, particularly the Republican Party, was developing more power to demand and get what it wanted. It was developing more power all over information because social media and other things have taken apart the monopoly of information that a few media companies and parties had. Small donor fundraising was changing who could be a viable candidate. Social media was changing what kinds of things work. The more competitive media environment through Fox News and Breitbart was changing the ways in which different political actors saw what the conservative base wanted. That was much more of a market in giving conservative voters the kinds of media and information they’re looking for. And Donald Trump just went out and market tested. I mean, he market tested with birtherism, with immigration. He saw that the things that were motivating base Republicans were really different than what Paul Ryan wanted them to be, were really different than what Mitt Romney wanted them to be. And he was summoned, right? Donald Trump, in his own ways, was a response to the market. And something that is scary about that then is, if you take that seriously and you listen to what Republicans are saying and doing now, you get somewhere very concerning. You wrote a really chilling line in a recent piece that I’m going to read here. “What people believe they should do is how they will eventually act. This is especially important because institutions, the courts, the political parties, the elections offices, the legislatures are merely people collectively deciding to act in a particular matter. When people change their mind about the rules, those will be the new rules.” The Republican Party has very clearly changed its mind about the rules of how elections should work. And it was the case that key people in the Republican Party who are in positions of authority were operating by the old rules. But the G.O.P. is trying to replace them right now. They are trying to get the people in those positions for next time to be in alignment with where the party has moved with the sort of post-Donald Trump idea that elections are legitimate only when we win them, rather than the pre-Donald Trump idea that elections are legitimate if they were done according to the rules set in advance. So how do you see this playing out, looking at the G.O.P. as a system?
Right, so the problem we have here is that this sort of overturning of the election attempt, which was kind of incompetent and ridiculous, but also deadly and very serious, came at the heels of increasing and persistent minority rule in this country. And that’s the problem. We have multiple things coming together. One of them is the Senate is structurally a minority rule. And traditionally, that didn’t correspond to an exact political line, right? The Democrats had some rural states. And the Republicans had some rural states. But it has since solidified into Republicans have the small white states, which means that they basically have enormous control over the Senate without a majority. They have used that to affirm their control over another branch of the government, which is the judiciary. We saw this when Obama tried to appoint a Supreme Court justice. He couldn’t even get a hearing. He didn’t even get downvoted. He couldn’t even get a hearing. But the same Senate could get another Supreme Court justice they liked in a couple of weeks right before the election. So they have used their Senate control through the minority — they don’t have numbers, but they just have the senators — to extend their control over another branch of government. And through gerrymandering, they are getting disproportionate control of the Congress. Plus, the gerrymandering means that since the seats aren’t competitive, the real competition is in the primary, right? It’s not like which party is going to get it. It’s which party’s candidate. And that means a small group of mobilized extremists can take over a primary much more easily. On top of that, you have cases in North Carolina, Wisconsin, and other places where when they do lose statewide office, which they can’t gerrymander, like the governor’s office, the lame duck state legislature, which they have gerrymandered, will go in and take away powers the previous Republican governor had, right? So even when you win elections, they go in using the gerrymander. It’s just this way the Senate has been used to take over the Judiciary. You see these gerrymandered state legislatures kneecap statewide offices that they can’t gerrymander and they lose. So you’ve seen this in a lot of places. It’s not one place. It’s not two. They end right after they lost the presidency. This time, you’re seeing, I believe, in Wisconsin and a bunch of other states, they now want to gerrymander their electoral college votes to try to give advantage. So these are not — this is like what we call democratic backsliding. In any other country, we would recognize this as a grave threat to the idea of free and fair elections and representation. But that’s how we’ve been doing it. And I didn’t even get into voter suppression, right? So this is at every level, we have non-representative rules entrenching itself as an idea and as an ideal and as sort of legitimate. We almost need the political scientists to study our eroded democracy the way they would have done it if it were any other place. So for the Democrats, it creates its own kinds of dysfunctions — difficulty holding their own coalition, the frustration, the detachment of parts of the wings of the party because you’re not allowed to govern. And they’re like, it becomes harder to mobilize people when there’s no outcome to all the efforts that you put in, and there’s despair. So I know I painted a difficult picture, but I think it’s really important to be realistic about how terrible things are, so that we can kind of go back to the crucial idea here in that people should be represented in government. And government should be accountable to the people that the government’s supposed to represent. Not to the wealthy, not to the minority that just happens to have structural advantage, not to this, not to that. So there’s this fundamental idea that is being undermined. And you don’t have rest of liberal democracy. I mean, liberal democracy isn’t just that. But if you don’t have that part, you can’t even start fixing all the other things that one is supposed to have to make a liberal democracy healthy.
Yeah, I always think about this in terms of threshold questions. There’s so much we debate in terms of this policy or that policy or this messaging decision or that strategic decision. But then there are threshold questions. And if the system simply doesn’t work, if it doesn’t work to create the conditions under which those other questions can be handled well, then it almost doesn’t matter what answers you come to. Because you’ll fail at that threshold. I think it’s a good place to actually to come to the final section here, which we call Recommendation Engine, which I ask you for a couple book, movie, in one case here, recipe recommendations on things that I think you’ll be able to guide us a bit. Are you ready?
OK, go for it.
All right, what is the best book, in your opinion, about systems thinking?
One of my favorite books I think is “Normal Accidents” by Charles Perrow, which is living with high risk technologies. And it’s about sort of accidents like Three Mile Island. But it’s really a nice sort of example of how things interact with each other. There’s a lot of concepts there about how things interact with each other in complex systems. And it’s looking specifically at systems that have potential catastrophic outcomes, but you don’t have to apply it just to that. You can apply that kind of thinking to a lot of things. And in fact, ideally, you’d have a field called systems thinking and how you think about these big systems, but you don’t.
Don’t some people say they’re in systems thinking, the field, right? Aren’t there people who teach that? I like Sidney Dekker, who comes out of safety work. He has a good book called “Drift into Failure.” But don’t some people say they’re complex systems thinkers and have a whole subfield about that?
Right, but there’s no academic departments about it. Complex systems thinking is another term for it. I learned a lot, for example, from ecology and biology. So what I’m saying, I get the concepts from across many fields because ecological systems have a lot of complex systems examples. I’ve always read about epidemics and pandemics, so it wasn’t like I was completely new to it. But I had read about them the way I read about ecology in that it’s very good for thinking and teaching systems thinking, right? I would use them as examples in my own classes to try to get my students to think about systems and complex systems. So it’s like distributed across many fields. It’s almost like the scientific method kind of thing, I think, but for the world.
What’s your favorite work of science fiction?
Now that one is easy because I’m an Ursula Le Guin fan. And hands down, “The Dispossessed” has always been my favorite work of science fiction, partly because I read it as one of the first books I read after learning English, learning to read in English. So I just love that kind of ambivalent utopia, as I believe that’s the subtitle for it. But it’s really interesting and about how do you try to change things, and how does it not work, and how does it work, and all those sort of complicated, fun things that I like thinking about.
It’s such a good book. If I’ve never read much sociology, which thinker shall I start with?
Oh, that’s a hard one. So sociology is traditionally described to have three founders, which is Weber, Durkheim, Marx. But they’re not the best writers, and they’re not all correct. So but they are really foundational thinkers there. But since then, we’ve had a lot of sort of important contributions. We have a very rich American tradition, so W.E.B. Dubois is an American sociologist and very important civil rights leader activist as well for understanding this country, I think. So you kind of have to — this is part of the problem, is, I think one of the things that sometimes I try to communicate is that I do what I do, I think, best partly because I don’t have too many favorites. Rather, I read across the board a lot, so it’s very hard for me to say, here’s the good one, because I’m going to read 100 and try to get what I can out of each one of them. So I can’t give a single current one, but I think even just reading some of the introduction to sociological theory kind of books, which kind of try to bring that together, a lot of them are very good and kind of way of thinking about the world that I think is useful because it’s not just about humans as individuals, but humans in social groups and institutions, which is how we live.
Final question here, which is, what’s your favorite vegetarian Turkish food?
Ha. So OK, I’ll give you a couple of things because it depends on if people want to cook them. So the Turkish cuisine has a whole range of chilled olive oil dishes that are all vegetarian. And you can do them with green beans. You can do them with eggplants. There’s all these versions of them that are not hard to do and that are completely outside of the Western experience because it’s a vegetable dish that is cold and cooked in olive oil. So you cook it, it will have green beans and tomato and tomato paste. And then you chill it, and you eat the cold version very often with bread. And it’s absolutely doable. It’s not something you need to kind of be very good at to cook. Another thing we have is lentil meatballs, which are vegetarian and that are really easy to make. Now, if you are really good at it, you would make the stuffed grapevine leaves, the dolmas. Those are also in Turkish cuisine, as they are in many other cuisines, like Greeks and Armenians. And sort of Middle Eastern cuisines have versions of it. It’s usually fairly similar. Those are difficult. So I love those, too, but I don’t try to make them because they’re a challenge to find the right grapevine leaf and do it and very time consuming. So that’s why I said a couple because if you want to cook, there’s a whole range of amazing, easy to make Turkish vegetarian food that is great. And then there’s the difficult ones, which I try to buy.
That’s probably always good advice. This has been wonderful. Zeynep Tufekci, thank you very much.
Thank you for the conversation. [MUSIC PLAYING]
Thank you to Zeynep Tufekci for being here. Thank you to all of you for being here. As always, if you enjoyed the show, please give us a rating at wherever you are listening to your podcasts. It helps us jump up in the algorithm. And as for building up the new show, it really does make a difference. “The Ezra Klein Show” is a production of New York Times Opinion. It is produced by Roge Karma and Jeff Geld; fact-checked by Michelle Harris; original music by Isaac Jones; and mixing by Jeff Geld. [MUSIC PLAYING]