I wanted to go back to something I wrote about in my last post. I devoted an extended portion of my last blog post dwelling on the CDC’s Vaccine distribution and some of its strategy's emerging critiques. I dwelt on it because as I finished writing the post, the controversy was bubbling up that morning in my news outlets/Twitter feed. I thought it dovetailed nicely with the subject I was exploring in general.
As it turned out, when I hit send, the discussion over the CDC guidelines was only the beginning of the discourse on the subject over the weekend, including a lot of pushback to the position I was pushing on last Friday. This made me question whether I should have written about said controversy: after all, I very much think the way that discourse unfolds on Twitter is toxic and addicting, and for that reason, I literally try to block my access to Twitter on my phone and block twitter on all my devices during the week. I get RSS feeds of different Twitter accounts to an app on my phone/CPU and uses that to stay informed without getting addicted. I purposely like writing in this format (email/blogging) because it is a more “slow-food” style of reading/writing, allowing me to reflect for longer periods of time before I put stuff out there. It also helps that I have a day job and a family, and it can take anywhere from days to months for an idea that pops into my head to become a fully formed blog post (wildfires part 3 is still a thing I will publish at some point!). So I regretted somewhat having my blog post last week come out mid-news cycle.
That said, looking back, I think much of what I posted holds up to scrutiny. I based my reading on three people whose work I follow closely: Nate Silver, Matt Yglesias, Zeynep Tufecki. None of the three are public health experts, but I have appreciated their commentary as public intellectuals on COVID. On the subject of Vaccine distributions, all three were making a similar argument (which I repeated in my post) that it made more sense to focus on getting the vaccine to the elderly over and above other high-risk categories. All three got considerable blowback over the weekend, mostly because of Nate Silver tweeting in a somewhat defiant way:
It appears they got some things wrong: it appears they misinterpreted the CDC’s published models on how different vaccination strategies will affect death tolls. Part of this has to do with assumptions that you make about how easily spread the disease is by those who have had vaccinations. If you assume that vaccines only stop illness but not transmission, we clearly should vaccinate the elderly first. If you assume that vaccines do stop transmission, vaccinating essential workers alongside the elderly ends up about the same level because you start to slow the spread, which does just as much to mitigate deaths in the long-run. Much of the argument about putting the elderly first did not really dive into these nuances, which (probably fairly) gave much fodder to the critics who said Tufecki/Yglesias/Silver didn’t know what they were talking about and shouldn’t be questioning the experts.
That said, the CDC over the weekend shifted their protocols, giving greater priority to people over the age of 75, while tightening up the definition of “essential” workers to just “frontline” workers:
I doubt that this discourse I cited influenced this discussion: all indications are the CDC panel was already moving in this direction. The spotlight may have hastened, rather than changed, a decision that was coming down the pike anyway. But I think it is worth taking a step back and using this as a frame to examine what has become one of the most popular slogans, especially on the left, during the pandemic “trust the experts.”
I have taken issue with this slogan. Of course, I 100% agree that we should be listening to the opinions and input from the public health community. After all, this is a public health crisis! But being an expert in public health does not necessarily make you best at figuring out how the whole of society (and most crucial to this conversation, how the apparatus of government) should respond to a pandemic. To determine the course that society should chart requires synthesizing conclusions across many different domains, of which public health is the most important. But decision-makers also have to consider economics, business, sociology, politics, psychology, communication, and numerous other fields who will be impacted, and thus, need to contribute to the conversation. I agree with Ross Douthat when he says the following:
One of many regrettable features of the Trump era is the way that the president’s lies and conspiracy theories have seemed to vindicate some of his opponents’ most fatuous slogans. I have in mind, in particular, the claim that has echoed through the liberal side of coronavirus-era debates — that the key to sound leadership in a pandemic is just to follow the science, to trust science and scientists, to do what experts suggest instead of letting mere grubby politics determine your response.
Trump made this slogan powerful by conspicuously disdaining expertise and indulging marginal experts who told him what he desired to hear — that the virus isn’t so bad, that life should just go back to normal, usually with dubious statistical analysis to back up that conclusion. And to the extent that trust the science just means that Dr. Anthony Fauci is a better guide to epidemiological trends than someone the president liked on cable news, then it’s a sound and unobjectionable idea.
But for many crucial decisions of the last year, that unobjectionable version of trust the science didn’t get you very far. And when it had more sweeping implications, what the slogan implied was often much more dubious: a deference to the science bureaucracy during a crisis when bureaucratic norms needed to give way; an attempt by para-scientific enterprises to trade on (or trade away) science’s credibility for the sake of political agendas; and an abdication by elected officials of responsibility for decisions that are fundamentally political in nature.
I think this is exactly right: while I think political leaders should have the humility to listen to experts with a narrow domain, it is wrong to assume that anyone group of answers will give us the right answers all of the time. There is fascinating literature on this subject of expertise demonstrating this. It is helpful to simplify it by saying that there are two kinds of problems in the world: “Kind” problems and “Wicked” problems. Kind problems operate in a relatively straightforward world, where the rules are the same every time and can quickly determine if you were right or wrong. Math problems are kind, as would be a bookkeeping exercise. These problems lend themselves well to narrow, specialized experts who can quickly use their knowledge to sort through the relevant data and quickly solve the problem. But “Wicked” problems are much harder for specialist experts; because rather than presenting the same way every time, they are dynamic, and often you do not immediately know if you have made the right choice or not. On an individual scale and a society-wide scale, most social problems are “wicked” problems that are so complex and devoid of timely feedback that narrow expertise is not nearly as helpful.
In his book Range, David Epstein talks about how this plays out in sports: for a period of time in the early 2000s, Tiger woods was cited as the paradigmatic reason why kids should specialize in a specific sport early: he was already excelling at golf as a toddler, and went on to be one of the greatest golfers ever (before his terrible real-life judgment derailed him). But Tiger’s success in golf serves as a poor example for most sports because golf is an extremely “kind” environment: a golf swing happens in a controlled environment, and you immediately find out if you have hit a good shot or not. Chess is similar: the board never changes; the pieces never change their movement patterns, so early life specialists do quite well.
But most sports don’t function this way. They are dynamic, ever-changing environments. Thus, research is starting to bear out that the best athletes in most sports are usually not specialists. Still, generalists learn skills from trying different sports, using their varied experience to 1) learn what they are best at while 2) learning skills to help them in other sports. To give a concrete example: while it has been widely noted that while the US is not a great soccer nation, we produce some of the best goalies in the world because unlike in Europe or South America, Americans goalkeepers play other sports (basketball, football, baseball) where they develop their hand/eye coordination alongside more specialized soccer skills like kicking/dribbling, which actually gives them an advantage to people who specialize as soccer goalies from a young age.
Likewise, one of my favorite basketball players in the NBA is Nikola Jokic, who is notable because even though he is a 7-foot tall center, he is one of the best passers in the game (and possibly one of the best in basketball history). Most tall centers in the NBA are narrowly taught to do a handful of things: score near the basket, rebound, and defend the rim. Jokic is so good because he is not overly specialized: he can do all of those things well, AND he can also pass and play on the perimeter. One of the reasons why Jokic developed those skill-sets is that alongside basketball, growing up in Serbia, he also played Water Polo, which involves an even higher-level of passing skill than basketball. When Water Polo coaches watch Jokic play, they note he passes “like a water polo player,” which makes him better basketball than someone who had only played basketball.
This is not just true in sports. In his book, Superforecasters, Phillip Tetlock documents how specialized experts are often terrible at predicting the future, even within their field. When Tetlock ran competitions to see who was best at predicting geopolitical events, he found that the best performers were very knowledgable amateurs, who often beat foreign policy experts and even intelligence experts within the US intelligence agencies. Modeling complex geopolitical events is incredibly “wicked” and fraught with ways to convince yourself that your bad predictions were due to circumstance, not a flawed model. Intellectually curious outsiders, who did not have the firm worldview of experts, were more able to quickly recognize when they made a mistake and update their “mental model” of the world. Experts were much more likely to stick to their priors, even when they were wrong. Tetlock actually found that the more credentials an expert have, the worse they were at predicting the future, presumably because the more easily they could find a way to explain how events fit their worldview. One of Tetlock’s “super-forecasters” described their strategy for learning from experts:
So if I know somebody [on the team] is a subject area expert, I am very, very happy to have access to them, in terms of asking questions and seeing what they dig up. But I’m not going to just say, ‘Okay, the biochemist said a certain drug is likely to come to market, so he must be right,’ Often if you’re an insider, it’s hard to get a good perspective.”
Adam Grant has shown that the majority creative inventions in our age come not from pushing the boundaries of one particular field but rather from taking the insights of one siloed group of research and applying it to another. You can see this play itself out in everything from “Moneyball” in sports to “behavioral economics” in the social sciences. Nobel Laurates are disproportionately like to have an artistic hobby alongside their specialized research area compared to their colleagues. Even basic research into Artificially Intelligence shows that while computers consistently find ways to “out-specialize” humans, it is tough for machines to achieve the same cognitive range as comes naturally to humans.
COVID and Decentralized Information
So what does this mean for COVID? What we have faced over the last 9 months is the definition of wicked: it is dynamic, involves synthesizing across many domains, and we often do not know what is happening or working until weeks after a decision is made. So we should not expect epidemiological specialists to be perfect at synthesizing all of these domains. Neither should we expert economists, or ethicists, to get everything right. On the other hand, this is not to say that generalists will always get it right: they often don’t! These are hard problems; we should expect many generalists to weigh in to get it wrong. We should neither take the strategy of the political Right (which is generally distrustful of technocratic experts) or the political Left (who relies exclusively on technocratic experts). The key is to have a strategy approaching this complex problem requires drawing on both insider and outsider perspectives. Experts have a key role in policy-making, while curious generalists have a different role to play.
To me, this really drives home the importance of having an open informational ecosystem. In the early days of the pandemic, there was some dialog about how authoritarian countries are much better set up to deal with this pandemic than democracies are. While the death toll in the US and Western Europe could serve as a proof point for this theory, I think this instinct is wrong. For one, as I laid out in my last two posts, I think the US failures have to do with problems with our governance and social trust, not our democratic nature. On top of that, the best responses (Taiwan, South Korea, New Zealand) involve some combination of democratic institutions alongside competent states and high social trust. Taiwan specifically has been a paragon of governments working with decentralized social movements, not against them. It also ignores that almost every innovation in fighting the pandemic was made possible by open informational ecosystems:
The original information we got about the pandemic in January only came out because of Chinese doctors' heroic efforts to circumvent China's closed informational ecosystem and get key information out to the world.
The WHO denied that asymptomatic carriers could spread COVID for most of the spring.
We only found out about the extended spread of the virus in the US in February because Helen Chu, a researcher at UW, disobeyed an order from the federal agencies and tested the samples in her lab.
Public Health institutions were clear and unambiguous in saying cloth masks did not prevent COVID spread until early April, even when most of East Asia saw a slower COVID transmission rate because of widespread masking. Many prominent people were questioning their reasoning.
Public Health experts originally came out against travel restrictions, both within the US and outside the country, for fear that it would create stigma, even while countries that did restrict travel saw far small and more controlled outbreaks.
Again, my point is not to degrade expertise: COVID was a new disease, and it was tough to get everything right, especially early on. Rather, my whole point is that because it is so hard to get things right, it is vital to foster the challenge of expert opinion from outsiders to sift for the best picture of what is happening. The past 4-5 years have seen much intelligentsia bemoaning all of the internet's downsides. There is increased energy (on both the right and left) around more government control of speech on the internet. Under pressure, Twitter has tried to actively police “misinformation” about COVID, which, has been pointed out, would have included labeling calls on people to wear masks before April as “misinformation.”
I think it is worth noting that rather than exclusively noticing all of the bad opinions and misinformation that proliferate on social media and the internet, it is incredibly vital also to acknowledge that the internet also creates way more good information and good opinions that we can access at our fingertips than ever before. Ben Thompson writes convincingly about this in Stratechery:
the Internet is in fact so resilient: despite the best efforts of gatekeepers, information of all types flows freely. Yes, that includes misinformation, but it also includes extremely valuable information as well; in the case of COVID-19 it will prove to have made a very bad problem slightly better.
This is not to say that the Internet means that everything is going to be ok, either in the world generally or the coronavirus crisis specifically. But once we get through this crisis, it will be worth keeping in mind the story of Twitter and the heroic Seattle Flu Study team: what stopped them from doing critical research was too much centralization of authority and bureaucratic decision-making; what ultimately made their research materially accelerate the response of individuals and companies all over the country was first their bravery and sense of duty, and secondly the fact that on the Internet anyone can publish anything.
Another point drives home is the imperative of having a culture that makes it OK to have these debates and disagreements in public. It seems to me that part of the culture of public health officials is the constant worry about how their statements will be interpreted by the public, to the point where they either:
Don’t voice legitimate disagreement within the field.
OR
Don’t give an honest assessment of the situation for fear of how it might be interpreted.
When Zeynep Tufecki wrote a piece for the NY Times in mid-March arguing for universal masking (against CDC guidance at the time), she wrote about the response she got from the medical community:
I got lots of doctors, medical people, epidemiologists, people from our government, other governments, just writing me, thanking me, telling me that they just didn’t dare say this, they couldn’t go against the CDC. The people in hospitals couldn’t even talk about the lack of masks in their hospitals and it was just against the idea of what they do. It’s their field, it’s their authority and they couldn’t just go out and contradict that without having the conversation change first. So instead of a backlash, I’m now in this very interesting state where I have an inbox full of people with the correct degrees and correct titles thanking me for just starting this conversation
This does not invalidate the good work that public health officials have and are doing. We would be in rough shape if we did not have the expertise of public health officials! But seemingly, too often, the profession has failed to reckon with the reality of an internet-based information ecosystem, where people expect a level of transparency from authorities. Experts should be willing to debate in public, where you can really have a true exchange of ideas for the public to see, rather than to toe the party line and have people lose all faith as they question you behind your back.
It is also worth pointing out that much of the energy directed at getting people not to question experts is done by non-experts and less well-known public health professionals. Tufecki wrote about how much of this is a status game that online discourse can create:
On social media, where there was a wide range of responses, but a more noticeable “this has no scientific merit/how dare you as an outsider” reaction compared to when I reached out to experts within the field.
One of the reasons for this discrepancy is that, on social media, the dynamic for in-group status assertions and status competition is strong—and getting stronger for this topic as the pandemic rages on. I’m no stranger to this dynamic, as it is something that’s very common among social movements (something I’ve studied at length) and, well, pretty much any human group. Who’s in and who’s out of the group is a key force for group-based species like ours and thus “stay-in-your-lane” assertions and wagon-circling against criticisms from perceived outsiders is forceful in any human group or profession. As the pandemic progresses, and as the field feels more and more under attack (and much of it quite unfair and terrible) the dynamic strengthens, often to the detriment of the field.
Many of those who are doing the most strong “wagon-circling” are not themselves experts, but people who want to get attention online. One easy way to do that within your peer group is to attack the outsider and see everyone on your side give you praise (or likes and retweets). Of course, this dynamic is not helpful for genuine dialog and intellectual curiosity, but it is great for status-seeking!
And lastly, for this all to work, we need better politicians. To truly make the complex decisions that they have to make, drawing on many different domains as true generalists, we need politicians who haven’t spent their whole life being specialists in one skill: marketing themselves and winning elections. It should concern us that our news ecosystem elevates politicians who are good at social media and cable news hits and not actual governing. Donald Trump’s one clear skill in life is marketing himself. Alexandria Ocasio-Cortez is probably the most famous congressperson in America, even though she was also one of the least productive in terms of legislating this past term:
Similarly, GOP Congressman Matt Gaetz of Florida literally said:
It’s impossible to get canceled if you’re on every channel. Why raise money to advertise on the news channels when I can make the news? And if you aren’t making news, you aren’t governing.
For the record, that comment by Gaetz makes no sense: almost none of the hard work of governing involves being on the news, but a lot of winning elections and running for higher office involves being on the news. This has always been true to some degree, but I worry that it has gotten worse in today’s more competitive and specialized world. We need to elevate people into public office who actually are competent generalists who have other skills other than selling themselves they can draw on when governing. I do not think it is a coincidence that one of the most successful politicians over the last 15 years in the world is Angela Merkel in Germany, who spent her early life as a Chemical and an Engineer, who then entered politics at 45. Not all who take this path to politics will succeed, of course, but I do think we as a public need to really think long and hard about how we evaluate our politicians along these lines.