If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:

Or download the MP3


When people in the US think of World War II, they think of Pearl Harbor, or Normandy or Hiroshima. What most people don’t realize is that the war had been already going on for two years when Pearl Harbor happened. During that time, the UK stood alone against Germany, and the situation looked grim. Even once the US had entered the war things still mostly went badly for them. They were just barely holding their own in North Africa against Rommel, and on the other side of the world there was the disastrous fall of Singapore. But finally nearly a year after the US had entered the war, and three years after it had started, the British finally got their first decisive win at the 2nd Battle of El Alamein. (AL-a-main)

Shortly after this victory Churchill was giving a speech, and in reference to this battle and the turning point it represented he said:

Now this is not the end. It is not even the beginning of the end. but it is, perhaps, the end of the beginning.

I bring this quote up because it feels like a decent description of where we are with COVID. It is not the end, or even the beginning of the end, but I think once the omicron surge passes that it will, perhaps, be the end of the beginning. As a consequence I thought it might be time to do a post about the beginning of the pandemic, both because I think we’re at the end of that beginning, but also because it’s an opportunity bring together some insights about the pandemic I recently gleaned from three books:

Nightmare Scenario by: Yasmeen Abutaleb and Damian Paletta

Premonition by: Michael Lewis

Viral by: Matt Ridley and Alina Chan

But first, before I forget, back on the 8th I made a prediction about when omicron would peak. I said that it would peak on the 15th, at around 2,500 new daily cases per million people. So how’d I do? 

It did in fact peak on the 15th at 2,409 cases per million. How’s that for an accurate prediction? 

Hopefully now that I’ve verified my expertise you’re ready to listen to me on all the other things I have to say about COVID. Speaking of other things, I obviously can’t cover everything from three books, or even everything I think is important and interesting. Instead I’ve picked five subjects where I think I have something important and interesting to add. Let’s get started with:

I- Schools

With the latest surge there was a lot of pushback from teachers unions, and even students on the fact that schools were not closing this time around. But it wasn’t just right wing governors that refused to close schools, Biden and democratic politicians were also emphatic that schools should remain open. 

For my own part I think going back to remote learning for a week during the very peak of omicron (which, as you may remember, I called) is fine, particularly if it’s a staffing issue rather than overactive risk management. The big question we’re grappling with now is not whether we should continue doing it, but whether we should have ever done it. Increasingly the answer seems to be that, outside of those first few weeks when information was scarce, it was always a mistake, and a huge mistake at that.

Recently shots were fired on this subject by Nate Silver, the noted statistician, who tweeted:

Suppose you think that school closures were a disastrous, invasion-of-Iraq magnitude (or perhaps greater) policy decision. Shouldn’t that merit some further reflection?

He later clarified that this was not merely a hypothetical, he did in fact think it was a mistake on the magnitude of invading Iraq. As you can imagine responses were all over the map, but I think Jonathan Haidt’s summation of the situation was on the money:

It is now indisputable, and almost undisputed, that the year and a quarter of virtual school imposed devastating consequences on the students who endured it. Studies have found that virtual school left students nearly half a year behind pace… learning loss falling disproportionately on low-income, Latino, and Black students…a million students functionally dropped out of school…caused a mental health “state of emergency,” according to the American Academy of Pediatrics. The damage…will be irrecoverable.

 

It is nearly as clear that these measures did little to contain the pandemic… 

 

By the tail end of spring 2020, it was becoming reasonably clear both that remote education was failing badly and that schools could be reopened safely.

 

What happened next was truly disturbing: The left by and large rejected this evidence. Progressives were instead carried along by two predominant impulses. One was a zero-COVID policy that refused to weigh the trade-off of any measure that could even plausibly claim to suppress the pandemic. The other was deference to teachers unions…

 

It is always easier to diagnose these pathologies when they are taking place on the other side. You’ve probably seen the raft of papers showing how vaccine uptake correlates with Democratic voting and COVID deaths correlate with Republican voting. Perhaps you have marveled at the spectacle of Republican elites actively harming their own audience. But the same thing Fox News hosts were doing to their elderly supporters, progressive activists were doing to their side’s young ones.

For me it mostly boils down to that last bit. If we’re faced with the horrible task of apportioning harm, and unfortunately we are. Then it seems logical to not apportion a greater share of it to the young. And not only is it logical, but viscerally I, and I think most people, recoil from the idea of sacrificing the boundless potential of the young for the limited potential of the old. (Which is harsh to say but ultimately true.)

So how did we end up doing just that?

Here’s where I can add something to the discussion, to do that I need to take you back to 2005. President Bush had just read The Great Influenza by John M. Barry about the Spanish Flu and decided the government needed to pursue pandemic preparedness. This was the first big effort taken by the federal government and they had to develop a lot of stuff from scratch. One of the things they lacked was a good model. The Premonition has a whole cute story about how the best model was developed by a child for a science fair project, which only starts to make sense when you realize her dad worked at Sandia National Labs. I don’t have the space to go into that story, the key point is once they started messing with this model they realized that they could significantly reduce transmission through non-pharmaceutical measures. This was the birth of social distancing. And if distance mattered for transmission, then the first places the model said you needed to shut down were schools. Nobody gets packed in tighter than school children.

I’m not sure how much this model drove school closures in the early days of the pandemic. I assume it had to have had some impact, but I don’t remember it ever getting mentioned. And of course even if the model was used to make this decision, it should have been updated as more information came in. Which is to say even if the model was used to justify closing schools I don’t think it could be used to justify keeping them closed.

As one final point as long as we’re on the subject of what sort of things were recommended in the years before the pandemic. When the group President Bush put together went to the CDC and recommended closing down things as per the model the CDC pushed back and said it would never work, people wouldn’t tolerate it. And to a certain extent that’s exactly what we’re seeing. Also the period the group recommended for shutting things down was relatively short. Which is also something we haven’t exactly done (particularly with schools).

II- Republicans vs. Democrats

I’m already a third of the way through the expected length of this post and I’ve only covered 1 of the 5 items on my list. I guess I’ll need to exercise more brevity going forward.

Like so many things COVID became a tribal issue with the blue tribe on the left and the red tribe on the right. And members of each tribe want to know that their tribe was the righteous one, while the other tribe was the wicked one.

Nightmare Scenario was written with the goal of proving that the Republicans, and particularly Trump were the wicked ones. Having also read Premonition as well as countless blog posts and tweets, I think this fixation blinded the authors of Scenario to the even greater failings of the FDA and CDC, institutions which are supposed to be the ones preparing for and dealing with emergencies like this. Which is to say while it would have been nice to have a President who’s great at handling a global pandemic, it shouldn’t be too surprising when we don’t. The president has countless jobs, the CDC only has a couple and the most important of those is handling a global pandemic, and on this count they were abject failures.

This discussion of tribalism and the CDC takes us to the story of Charity Dean, one of the three stars of The Premonition. In the years leading up to the pandemic, Charity was a public health official, a very talented and dedicated one, exactly the kind of person you wanted in charge during a pandemic. The book relates numerous stories about her, but two are germane here.

First, in the months immediately before the pandemic Charity had risen to be the number two  person in the state of California for public health, and when the top position, Director of the California Department of Public Health, opened up in 2019, Charity assumed she would be appointed. She was not, and she discovered later that as a white woman with blond hair she wasn’t even in the running, despite being the number two person. Instead Dr. Sonia Angell was given the job. Her primary qualification appeared to be the fact that she was Latina. In other words it was an affirmative action hire. Had it been a successful example of affirmative action, raising someone with the necessary skills who had previously been overlooked, then we wouldn’t be talking about Dr. Angell, but that was not the case. Lewis described her as being “monstrously incompetent”, so much so that she ended up abruptly resigning in August of 2020.

Republican ideology gave us Trump, and he obviously made some big mistakes (more on that later) but the presidency is not our primary line of defense. The public health bureaucracy is, and Democratic ideology undermined that, and not merely in the case of Dr. Angell. I think one of the biggest hits the public health authorities took was when they backpedaled all their guidance on gatherings when people started to protest in the wake of George Floyd’s murder.

The second reason I wanted to bring up Charity Dean was to highlight her views on the CDC, since you might need further convincing. Lewis describes how Dean, through her connection with other characters in the book, was finally given the opportunity to put together a plan for how to deal with the pandemic. Someone, in an effort to help, added a small role for the CDC in her plan. Charity wrote back:

No…The single most important part of this plan is IT IS NOT RUN BY THE CDC… The entity/agency/figurehead leading this must be a Churchill not a Chamberlain.”

III- The FDA and CDC

So what made the FDA and CDC so bad? What did they do that they shouldn’t have and what should they have done that they didn’t? 

To explain that I’ll first need to take you on a brief aside. Back in 2014 Scott Alexander of Slate Star Codex decided to take on the effects of marijuana legalization, in a post titled, Marijuana Much More Than You Wanted to Know. As is usual with his “Much More Than You Wanted to Know” posts he went really deep, looking at the issue from every angle you could imagine. But despite pot affecting everything from IQ to organized crime, one consideration overwhelmed everything else: traffic fatalities. Does pot increase the number of traffic fatalities because more people drive under the influence? Or does it decrease the number of traffic fatalities because people substitute pot for alcohol and while driving under the influence of pot is bad, it’s vastly better than driving drunk? In the final analysis death is so bad, and traffic fatalities are such a big cause of death that even changing it by a tiny percentage one way or the other overwhelms every other consequence the legalization of marijuana has on society.

This is the position the FDA and CDC were in. They probably did a lot of things right. There are probably many criticisms which are unjustified. But in the final analysis the consequences of the few bad things they did were so bad they overwhelmed all the rest.

The reason these things were so bad is that when you’re talking about exponential growth, what happens at the very beginning of that curve has a far more dramatic impact than anything that happens later on, after the curve is established. I’ve seen evidence that they did lots of things wrong at pivotal points where speed was essential, but here are the two biggest examples in my opinion, one from each agency:

For the CDC it was obviously creating the initial test. If you haven’t heard this story Buzzfeed actually has a great investigative piece on it. But basically they refused to rely on tests that had already been created in other countries, they refused to let private companies market tests they had developed, the supply of their test was horribly limited and to top it all off it didn’t work! Certainly even with good testing there was no way for the US to avoid the pandemic, but we seeded it far more deeply and more broadly than necessary because in the very earliest days we couldn’t test for it.

For the FDA it was the vaccine, and in some respects this is less forgivable. The CDC might plausibly claim that, early on, it wasn’t really clear how important the test was going to be, the FDA had no such excuse when it came to the vaccine, everyone knew exactly how important it was. And yet the FDA’s Vaccine Advisory Committee seemed almost leisurely in their approval approach. The vaccine was ready for approval by November 20th, but they didn’t meet until December 10th, nearly three weeks later. What possible reason could there be for not meeting on November 21st or the evening of November 20th? 

As I said, I think these are the biggest mistakes, but they are by no means the only mistakes made by the FDA and CDC. In fact when it comes to the vaccines there may be an even bigger scandal, but for discussing that, let’s turn to: 

IV- Trump and the Vaccine

The same thing I said about the CDC and the FDA could be applied in reverse to Trump, and his team. However many mistakes they may have made, in the final analysis, the speed of vaccine development was going to be the measurement that mattered the most. And in this respect everyone basically agrees that Operation Warp Speed did an amazing job. The obvious objection for those not inclined to give Trump any credit is that he played only a very small part in the operation. That’s a fair point, but in considering it we should remember what Kennedy said after the Bay of Pigs, “Victory has 100 fathers while defeat is an orphan.” Lots of people now want to take credit for starting or managing or having the idea for Operation Warp Speed, but if it had failed everyone would have blamed Trump. This being the case, should he not therefore get some of the credit?

Even Nightmare Scenario which was written specifically to savage Trump, admitted that this sort of “hail mary” approach was something the Trump administration was actually pretty good at. The book’s chief criticism of such efforts was not that they were ineffective, but rather that by routing around the normal procurement process, such efforts wasted money. This complaint about the government wasting money made me laugh, a lot.  Other more serious criticisms include duplication and a lack of focus, and there was certainly a lot of that, but in the end we got the vaccine months and months sooner than anyone thought possible. Though we could have gotten it even sooner and therein lies the potential scandal. 

There’s lots of evidence of people working to move the announcement of a successful vaccine from October to November, enough that the fact of it happening really isn’t in question, what’s in question is why?

The most benign explanation is these people were worried about anti-vaxxers and vaccine hesitancy, and they wanted to make sure the safety data was ironclad. 

However, lots of people (including the aforementioned Nate Silver) find the timing to be very suspicious. The fact that moving it from October to November happened to move it from before the election to after the election makes it look like it was largely motivated by a chance to hurt Trump’s reelection efforts.

Other people don’t care about the motivation, they just think that it was a bad trade off, that the delay was never going to have that much of an impact on vaccine hesitancy and anti-vaxxers, but it was always going to have a huge effect on the number of people who died. Who cares what the motivation was, the decision to delay ended up killing thousands of people.

The problem with much of the discussion is that it ends up being broken into a binary. Either it was totally reasonable for the FDA to wait 60 days to see about side effects or it was monstrous and politically motivated. And what’s lacking in both the current discussion and the toolkit of bureaucratic capabilities is any capacity for nuance. Could we have authorized it for high risk old people in October? People who felt that the protection from not dying was worth the potential that there might be unknown side effects?

V- The True Beginnings of the Pandemic 

Here we arrive at the ultimate example of the “traffic fatality effect”. Despite everything that has been said so far, when it comes to the pandemic all other considerations are secondary to the question of where the virus came from, because if it didn’t happen in the first place every other issue is moot. Thus far I have mostly drawn from The Premonition and Nightmare Scenario. This section is going to draw almost entirely from Viral. With a few other quotes here and there. Speaking of which, here’s Matthew Yglesias to start us off with a quote saying essentially the same thing:

(Or not… I had this quote in my notes I thought it was from Yglesias but it wasn’t, and now I don’t know where I got it… That said I still think it’s 100% true.)

I’m not saying it was definitely a lab leak. I’m saying that answering that question is one of the most important tasks of the post-mortem, and anyone who says it definitely wasn’t a lab leak is not trying to answer it they’re engaged in the culture war.

Before we get to a discussion of the actual evidence, this is a perfect example of what I talk about over and over again in this space. You might think that in a situation where something only has a 5% chance of happening that you should act as if it won’t or if it didn’t. But my argument has always been if the consequences of it happening are bad enough, you should act as if it will or that it did. Applying this the lab leak hypothesis they should have never been conducting coronavirus research in a biosafety level 2 lab (4 is the highest) and going forward we should probably stop such research all together, certainly anything that involves gain of function. But that’s precisely what we’re NOT doing. To quote from Yglesias again:

By contrast, on something like Covid-19’s origins, we’ve had a decent amount of coverage of the lab leak controversy but essentially no coverage of what is being done to prevent future lab leaks (basically nothing) or to prevent future zoonotic crossover events (again, nothing). But the reason these are our two contenders is that as far as we know, these are the routes through which new viruses emerge…either way, we’re not doing anything to counter either route for transmission, and that (shocking! alarming! insane!) fact gets way less attention than the latest round of “who’s yelling at whom about masks?”

We should be taking all of these measures even if the chance is only 5%, but having read Viral I would argue that the chance is far higher than 5%. In fact it might be the inverse of that, I might put the odds at 95% lab leak, 5% zoonotic origin. I’m sure you’re interested in the evidence backing up that assertion. As this post is already running long I’m going to steal Scott Aaronson’s list, and briefly offer my own commentary. And if you’re interested in going deeper you should read his whole review. But this is his list of things Viral proved beyond reasonable doubt, a list I entirely agree with:

Virologists, including at Shi Zhengli’s group at Wuhan Institute of Virology (WIV) and at Peter Daszak’s EcoHealth Alliance, were engaged in unbelievably risky work. 

This gets to the point I was making previously. We were already taking risks we shouldn’t have been. Lab leaks happen all the time, and rather than focusing on the huge negative black swan of potentially creating and releasing a viral pandemic, they were focused on the small gains of having a better viral database. The claim was it would help us fight future pandemics, but:

Even if it didn’t cause the pandemic, the massive effort to collect and enhance bat coronaviruses now appears to have been of dubious value. It did not lead to… useful treatments, vaccines, or mitigation measures, all of which came from other sources.

Another point I make all the time, the benefits of technology are almost always oversold while the potential harms are generally entirely invisible. Particularly once we have done all the obviously beneficial things (modern sanitation) and we’re moving on to things of more dubious value (harvesting exotic viruses and studying them).

There are multiple routes by which SARS-CoV2, or its progenitor, could’ve made its way, otherwise undetected, from the remote bat caves of Yunnan province or some other southern location to the city of Wuhan a thousand miles away, as it has to do in any plausible origin theory. Having said that, the regular Yunnan→Wuhan traffic in scientific samples of precisely these kinds of viruses, sustained over a decade, does stand out a bit! On the infamous coincidence of the pandemic starting practically next door to the world’s center for studying SARS-like coronaviruses, rather than near where the horseshoe bats live in the wild, Chan and Ridley memorably quote Humphrey Bogart’s line from Casablanca: “Of all the gin joints in all the towns in all the world, she walks into mine.”

I had to quote this point in full. It really elegantly encapsulates the whole argument. And what about the wet market?

The seafood market was probably “just” an early superspreader site, rather than the site of the original spillover event. No bats or pangolins at all, and relatively few mammals of any kind, appear to have been sold at that market, and no sign of SARS-CoV2 was ever found in any of the animals despite searching.

My sense is that those who hold on to the zoonotic origin idea, imagine that the wet market was this crazy lawless area where anything could have happened. Actually it was closely monitored, and because of other diseases they were carefully tracking what was sold and when, and as Aaronsson says, none of the potential carriers were even sold at that market.

Most remarkably, Shi and Daszak have increasingly stonewalled, refusing to answer 100% reasonable questions from fellow virologists… They’ve refused to make available a key database of all the viruses WIV had collected, which WIV inexplicably took offline in September 2019. 

If that database had been taken offline in November of 2019, this would be the smoking gun, but even September seems very, very strange. But of course the most inexplicable thing is why it has never been made available since the pandemic started. The whole point of the database was to assist people in fighting pandemics, and when an actual pandemic comes along it’s permanently made unavailable. The list of other strange oversights and evasions is nearly as baffling. And of course on top of it all:

The Chinese regime has been every bit as obstructionist as you might expect: destroying samples, blocking credible investigations, censoring researchers, and preventing journalists from accessing the Mojiang mine.

Sometimes people imagine coverups where there probably aren’t any. Or there is a coverup, but they’re covering for something different than what you imagine. Consequently I try to be somewhat skeptical when an organization or a person is accused of acting in bad faith. The world is complicated, and incompetence is ubiquitious, but for this particular issue there are so many obvious good faiths steps would could be taken but haven’t been:

  • WIV could restore access to the database.
  • China could allow an international team to investigate the cave.
  • Daszak could confirm and explain facts that are already out there.
  • They could stop any remotely dangerous viral research now that it’s been shown that it may have caused significant harm, and it didn’t provide any significant help.

Of course this lack of good faith engagement illustrates the entire problem, not just with finding the actual source of the pandemic, but with everything about the pandemic, and indeed everything about the modern world. We have turned everything into tribal warfare, and the only thing that’s important is that our tribe wins.


Speaking of fracturing into tribes, sometimes I feel like Treebeard, from the Lord of the Rings, when he was asked by Pippin whose side he was on, “Side? I am on nobody’s side, because nobody is on my side…” If you’d like to be on my side with things like the pandemic, and the fragility of modernity more broadly. Consider donating