Pandemic Woo

Here we are part-way through 2020, the year humanity started a war against the common cold, and lost.

“How DARE you, it’s much worse than the flu!” someone howls.

Yes, sure, maybe in a certain context. It’s more contagious than a flu, it’s new so there is no vaccine, and it will lead to a great many lower respiratory tract infections and deaths, possibly more than we have seen in decades. Any virus that jumps to humans from another species has the potential to be quite a bit more dangerous until it evolves toward its attenuated endemic state. But contrary to what prominent experts still say, COVID-19 and all endemic human coronavirus diseases are colds. And although colds are bad, the long-term lethality of every new species is wildly over-estimated.

“You’re not a scientist, you know nothing about how SCIENCE works,” someone else heckles.

“I fucking LOVE science!” contributes another.

Here’s the problem: Science as you know it is dead. Unplug the ventilator because this was it, this was the last straw, there will be no coming back from this. The folks running the show do not know what they are doing, they have made a huge mistake, and the consequences will be too big to ignore. I’ll explore this claim in detail, and I did sit through the quarantine, so now it is only fair that you have to sit through my rant.

Science is dead

We’re witnessing the end state of a toxic culture that punishes disagreement, rewards sycophancy, and worships consensus. I’m an engineer but I do know how this culture works, and I’ve avoided it like… some sort of plague. Being a scientist today is about the worst possible job for someone with more than a teaspoon of curiosity, because most of what you know as “science” is a sham, run by the pettiest gang of bullies on the planet. Disagree with the favored consensus? Good luck applying for the zero projects that will talk to you, loser. Slog through six months of research? Congratulations, your boss will steal every discovery to get tenure.

Thomas Kuhn said basically the same thing in The Structure of Scientific Revolutions. It’s depressing to think that even if Popper was right about how science should work, Kuhn was right about how it actually did work – right up until it died. Which was just now.

Aren’t we talking about a virus?

Of course! The virus.

Respiratory viruses have been mutating and jumping between host species since before humans (or heck, mammals) evolved. Until the original SARS-CoV, there were four known species of coronavirus circulating in humans, and they’re all still around causing common colds year after year. We have no idea how many other species went extinct in the past. Remember that molecular biology is less than a century old, and the tools needed to analyze DNA “really well” have only existed for a few decades. With more humans on Earth, and more animals living alongside humans, viruses might be jumping to humans more frequently than before, but we still barely understand this phenomenon.

As mentioned above, a virus can be much more virulent (deadly, basically) in a new host species right after it jumps to that species. But if something can adapt enough to jump between species, that same flexibility means it can continue evolving, and probably in a less-dangerous direction, because the strains which do not kill or incapacitate their hosts right away are fitter, i.e. the host will have a chance to show up at work and cough onto other hosts. There are lots of similar theories in the literature, but I’ll add that out of seven examples, we have never seen a coronavirus that is both contagious enough to become endemic, and more deadly than a flu. While this new one may well become endemic, in that case I would be very surprised (and we would be in a lot of trouble) if the higher mortality estimates hold.

Personally I don’t think that millions of healthy people will ever drop dead from one of these diseases in the space of a year, like they once did from a flu in 1918. But that’s speculation – stay tuned for more in a bit.

Theory and experiment

This post-modern mess that is contemporary science bothers me so much because I come from an old school, one that doesn’t get along with Kuhn. He argues that science in its normal phase always operates with respect to a dominant paradigm, such as the Geocentric Universe or the Standard Model. Problems inherent to that paradigm will cause disagreement between scientists, and eventually a rebel faction will install a whole new dominant paradigm in a coup. There is no paradigm without problems (see incompleteness), and there is no objective way to judge one paradigm against another. Bad news all around!

If no-one can arbitrate whether a contending paradigm is better than the dominant one, then meaningful scientific progress can only happen by political revolution. And if scientific achievement is the process of winning a popularity contest, then nothing about science makes it special or different. Kuhn’s science is basically plain-old politics with silly customs. This was famously implied by Paul Feyerabend in Against Method, even though some academics have tried to interpret it otherwise.

Before all of the silliness, science actually was special. It was the light of modernism, finally illuminating the demon-haunted world. It was the one institution that was supposed to be immune to human nature, because it was not a government but a method, an idea:

Imagine a theory about how the world works in some small way.
Do an experiment to test your theory. If it fails, discard it.
Repeat until you collect the complete Theory of Everything or you die, whichever comes first.

Karl Popper distilled “science as falsification“, the idea that every meaningful experiment must attempt to disprove a theory, and an early version of the null hypothesis. In 2010 Moshe Vardi criticized the encroachment of numerical methods and computer models into the theory-and-experiment core of the scientific method. I believe strongly that science should be understood as this ideal, because it has no value as anything else.

The reputation of science was built by people who treated it like an idealized method, too. In 1925 religious modernists sold the method to the American public at the Scopes Trial. Prestige accumulated in various wars for most of the 20th century. By Kuhn’s time, the scientific method had been applied (with intent and at unprecedented scale) to help win freedom around the world using technology like radar, metal alloys, atomic bombs, and more.

Any random post-modernist theory couldn’t put a dent in that reputation, but Kuhn’s idea had a profound effect on the behavior of scientists themselves. Many more started cutting corners with the method, hustling their pet theories and paradigms without experiments to support them. Or arguably even worse, substituting computer models for the experiments and pretending that their papers were something other than fiction. We have to admit that Kuhn accurately described the behavior of a community of people. The tragedy is that the average response was not to work on improving that behavior, but to double down and abandon the only reason why anyone cared about the community in the first place.

Whether Kuhn’s vision was a self-fulfilling prophecy or just an inevitable result of natural sociological law, we cannot tell at this time. In either case, science has since become thoroughly post-modern, lost its reputation, and died.

Right, because of a virus!

Not quite. Perhaps you are wondering why this particular episode signals the end, and not the replication crisis, the global warming business, or something else. Well, it’s because this time people have actually gone ahead and done shocking and terrible damage to civilization on the recommendation of the scientists. That damage will be impossible to ignore, it will force us to investigate exactly what went wrong, and we will find out, uh that the wizard behind the curtain has no clothes or whatever.

Surely this can’t be true! The only alternative to radical suppression and mitigation would be utterly devastating, that’s what the president said, right?

Well no, this is almost completely backwards. States and countries are pulling back on their lockdowns, and the peasants are rebelling against quarantine, but we will fail to see the “second wave” that was predicted (although cases may rise again in the winter). While I’m not going to try to unpack a bunch of preprints which are hinting in this direction, it’s basically because we’re closer to herd immunity than prominent experts are interpreting the available data to suggest.

[NOTE: This turned out to be way off. I underestimated both the seasonality and the speed of mutation, meaning we never got to herd immunity for a variant before the next one evaded immunity enough to spread again.]

Meanwhile, the damage done by radical suppression and mitigation will be far worse than those same experts are willing to acknowledge. Let’s review some forms that this damage may take – most references are anecdotal news reports because these effects are not being studied by public health professionals yet:

Many people with heart attacks, strokes, and other medical emergencies are avoiding the hospital because they are afraid that catching COVID-19 is a bigger risk than staying home. Something like a stroke is very sensitive to how quickly it is treated, and these patients are dying much more often when they wait to seek treatment. Many of the survivors will suffer heart damage, brain damage, and shortened lives as well.
Similarly, people who are supposed to have elective procedures like cancer treatments or heart bypass surgery (yes, that’s usually an elective procedure) now have to wait until either hospitals will accept them again, or their condition becomes so bad that the procedure is no longer classified as elective. Delay in these cases will also lead to additional deaths.
Many people who are right now turning to drugs and alcohol in quarantine will relapse or become addicted for the first time. Many of those will then never recover from their addiction, and some will overdose on hard drugs or die of alcohol poisoning.
Many people are committing suicide, both because of economic hardship, and because this is by far the scariest and/or loneliest experience of their lives. Increased suicides will continue after the quarantine is over, because some people will not get their jobs back, and some people will slip into a rut of depression from which they never recover.
Panic attacks can cause real heart damage. In general, stress and fear can be very bad for health.
More people overall may be dying in car crashes. How can this be happening when there is less traffic on the roads? Well, more people are speeding because the roads are empty and the cops are busy arresting people for not wearing face masks!
Most victims of domestic abuse will not be killed by their abusers, but these people are now trapped in quarantine with their abusers and suffering. These people will bear physical and emotional scars for the rest of their lives, some will turn to substance abuse or suicide to deal with their trauma, and some abused children will grow up to continue the cycle of abuse.
Some families that would have managed to stay together otherwise will be stressed to the point of breaking by constant contact. Others will benefit from the time spent together, some families should break up, and many children probably will be conceived during quarantine, so it’s hard to say for sure whether this will be a net-negative.
School-age children in every household are staying home, of course. Many children from lower-income households rely on school-provided meals to stay healthy, and in general the disruption to their education will have downstream effects.
It won’t happen as much in the first world, but there will be surges in diseases like tuberculosis, as well as starvation, in poor countries due to the lockdown. I would bet that these additional deaths alone will at least double the (official) total killed by the virus globally.
As we move further out along this limb, we can guess that some parts of the world will see tyranny and the general breakdown of social order, or even outright war. Almost every country has slipped at least one small notch in this direction with the imposition of questionable and/or unconstitutional lockdowns. Some are already going much further – Hungary’s prime minister basically suspended parliament, for example.

Finally and most controversially, deaths from COVID-19 could end up being greater due to the lockdown and panic. This is for at least three possible reasons:

In several hotspots, nursing homes were compelled to accept recovering COVID-19 patients, due to fear of scarce hospital beds that never materialized (sadly ICU beds are a different story). Apparently the NY government wants you to forget that this happened.
Mitigation can only slow the spread. If the pandemic does not end with a vaccine, roughly the same number of people will be exposed to the virus regardless of mitigation efforts. This means that people not at risk but still in quarantine (mostly the young) will be susceptible or infectious for longer, and it is possible that many more vulnerable people (mostly the old) will therefore be exposed, before herd immunity is achieved.
Most people who are diagnosed with COVID-19 will be overcome with anxiety. As a result their prognosis will be worse than it would be if they thought they had a mild disease. It has even been hypothesized that fear can be the most significant cause of a person’s death. I am not claiming that the severity of COVID-19 is “all in our heads”, but ignoring this factor would be foolish.

Reckoning

Let’s imagine that the effects of radical mitigation do in fact turn out to be net-negative. The immediate question is then: What could we have done differently? How should scientists and policymakers have acted instead?

With the growing powers of hindsight come answers. First and most outrageously, prominent experts made zero effort to communicate the true limits of their understanding, and they should have known better. Most laypeople won’t dig through research papers looking for stated assumptions, but they’re not less intelligent than us, folks. We need to admit that we’re unsure about every one of our assumptions, every time. We need to interrupt and correct reporters when they get things wrong. If we do not do this, ordinary farmer-joe type citizens will be suspicious, and they will be right to be. The confidence game of pretending to know everything is over.

But even if we were looking honestly at the data, how could we have avoided a wave of sick people overwhelming hospitals, without extended lockdown? In that case, as soon as we learned that the elderly and immunocompromised were the people seriously endangered by this virus, we could have quarantined them and gotten to work exposing everyone else as fast as possible. Most professionals backing this idea suggest that it could be done in stages based on risk tiers, which would roughly approximate age groups from young to old (excluding the immunocompromised). Essentially, we could have been racing for herd immunity, against the spread into the most vulnerable populations.

Prominent experts and media people are not recommending this right now, and some have gone further to insult anyone who suggests that this risk could be acceptable and even strategic. These people are cowards. They should, and they soon will, lose all credibility in the domain of public health policy.

The fate of scientific discovery

Let’s take an abbreviated (and slightly unfair) tour through the corpse of the sciences, and conclude with an attempt to solve Kuhn’s conundrum.

Theoretical physics: We need another trillion dollars to build the next-size-up particle accelerator, so that it can probably just tell us the Standard Model works, again. What do the finely-tuned parameters in the Standard Model actually mean, anyway? And what is renormalization, other than a glorified fudge factor? Hey look at this clunky new alternative to String Theory, which also predicts nothing new, testable, and correct.
Applied physics: We don’t get the biggest budget, but we did develop some lasers and rockets which could be useful. Unfortunately however, our research is not likely to yield an answer for the meaning of life.
Astronomy and cosmology: Heyy, we got a bloop! It’s definitely, positively, absolutely, two black holes merging. One is, uhhh, 25 solar masses, and the other is, uhh, 17 solar masses. The data fits!
Chemistry and biology: Actually doing gangbusters over here. We just figured out how to chop up DNA and rearrange it, so the future is going to be terrifying.
Sociology: Kuhn proved that science is a social phenomenon. Also, everything else is a social phenomenon.
Climate science: Science is in the name, you can’t get more science than that! We are currently researching why white Republicans are in denial about it.
Economics: We’re just as much scientists as those theoretical physicists! Why don’t we get a trillion dollars too?

In the not-too-distant future, “science” in the popular understanding will revert to being just a method, and the academic community built around it will evaporate onto internet message boards. Much of the same work will happen, and different paradigms will remain incompatible with one another – Kuhn was right about that. The big prize, that prestige which had been bestowed on the former scientific community, will be claimed by the applied sciences and engineering disciplines, where it belongs. Everything else will be understood as something like “natural philosophy” (which is fine too).

We’re left with a serious problem: All paradigms are incomplete, and so the ideal scientific method must always have discontinuities in the real world. How can any field move from one paradigm to the next without the method falling apart? Are we doomed to flail forever in the eternal darkness of political quagmire?

No! In fact the applied sciences are precisely those which can always be applied, so for them at least there is a solution. We can evaluate every paradigm in the applied sciences, and indeed compare them against one another, by observing whether they are useful. Call this process “paradigm renormalization” if you like:

Before a paradigm is useful, it can be superseded by any other paradigm which is already useful (if there is overlap).
A useful paradigm can in turn only be superseded by another more useful paradigm (if there is overlap).

Look through history, and this same pattern is followed by successful paradigms until the mid-20th century. Ballistics is useful. Fourier analysis is useful. Oxygen theory is useful. Relativity and Quantum Mechanics are both useful. Evolution wasn’t useful for a long time, but today we use it to develop flu vaccines. Utilitarian science is not a brand-new concept, and it bears some similarity to the ideas of Imre Lakatos.

We have an interesting decade ahead of us. Will we actually lose this war against respiratory viruses? New treatments and cultural changes will make a big difference, which is great news. But I’m not so sure that we will be able to beat evolution at its own game anytime soon. Be safe, wash your hands, and enjoy the ride.