Global warming is the first predictive catastrophe, by which I mean a catastrophe that requires action decades before those actions can make an impact. The ability to respond to predictive catastrophes is important. For complex societies to survive, they must be able to deal with complex threats.

There are lots of examples of making changes after a crisis. Deadly North Sea flooding in 1953 led to the Delta Works in Holland. The Tay Rail Bridge collapse in 1879 (wind loading) and the Tacoma Narrows Bridge in 1940 (aeroelastic flutter) led to engineering changes. But in all of these prevention follows a disaster.

But global warming predictions are based on decade-long time series analysis and fluid dynamics simulations. No large public policy decision has ever used these as inputs. And trusting experts to predict the future is scary for us. (More here on politics, experts and fluid dynamics).

Smoking is also a decades-ahead predictive disaster, but with smoking, individuals or families can stop on their own. Smoking shares with climate change a contentious information space – it resisted consensus for a while, in part because of disinformation from big tobacco.

I’ve been trying to write this post for months but it didn’t feel timely or relevant. Then coronavirus happened.

Coronavirus quarantines are predictive

Pandemic isn’t a direct analogy for climate change, and the long lead times are missing, but here’s what’s the same: leaders are forced to balance two harms, the immediate medical risk to life from not quarantining vs long-term economic shockwaves from retail lockdowns.

(With climate change the harms are the short-term economic effect of abandoning cheap energy vs the long-term impact of two degrees, but you get the idea).

In a fast-evolving situation, there are no best practices. Also, we don’t understand yet the differing death rates in China and Italy. Theories include cohabiting elderly relatives and different testing / reporting rates.

Because there’s no way to avoid acting before we have the details, you can see societies testing their ability to do expensive collective prevention. There are public arguments happening in every country now, and the ones that work will permanently enter our vocabulary for governing – not just in epidemics but for all kinds of complex, preventable threats.

Companies and small teams

I worked on a product, long since abandoned to the watery depths, that saw a linear decline that was consistent over months. And the team knew, had always known in fact, what we wanted to do: bootstrap off the first thing and develop more products. Instead, we continued operating the first product, and when nothing worked, we rode it into the ground.

We constantly tweaked our estimates, I guess for morale reasons but IDK. We altered models so we could claim to be operating profitably well below the actual point of profitability. We blamed the decline on measurement changes to convince ourselves it had stabilized.

My point is that human cultures have countless effective tools for continuing with business as usual in almost situation that can’t last.

I have a lot of ‘what is strategy’ conversations and part of it is this – seeing things that haven’t happened yet but logically follow from current events, assuming no change.

In the 70s, to avoid this kind of thing, companies including Shell Oil embraced scenario planning, the practice of creating plans for unlikely events and then investing in those plans according to likelihood and expense.

To a surprising extent, future planning of groups relies on trends and ignores the risk of trends changing. It takes a lot of vision to see something coming. That’s why people like Nassim Taleb and Nouriel Roubini, who correctly called 2008, get a lot of credit, even though there’s always someone predicting catastrophe and a stopped clock is right etc.

To survive complex threats, we need to bring the concepts and vocabulary of black swan / oncoming bus situations into the Overton window.

Might have beens

I read the first half of the Joseph Tainter collapse book and was expecting it to be about external triggers for collapse, and was surprised that he devoted a lot of space to how societies legitimize their social order.

Like we always do in the presence of uncertainty, in the aftermath of covid people will place blame and other people will try to prove that their actions were correct. But there’s no way to discuss correctness without might have beens, counterfactuals, and understanding ‘who knew what when’. And even hindsight isn’t 2020 in these cases.

This whole UK page on covid evidence is about legitimacy – as I read it on 3/20 it doesn’t say what the UK’s policy is, or even contain any arguments about how to go from observational evidence to policy. It does talk a lot about confidence, polling, and transparency.

This year will be a major test of our ability to toe the line between medical & economic danger. The public relations side is a big part of it. Concepts like ‘flatten the curve’, exponential growth and safe gathering size are entering the public consciousness and will be reusable to think about other catastrophes.

Coronavirus is rewriting our social code in the same way bracoviruses rewrote moth DNA to make butterflies.

Hiding & delaying an emergency is another way to handle it, and I suspect we’ll discover that some countries did too much of that. There’s already a censorship conversation happening in China, and they need to take the conversation seriously. (Vs disappearing some folks for speaking out).


See also: privacy, my personal white whale. People who campaign for it are endlessly stumped by the ‘nothing to hide’ crew who is happy to trade privacy for convenience or free TV.

Legislators are trying. California got something on paper during the Cambridge Analytica coverage, but it’s still hard to read anything or buy anything without telling a chain of companies what you did, and getting advertised to about it the next day.

Even as actual harms from ad-supported, data-rich products are coming to light, we’re having trouble pointing to the harms as a reason to blanket-prevent the bad practices.

I don’t know if the end of privacy is a complex threat that will endanger our civilization, but the point is that we can’t really know until it’s too late. Maybe corona containment efforts and their aftermath will give us a framework to think more seriously about prevention.