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
If you haven’t already sign-up to receive this newsletter in your in-box!
These days everyone worries about the dangers of technology. With the Russian invasion of Ukraine these worries have become very focused on one specific technology: nuclear weapons. Despite this danger and the other dangers technology has introduced, there are still many people who expect the exact opposite, that technology will be our salvation. I brought this dichotomy up in my very first newsletter. Looking back I might have given the mistaken impression that whichever it ends up being, salvation or destruction, that it will be simple. We will either be permanently saved or permanently destroyed.
This is not just my mistake, most people make this mistake, particularly when it comes to our current worry, nuclear war. They take a horribly complicated event and simplify it down to a single phrase: “The end of the world.” And nuclear war is not the only technological danger where this simplification happens. People often use similar language when talking about climate change.
On the other side of things, the imagined salvation is perhaps not as dramatic or as sudden, but it is imagined as being just as straightforward. Last week I attended a lecture by Steven Pinker, who made the argument that progress is continuing and things will just keep getting better, a subject he has written several books about. In support of this argument he offered numerous graphs showing that trends in everything from violence to wealth have been steadily improving for decades if not centuries. From this he asserted that there is no need to worry, just as we solved all of our past problems we will solve all of our future problems as well.
The belief in humanity’s unstoppable progress and the fear that we will annihilate ourselves in a nuclear war represent the extremes of optimism and pessimism. On the one hand is the claim that science and progress have solved or will solve all of our problems, on the other hand is the claim that if the situation in Ukraine escalates 7.9 billion people will die. Neither of these claims are true, but we have a tendency to think in extremes because they’re easier to understand.
As it turns out, even a war involving all of the nukes will not kill everyone. Recently a Reddit user put together a simulation which predicted that around 550 million people would die from the war, and the ensuing fallout and nuclear winter. That’s about 7% of everyone. Obviously the simulation could be wildly inaccurate, though it does claim to be based on data from the International Atomic Energy Agency, the UN and CIA, but even if it was off by an order of magnitude that would still only be 70% or 5.5 billion people, leaving 2.4 billion people alive. An inconceivable tragedy, but not the end of the world. Also, these people might wish they were dead, because living after a nuclear war would be exceedingly difficult.
However, historically life has always been exceedingly difficult, not to mention messy. The Native Americans survived the loss of 90% of their total population. During the Black Death, Europeans survived death rates of up to 50%, with some people suggesting it was as high as 60%, very close to the extreme estimate of 70% above.
Despite this sort of messy middle being the historical default, we don’t like it. We want either the steady and implacable march of progress, or a quick end that absolves us of hard work. Even when we imagine surviving “the end”, we cut out most of the messy stuff, like raising crops, and making tools in favor of more simple apocalyptic stories, where there’s always plenty of canned food and lots of guns and ammo—even when we imagine a gigantic mess, we cut out all the truly difficult bits.
The modern world has made a lot of things easy that used to be incredibly complicated. It has made a lot of things possible that were previously impossible. In the process it has weakened our ability to deal with complicated and messy situations. We want the pandemic to go away if everyone just wears a mask, or if everyone gets vaccinated, or if we just ignore it. We want the invasion of Ukraine to stop if we implement the right level of sanctions, or institute a no fly zone, or, again, if we just ignore it. But the truth is that simplicity and ease are temporary aberrations, messiness has returned and we’d better get used to it.
You may not have realized that nuclear war would only kill 550 million people. If you feel any appreciation for this comforting fact, and would like more comforting facts in the future, consider donating.
550M makes sense. But the US is about 330M, Russia 100M and the EU 440M. So that’s enough to make te US and Russia extinct, or drop us to about 37% of the population if the EU joins us in one. I suspect the simulation isn’t capable of really capturing the long run consequences of removing the US from the world, I think it would add hundreds of millions in deaths from increased poverty, world choas and other problems. Also if all the nuclear powers take this as an opportunity to settle scores…China, North Korea, Iran, India and Pakistan… One might be able to get much closer to human extinction if they could personally target every single nuclear weapon in a way that optimized killing everyone. Even then I think once you got rid of 90% or so, it would be very hard to eliminate the rest. That would roll human population back to maybe the 1790 level or so.
Speaking of which, I’m wondering how well weapons would actually perform since Russia’s conventional performance seems pretty shoddy (but lethal enough). I’m wondering how many missiles really were checked and repaired all these years by diligent workers versus how many times did people just check a form that they did something. If the missiles launched, how many would actually work as intended?
An excellent book I read long ago on this was Command and Control. It was the story of how a dropped wrench in a missile silo in Arkansas in 1980 set off a set of consequences that resulted in the missile exploding killing 1 and leaving 21 others hurt…and a live nuclear warhead laying exposed on the ground for news helicopters to broadcast. Alternating chapters told the macro story of how the US military figured out how nuclear weapons should be controled. How do you make sure some dope doesn’t set one off by accident or on purpose while at the same time not putting so many codes and locks on them that they can’t be used quickly if war breaks out. Answer: Not as many locks as they probaby should.
Relevant to this, the US military developed total overkill on nuclear weapons for rather banal reasons. The army, navy and airforce all felt they should control nukes. Rather than giving it to one branch, all the branches got nukes. The airforce initially got bombers, but when missiles were developed they didn’t replace the bombers but added missiles.
At one point they finally did a simulation of what would happen if each branch carried out their nuclear strike plans. The result was absurd overkill…key targets would get hit dozens of times, and total non-coordination. Bombers approaching Moscow would be vaporized by incoming warheads.
If the Russian forces have a like measure of absurdity, the ‘good news’ is that a nuclear war may kill even less people but leave large portions of key centers (like where I am) not just devasted but with 100% extinction. Tomorrow is Monday.
I grabbed Command and Control, sounds pretty interesting. I too wonder about the reliability of Russian nukes. They claim they’ve been engaged in a decades long modernization project:
https://thebulletin.org/premium/2022-02/nuclear-notebook-how-many-nuclear-weapons-does-russia-have-in-2022/
But they also claimed they’d roll over the Ukrainians. I suspect that they pay slightly more attention to their nukes than to their trucks:
https://threadreaderapp.com/thread/1499164245250002944.html
But still any rational estimate would have to encompass a large amount of uncertainty on performance and reliability.
Something to keep in mind with those simulations: they assume every single missile will fire and that every single warhead will hit a target. In an actual nuclear war, you’d have missiles failing to go off; even the Trident II has had an 8% failure rate in tests. Doesn’t sound like much, but it adds up.
Currently, America and Russia have around 1,600 operational strategic weapons. Considering we had something like 12,000 targets during the Cold War, it’s a smaller number than it sounds. You need a couple warheads to destroy an airfield, than one ground burst for each runaway, for example. Many warheads would be destroyed and some would be intercepted. Perhaps not a lot, but our defenses are improving.
So you wouldn’t have many to spare for hitting cities. In the end, most of the deaths would likely be outside of the United States and Europe. Not because of fallout or nuclear winter, but because world trade would disappear and those dependent on food exports would starve. There’s already a serious risk of famine thanks to the current war. A nuclear one would make this vastly worse.
Excellent points, and I definitely think simulations (and definitely people’s imaginations) overestimate the immediate casualties and underestimate the long term casualties.
Could you clarify why multiple warheads are needed to destroy an airfield? Consider a single one megaton hit on an airport or runway. That airport isn’t coming back to use during the conflict. Perhaps planes could do emergency landings on the unhit runway a few days after the strike, although I’m not sure how helpful that is given you’ll probably have a ground radiation problem. I suppose after a year or so it would be possible to clear the wreckage and use the foundations and runways to rebuild the airport with less effort than doing it from scratch. Even if we are talking about an airforce base, how much of a typical base keeps the planes and men in hardened bunkers and hangers?
It makes sense to target a base with several warheads because if you just use 1, the base will be at 100% capacity if that 1 warhead is a dud. But if you pull off a single explosion even against a hardened base, I’m thinking you’re going to reduce its capacity down to something like 20% at best. Unless you’re thinking of missile silos and I see how that could cause thousands of warheads to be ‘wasted’ exploding over relatively less populated areas since a silo needs a direct hit to really take it out but at the same time the odds are by the time the warheads get to a silo its missile will have already been launched.
Quick googling says 83% of the US population lives in urban areas and there are 300 places with 100K+ people in the US. That tells me if you wanted to reduce the US population by at least 50%, you just need to ask your military to let you target 300 of their warheads. In fact I bet you could pull it off on a budget of just 100 warheads.