Book Review of Nuclear War: A Scenario by Annie Jacobsen

A review of Annie Jacobsen’s terrifying book about nuclear war

Hi there. This week I’m reviewing Nuclear War: A Scenario by Annie Jacobsen. Written in 2024, this title does exactly what it says on the tin. Short of an asteroid strike, says Jacobsen, this is the only scenario that could end the world in a matter of hours.

It seems fitting to be discussing this book on the anniversary of the September 11th attacks on the USA. I’m not saying that 9/11 was in any way comparable to a nuclear war, of course, but recalling that awful day might give us an inkling of the Earth-shattering upheaval people would feel.

Jacobsen does an excellent job of illustrating just how unspeakably awful a nuclear explosion would be for so many people. This vivid style of disaster writing is the book’s strongest point, but it’s overdone at times. I’ll focus the first part of my review on this. Then I’ll move on and discuss the narrative structure of the book, which is a minute-by-minute and second-by-second account of the unfolding disaster. After that I’ll get even more pedantic and question some of the assumptions and the ultimate conclusion of the book.

There are spoilers ahead, so be warned.

As always, before we start, let me encourage you to subscribe to the blog using the link below. As always, I’d love to see your comments below, or you can contact me via webform here or email here. Finally, if you’ve enjoyed this post, please share it with a friend. You might also be interested in my article all about nuclear weapons as part of the “CBRN” deadly quadrangle (and by far the scariest part).

If you enjoy this blog and want to support it, please consider a donation. Keeping this blog going doesn’t cost much, but it isn’t free either, so any help would be very much appreciated👍

Jacobsen describes a nuclear explosion in terrifying detail

The book opens with the biggest, strongest attention-grabber possible: the incineration of Washington D.C. by a 1-megaton¹ nuclear warhead. After describing the vaporisation of the Pentagon, the obliteration of Arlington Cemetery, and the instant disintegration of the Lincoln and Jefferson memorials, she turns her gaze further afield:

It has been three seconds since the initial blast. There is a baseball game going on two and a half miles due east at Nationals Park. The clothes on a majority of the 35,000 people watching the game catch on fire. Those who don’t quickly burn to death suffer intense third-degree burns. Their bodies get stripped of the outer layer of skin, exposing bloody dermis underneath.

…

Within seconds, thermal radiation from this 1-megaton nuclear bomb attack on the Pentagon has deeply burned the skin on roughly 1 million more people, 90 percent of whom will die… Most won’t make it more than a few steps where they happen to be when the bomb detonates. They become what civil defense experts referred to in the 1950s, when these gruesome calculations first came to be, as “Dead When Found.”

Jacobsen brings the horror of a nuclear explosion to light by giving the most narrative space to people who will experience the agony of a slow and painful death because everything which existed before to care for them is also burned beyond recognition.

In this, I presume that the author was inspired by eyewitness accounts of the WWII fire-bombings of cities like Hamburg, Dresden, and Tokyo, and the atomic bombings of Hiroshima and Nagasaki. Any of the accounts linked above make for harrowing reading, and these are only the top results for a handful of cities during this war. Jacobsen’s great talent in Nuclear War is to distil and then magnify that horror so we can easily imagine it happening in a city and time much nearer to us.

The author uses the same style throughout the book, however, and it quickly starts to grate. It’s like she feels the need to remind us on every page that the stakes are, indeed, very high. So we get things like:

Every fraction of a second matters. Every byte of information counts.

Or:

The outlook is catastrophic.

A worst-case scenario beyond comprehension.

Or:

Here, now, there is no room for error in judgement.

No room for error of any kind.

This isn’t a cardinal sin—the stand-alone paragraphs are easy to ignore and skim over, for one—but it does distract from the narration. We know that nuclear war is a serious business, we don’t need to be told at every turn.

Breathless though the writing is, however, it’s easy enough to forgive. It’s a mere distraction for the bulk of the book, but comes to life when the author is describing the cataclysmic set pieces which we’ve all come along to read. A nuclear explosion involves destruction on a scale we’re (thankfully) not used to comprehending, so Jacobsen is to be commended for bringing this to life in all its terrifying glory.

She follows the action second by second

The bulk of this story consists of a second-by-second and minute-by-minute description of the sequence of events from the first missile launch until the, well, end of the world as we know it. It’s a scarily short timeframe, in Jacobsen’s telling:

She gives this narration in a techno-thriller style. For example, we follow a satellite in geostationary orbit above North Korea as it picks out the infrared launch plume and sends an alert to NORAD in Colorado over the course of a fraction of a second. Then we check in with a bewildering array of command centres, radar tracking stations, and strategic military headquarters around the globe. Think Tom Clancy, with occasional quotes from experts or sidebars to describe a concept in more detail.

While we’re watching the unfolding panic and confusion on the ground, the author also makes sure to periodically check in on the progress of the missile, and later, the missiles. We see just how short the boost phase of an intercontinental ballistic missile is (3 minutes in the example below), and therefore how limited is the scope to take early action, when it’s most vulnerable:

In the diagram above (from original source):

1. The missile launches out of its silo by firing its 1st stage boost motor (A).
2. About 60 seconds after launch, the 1st stage drops off and the 2nd stage motor (B) ignites. The missile shroud (E) is ejected.
3. About 120 seconds after launch, the 3rd stage motor (C) ignites and separates from the 2nd stage.
4. About 180 seconds after launch, 3rd stage thrust terminates and the Post-Boost Vehicle (D) separates from the rocket.
5. The Post-Boost Vehicle manoeuvres itself and prepares for re-entry vehicle (RV) deployment.
6. The RVs, as well as decoys and chaff, are deployed during backaway.
7. The RVs and chaff re-enter the atmosphere at high speeds and are armed in flight.
8. The nuclear warheads detonate, either as air bursts or ground bursts

With this real-time narration, Jacobsen shows us just how little time decision makers on the ground have to make the most consequential choices of their lives—about half an hour. If anything, I think she is being too generous here. She stars off this book with a list of the dozens of notable military, political, and academic experts she interviewed for this project (normally you’d expect to see something like this in an appendix or end notes). It’s very clear that she’s received and is giving back to us as comprehensive a description of the US response to a nuclear attack as possible within the bounds of information security.

I think it’s over-generous because she makes no allowance for the kind of friction you would encounter along each step of the chain. She assumes the best possible timelines for information flow and decision making. Nobody is asleep, or on another call, or on the toilet when this sequence starts. To give a few examples:

• Six seconds after launch, she has three different generals already barking orders in a bunker underneath the Pentagon
• Fifteen seconds after launch, fighter pilots are running toward their waiting aircraft
• Two minutes after launch, the Secretary of Defense WAR and Chairman of the Joint Chiefs of Staff have scurried from their offices to the war room under the Pentagon

This is an observation, not a criticism, because she can only work with the information she’s given and can’t start making negative assumptions about how it would work in real life. If anything, it reinforces her point about the lack of time for decision making: the half an hour mentioned above assumes that every person and computer system is coiled like a spring, ready to go when the balloon goes up.

Contrast this with the early scene in Dr. Strangelove, when General “Buck” Turgidson’s deputy tries to reach him to let him know that a nuclear war is about to break out:

Yes, Dr. Strangelove is a comedy, and communications technologies have evolved somewhat from the 1960s. Nevertheless, I think the real response to a nuclear launch would have elements of this farce interrupting the slick machine-like operation Jacobsen describes.

Friction and farce aside, there are a few decisions in Jacobsen’s scenario which we need to examine in more detail.

Many people make questionable choices

This book paints an alarming picture of how a world-ending nuclear war might come about. There’s no question: it’s scary, and all too plausible. However, Annie Jacobsen wants us to think that it’s inevitable. It’s only inevitable if lots of people make lots of silly decisions. Let’s break down some of these decisions.

I’m not necessarily saying that the scenario isn’t still scary or believable. We have plenty of examples of world leaders making silly decisions in real life. But then again we have always had this problem, and nuclear war hasn’t broken out… yet.

Anyway, onto the silly choices which chip away at this scenario’s inevitability. Let’s start with:

• North Korea. This is the big one. Why, oh why, would the North Koreans launch a nuclear attack on the USA? Jacobsen hand-waves this a bit in the book by implying that they’re unpredictable and irrational, and miffed at being perceived as backward. Perhaps… but are they suicidal? Moreover, if they are intent on a first strike on the USA, then why with only three weapons? And why target the second missile at a nuclear power plant instead of a major city (I’ll deal with that one below)?
• Russia. Okay, Russia is paranoid and has a centralised command and control system which deters initiative and is liable to overreact. Also, its early warning radars are bad, so they mistake the American counterstrike against North Korea as an attack on them. Even though the Americans tell them that they are attacking North Korea, and they (the Russians) have seen a video of a nuclear explosion in California and know that they were not the ones responsible. It seems like a stretch that they would throw in the kitchen sink and doom themselves in the bargain just because they can’t get the President of the US on the phone.
• USA. There are a few ones here, from major strategic choices to tactical decisions which have major consequences:
  • Targeting North Korea with over-the-pole ICBMs. This seems dumb, when you know the Russians are paranoid and you know their tracking systems are sub-par and you haven’t spoken to their President on the phone yet. Use aircraft or sub-launched ICBMs or cruise missiles instead. The US has been hit with two nukes. If they had any more to throw at you, they would have done it already. You can take a small bit of time before delivering your nuclear retribution.
  • The President giving up his universal launch code so quickly. I struggle to imagine any leader giving their military commanders carte blanche to deal out as many nuke strikes as they want, even in a situation as dire as this.
  • The President’s Secret Service detail doing everything in their power to stop him from doing his job (making life-or-death decisions and speaking to the Russians in the few minutes he has left to live), instead physically carrying him on a madcap tour of various strong locations around the White House, before eventually jumping out of a helicopter with him. Net result, he still dies, the world still ends.

To be fair to Jacobsen: each of these are plausible mistakes for people/nations/systems to make (with the possible exception of the North Korea one). But it’s less plausible, while still possible, that all the mistakes line up in the way she describes here. It’s a classic Swiss cheese model of accident causation:

So the end result in this scenario (everyone firing nukes at everyone) is plausible, but not (in my opinion) the most likely outcome. Let’s say the holes in the Swiss cheese do line up, for the sake of argument. Then what? Does the world end? Let’s discuss some of the technical issues next.

There are some technical details to nitpick

Because I’m a pedant and you expect no less from me, I’m going to spend a few minutes quibbling about minor details. Only because, while reviewing this book, I came across a paragraph that was full of howlers and couldn’t let it go. I’m going to attack it as if it was one of my students’, back when I used to be a pain in the ass professionally:

She’s right, however, that it’s quite impossible for the human mind to comprehend the size of a nuclear blast. A useful comparison for me is Operation Sailor Hat, a series of tests which the US carried out in 1965 to simulate the effects of nuclear explosions using lots and lots (and lots) of conventional high explosives. I wrote about this previously: the amount of TNT you see in the picture below is 500 tonnes, or half a “kiloton.”

This is a lot of high explosives. The video shows just how much. Three hundred kilotons is the same as 600 of these giant stacks of TNT. The 1-megaton device which she describes incinerating Washington D.C. would be 2,000 stacks.

Another point I have to deal with is the North Koreans targeting Diablo Canyon nuclear power plant in California with a submarine-launched ballistic missile. This, according to Jacobsen, is the “Devil’s Scenario.” Is it really? Surely the Devil’s Scenario is a nuke hitting somewhere like downtown Los Angeles?

Let’s be clear (and I’m not saying Jacobsen makes this mistake): a nuclear explosion is a cataclysmic event with no man-made parallel. A nuclear power plant produces energy from nuclear fuel. But there’s no way that this nuclear fuel would contribute in any way to the blast of an existing nuclear explosion.

The problem it does give you is fallout, which is described in the book. But it’s fallout which is mixed with spend nuclear fuel, which is radioactive. The assumption is that all this nasty stuff gets swept up into the atmosphere and deposited over a huge area, rendering it uninhabitable. I’m not an expert in radionuclides, but I do know that materials either emit lots of intense radiation, but become stable quickly, or emit lower levels of radiation and stay radioactive for a long time. It seems unlikely, therefore, that the “radiation” effects of a nuclear explosion, even centred on nuclear power plant, would come close to the “fireball and blast” effects of that same explosion. That’s just my understanding, though, and I’d love to hear your thoughts in the comments.

Finally, the scenario ends with a nuclear winter. Again, I’m no expert, but I do believe that many of the direst warnings relating to the climatic effects of nuclear weapons are at least contested. That’s not to say that the endpoint of Jacobsen’s scenario is a rosy one, but that the destruction of hundreds of cities and the deaths of hundreds of millions is the really scary part, not the precursor to global extinction.

Conclusion: Nuclear gigadeath is plausible, and that’s scary enough

Following on from the nuclear winter point above, I think the true horror of a nuclear war is not that it marks the end of the world, because it probably doesn’t. The planet will certainly survive whatever we throw at it, and enough of humanity will survive as well. And not in a “survivors will envy the dead” kind of way that we see in lots of post-apocalyptic fiction, but in identifiable societies and communities, albeit ones which have been kicked back a few decades on the developmental ladder.

Twenty four years ago, two airliners tore the heart out of the centre of the world’s economy, killing thousands and scarring millions psychologically. A nuclear war would kill hundreds of millions and scar billions. But life will go on, and the survivors will build on the ashes and eventually, many generations later, repopulate the empty cities. There’s hope there, but also a grim reality that, once you’re gone, you’re replaced and (largely) forgotten.

In some ways, the idea that nuclear war = total annihilation is a sort of comfort blanket. “If it happens,” we say, “at least we’ll all go out together.” No, we won’t, and that’s what makes it scary. It could happen, and you might still be here afterward and have to pick up the pieces.

Regardless of how it ends, Annie Jacobsen’s scenario is a riveting account of how we might get to that point. If there’s a lesson here for strategists and policymakers, it’s that “launch on warning,” the policy which is at the heart of this book, is a recipe for runaway disaster. Most US Presidents in recent years, according to the author, have pledged to get rid of the policy when campaigning. Once elected, however, they quietly keep it.

If you’re interested in geopolitics, strategy, deterrence, or war in general, you’ll enjoy this book. The input from world-leading experts shines through on every page, even if the prose suffers at times. I’ve yet to see a better fictional description of a nuclear blast: it’s terrifying.

The technical issues I have are minor enough, and the only major criticism I’d level is that a lot needs to go wrong for this particular scenario to play out. It only needs to go wrong once, however.

That’s all for this week, folks. Thanks for reading and please remember, if you haven’t already, to subscribe using the link below. Please also share this article with a friend. See you next week.

Featured image: Castle Bravo: The Largest U.S. Nuclear Explosion, Ariana Rowberry, via Brookings Institute (February 27, 2014)

1. I explained the “megaton” and “kiloton” measurement system in my post about nuclear weapons. ↩︎
2. I should note that Wellerstein was one of the experts consulted by Jacobsen for this project. ↩︎