The week before last, at the International Journalism Festival, I talked briefly with a journalist who left a major network/publication to become an independent journalist with an apparently huge Substack and YouTube channel. I asked them how that worked out the way it did? They gave a few answers (of varying levels of persuasion, for me — “luck” was not among them), that mostly came down to “hard work” (i.e., they do updates of everything all the time, daily, etc.). It sounded fairly hellish, I will admit; it is hard-enough for me to get done the things I want to get done and have a weekly blog entry, I cannot imagine trying to do it on a daily basis.

One of their other things they said was that they ran all of their writing through ChatGPT or some other LLM/AI service to make it punchier, identify typos, and otherwise improve the chances it would do something in the world. I am pretty opposed to using LLMs for written work for a lot of reasons, pride in the craft being one of them, but I admit the thought stuck in my craw. Because I do hate the typos that pop up in my writing online, a function of a fast schedule (weekly content that is mixed into a lot of other work) and the individual nature of it (there’s no staff over here!). And I do love when I can write for a venue that has a good copyeditor, because they always have a better sense than I do of how to cut and condense this stuff (I just find it all too interesting).

What if, I wondered, I experimentally uploaded my blog post for this week to ChatGPT and asked it to work its wonders? Would it be impressive? Would it make something that would give me that feeling of a good editor? Would it at least find my typos? Am I being too much of a curmudgeon about this stuff?

So I took this week’s blog post (linked below), put it into a MS Word file, and uploaded it to free version of ChatGPT. I told it to tighten up the writing and identify typos, but to not change its tone of voice to the insipid one that ChatGPT usually uses. “OK!” it said, insipidly, and then after a second of thinking gave me a new MS Word file with its edits.

I told MS Word to create a composite version of my original file and the new file with the changes highlighted, so I could see its magic. It asked me who the author the changes should be listed as, and I suggested “Clanker.” It did identify a couple clear typos (I somehow misspelled “annihilation” once and ignored the red underline, I once I mis-typed Ellsberg’s name), but it also let a few through that I noticed on a re-read (e.g. “would bu” was left in, where it should have been “would be”).

But otherwise it almost changed nothing. A few words snipped here and there, but no actual editing of importance was done. Except for one, persistent change, that is so strange that I feel the need to provide a screenshot of it, because it seems so absurd:

That’s right, it changed the word “discovery” to “disco” multiple times in the document. “The disco of nuclear reactions.” “The disco of induced radioactivity.” “The disco of nuclear fission.”

I admit that I did laugh at loud, and then laughed at myself for ever thinking this was a good idea. Now, I am sure that it doesn’t always give results like this. And perhaps if I were paying for the more expensive version of ChatGPT it would do better. Or another service. But the whole thing reminded me why I am so loathe to rely on this kind of thing in the first place. Aside from the absurdity of wasting gobs of energy to query a statistical database full of copyright-infringed texts (including, I am positive, my own!) in a business model that seems guaranteed to destroy the economy no matter whether the predictions of its advocates come true or not, for an industry run by anti-human tech-bros who would set all of human culture on fire for a dollar and their science-fiction dreams… aside from all that, it also is just so often not that good.

Imagine giving a human being the task of editing this and aside from the lightness of it, they changed all instances of “discovery” to “disco.” You’d wonder if it was a joke. You certainly wouldn’t be encouraged to pay them anything. Or have them teach your children. Anyway, I am not going to be using LLMs in this capacity anytime soon, I don’t think — so I just hope that you will see my typos for what they are, a consequence of an all too human workflow!

Well, I am an AI grump, I know this. And of course I am participating in a selection bias, here. I have found a few uses of LLMs in my own life that, all of the ethical stuff aside, I find good-enough. They all involve uses that are very easy for me to check the veracity of the output of, like translating a statistical function to Javascript for me, which I can then look over the logic of, and test that it gives the correct outputs — even this often requires some back-and-forth as the LLMs make assumptions about what I want and diverge from instructions. Or translating English into French for the purposes of a formal e-mail (even then, I look it over closely).

None of these uses involve anything I enjoy doing, like writing or researching or even the more enjoyable parts of programming. And I would be happy to give them up in a heartbeat, really. But I simply do not understand why people would want to use them for things like that, but I suppose people come in all sorts of different packages…

The post for Doomsday Machines that I subjected to that experiment is the one I posted on Friday, about the topic of “atmospheric ignition,” that is, setting the Earth’s atmosphere (or oceans, or crust) on fire through an endless, out-of-control nuclear reaction:

You might think, if you are of a certain frame of mind, that you have read everything there is to read about that topic — e.g., that during World War II, the Los Alamos scientists contemplated whether they could set the atmosphere on fire, and did the math and found out they could not.

This is only part of the story, however. The trope goes back to at least 1903, as far as I can tell, and the actual “math” behind the ignition problem is more interesting than I think has been appreciated anyway. More importantly, as Dan Ellsberg emphasized in his 2017 book The Doomsday Machine, there is at the core a question that doesn’t go away once you conclude that you can’t burn up the world that way: a question of how much risk is acceptable for scientists to take.

I am not sure exactly what I will write about next week, but I stumbled across a very revealing document from the mid-1950s that shows how the generals thought about nuclear war at that time, so I may write something that centers around it… we shall see…!

Read more

“Only a nightmare dream”

The earliest case I am aware of this idea being put into circulation dates to 1903, where the atomic pioneer Ernest Rutherford was reported by a colleague to have suggested that “it was just conceivable that a wave of atomic disintegration might be started through matter, which would indeed make this old world vanish in smoke.” This supposedly referred to a “joke” of Rutherford’s that “some fool in a laboratory might blow up the universe unawares.” It was, the colleague reported, merely a “nightmare dream.”1

Rutherford’s collaborator, the chemist Frederick Soddy, also invoked the "nightmare” of death-by-transmutation in his public lectures. In a 1903 lecture on “Some Recent Advances in Radioactivity,” Soddy considered that:2

The knowledge [of radioactivity must] make us regard the planet on which we live as a storehouse stuffed with explosives, inconceivably more powerful than any we know of, and possibly only awaiting a suitable detonator to cause the earth to revert to chaos.

The next year, Soddy gave on radium the next year at the Corps of Royal Engineers, in which he suggested that:3

It is probable that all heavy matter possesses—latent and bound up with the structure of the atom—a similar quantity of energy to that possessed by radium. If it could be tapped and controlled what an agent it would be in shaping the world’s destiny! The man who put his hand on the lever by which a parsimonious nature regulates so jealously the output of this store of energy would possess a weapon by which he could destroy the earth if he chose. […] The fact that we exist is a proof that [massive energetic release] did not occur; that it has not occurred is the best possible assurance that it never will. We may trust Nature to guard her secret.

Soddy did not mention the possibility of either a weapon or global destruction in his 1909 book, The Interpretation of Radium, which is an interesting omission. The idea was resurrected and further popularized some years later, in May 1922, by the British chemist Francis William Aston, who was awarded the Nobel Prize in Chemistry that year for his work on isotopes. Aston gave a lecture at Philadelphia’s Franklin Institute in which he warned that:

Should the research worker of the future discover some means of releasing this [atomic] energy in a form which could be employed, the human race will have at its command powers beyond the dreams of scientific fiction; but the remote possibility must always be considered that the energy once liberated will be completely uncontrollable and by its intense violence detonate all neighbouring substances. In this event the whole of the hydrogen on the earth might be transformed at once and the success of the experiment published at large to the universe as a new star.

And Aston repeated this paragraph in his Nobel Prize acceptance lecture later that year. Aston’s lectures led to a spate of newspaper articles, not unlike an anonymous “Oxford scientist” of the same year who warned both of atomic bombs and possible global annihilation: “it is conceivable, too, that such terrific force might eventually be liberated as to blow up the world.”

Newspaper headlines globally ran headlines about these warnings: “Tinkering With Angry Atoms May Blow Up the Earth,” “Earth as a Bomb,” “Will the Atom Blow Up the World?” But dismissing this idea, in the 1900s-1920s, was, as Soddy illustrated, simple enough: if it was that easy to turn the world into a run-away nuclear reaction, then surely it would have happened at some point over the course of the Earth’s long history. What are the odds that humanity, that johnny-come-lately species, is going to be the one who set it all aflame?

And it was all still quite hypothetical, as humanity had no sign of this level of control over transmutation, no “detonator.”

“Atmospheric ignition”

This would change in the 1930s. The discovery of induced radioactivity in 1934 meant that the possibility of chain reactions was possible: that one nuclear reaction could lead to many more nuclear reactions. The discovery of nuclear fission in uranium in late 1938, and the fact that neutron-induced fission reactions produced additional neutrons, gave a practical way to accomplish nuclear transmutation at a whim. That the world had not already in the long past exploded from nuclear fission reactions suggests these reactions are not easy to start or propagate. And, indeed, the issue was quickly discovered: only some isotopes of uranium are easily fissioned, and most are not, and so it requires quite a lot of artifice to create the conditions for such chain reactions to occur.4

But the possibility of explosive nuclear reactions was, eventually, pursued for military purposes. And so we jump ahead to the famous meeting at the University of California, Berkeley, in July 1942, where J. Robert Oppenheimer assembled a coterie of nuclear physicists to discuss the theoretical basis for designing an atomic bomb, prior to the formation of the Los Alamos laboratory. And into this mix, the Hungarian physicist Edward Teller suggested a new nightmare: “atmospheric ignition.”

Accounts of the discussion vary. Daniel Ellsberg’s The Doomsday Machine (2017) contains an excellent summary of many of them.5 The gist is that Teller laid out the possibility that in the event that they created and detonated a bomb powered by an explosive uranium-235 fission chain reaction, it would produce, among other things, an intense and concentrated source of heat. This heat would be enough, he argued, to cause fusion reactions in light elements, like hydrogen. Teller’s primary interest, here, was the idea of the Super, or hydrogen bomb.

But this was linked with another idea, that these fusion reactions would also take place in the Earth’s atmosphere, notably hydrogen (in its water content) and nitrogen (which makes up 78% of our air). If the atomic bomb started such a fusion reaction in the atmosphere, and the fusion reaction was hot enough to cause additional fusion reactions in that same atmosphere, then a runaway fusion reaction would, within seconds, traverse the globe, burning everything, and leaving dead rock in its wake.

This kind of problem, to a nuclear physicist, is equal parts interesting and distressing. The assembled group, which aside from Oppenheimer also included the meticulous German physicist Hans Bethe — who would later win the Nobel Prize in Physics for his work in the 1930s in uncovering the fusion reactions that power stars — picked over Teller’s equations and reasoning. A fatal flaw was soon found: the heat of any fusion reactions created would dissipate quickly in the atmosphere. This would likely make such a run-away fusion reaction impossible, as the cooling would prevent it from spreading. Thus Bethe, nearly instinctively, judged the scheme as “impossible.”

“The ultimate catastrophe”

However, that this was considered far from proven. Oppenheimer himself subsequently traveled to Chicago talk with Arthur Compton, the Nobelist physicist running the Metallurgical Laboratory to create the world’s first nuclear reactors there, to share news of the possibility. Compton wrote of this some 14 years later in his memoir:6

I’ll never forget that morning. I drove Oppenheimer from the railroad station down to the beach looking out over the peaceful lake. There I listened to his story. What his team had found was the possibility of nuclear fusion—the principle of the hydrogen bomb. This held what was at the time a tremendous unknown danger. Hydrogen nuclei, protons, are unstable, for they could combine into helium nuclei with a very high temperature. But might not the enormously high temperature of an atomic bomb be just what was needed to explode hydrogen? And if hydrogen, what about the hydrogen of sea water? Might the explosion of an atomic bomb set off an explosion of the ocean itself?

Nor was this all. The nitrogen in the air is also unstable, though in less degree. Might it not be set off by an atomic explosion in the atmosphere?

These questions could not be passed over lightly. Was there really any chance that an atomic bomb would trigger the explosion of the nitrogen in the atmosphere or of the hydrogen in the ocean? This would be the ultimate catastrophe. Better to accept the slavery of the Nazis than to run a chance of drawing the final curtain on mankind!

This is, as Ellsberg later noted, quite a proposition: if the atomic bomb had any chance of killing the entire planet, then it was better to accept the possibility of a Nazi victory than total suicide. We will come back to this proposition, later.

In many later recollections, particularly in the 1970s when this issue became raised again, many Manhattan Project participants, like Bethe, outright dismissed the idea that they ever considered atmospheric ignition even a remote possibility. One finds this sentiment in many modern-day discussions of it. Ellsberg argued, and I think he is correct (for both his reasons, and others that I will come back to later), that this latter-day denial is bad history.

It is perhaps quite true that Bethe himself never took it seriously, for reasons that we’ll get to below. But it is also clear that there were some who did take it more seriously, and that even up to the moment of the Trinity test there were those who feared, in some small way, that the “ultimate catastrophe” might still be in the cards. And the insistence that this was not a legitimate concern was in part an effort to retrospectively paper over that any kind of uncertainty about a risk that might have been taken.

Because to acknowledge any sense of uncertainty on the issue is to suggest that these scientists were willing to take a secret gamble on humanity even more so than they did by merely ushering these weapons into the world. And to acknowledge that would be to undermine the story that the atomic scientists were telling after World War II that they were the responsible experts whose views should be taken seriously when shaping the future of the world.

In 1959, Compton told the writer Pearl S. Buck that:

scientists discussed the dangers of fusion but without agreement. Again Compton took the lead in the final decision. If, after calculation, he said, it were proved that the chances were more than approximately three in a million that the earth would be vaporized by the atomic explosion, he would not proceed with the project. Calculation proved the figures slightly less—and the project continued.

This suggests the issue was a more drawn-out discussion than the “never took it seriously” narrative would imply, and also the firm-but-arbitrary limit of three-in-a-million as the “acceptable odds” for this catastrophe. Those are indeed long odds, but not as long as winning lottery jackpots.

What was the source of these estimates? Teller did write up a paper in 1943 on the subject (cited in some literature as LA-001), but it has never been made available that I have seen. The internal, top secret Manhattan Project History, commissioned by General Groves, with its Los Alamos section written by the physicist David Hawkins and finished in 1947, describes the work thusly:

It should be mentioned at this point that in the early period of the project the most careful attention was given to the possibility that a thermonuclear reaction might be initiated in light elements of the Earth’s atmosphere or crust. The easiest to initiate, if any, was found to be the reaction between nitrogen nuclei in the atmosphere. It was assumed that only the most energetic of several possible reactions would occur, and that the reaction cross-sections [probabilities] were at the maximum values theoretically possible. Calculation led to the result that no matter how high the temperature, energy loss would exceed energy production by a reasonable factor. At the assumed temperature of three million electron volts the reaction failed to be self-propagating by a factor of sixty. This temperature exceeded the calculated initial temperature of the deuterium reaction by a factor of one hundred, and that of the fission bomb by a larger factor.

The impossibility of igniting the atmosphere was thus assured by science and common sense. The essential factors in these calculations, the Coulomb forces of the nucleus, are among the best understood phenomena of modern physics. The philosophic possibility of destroying the earth, associated with the theoretical convertibility of mass into energy, remains. The thermonuclear reaction, which is the only method now known by which such a catastrophe could occur, is evidently ruled out. The general stability of matter in the observable universe argues against it. Further knowledge of the nature of the great stellar explosions, novae and supernovae, will throw light on these questions. In the almost complete absence of real knowledge, it is generally believed that the tremendous energy of these explosions is of gravitational rather than nuclear origin.

It is not clear whether this perfectly reflects the reasoning of 1943-1945. Ellsberg interviewed Hawkins about the question in 1982, and says that Hawkins told him that he had “done more interviews with the participants on this particular subject, both before and after the Trinity test, than on any other subject,” and that the atmospheric ignition question was continually “rediscovered” by other, younger researchers, and had to be dismissed anew again and again. This suggests another origin of the early defensiveness about this topic — and part of the its seemingly conclusive inclusion in the Manhattan District History, which was explicitly meant to serve as a classified record of activities to be consulted in the event that there was a need for justifications of actions taken during the war. Hawkins would tell Ellsberg that by “impossibility,” he had really meant “for practical purposes,” as opposed to literal “impossibility.”

“A colossal catastrophe”

In March 1946, the question was raised from another source. In the months prior to Operation Crossroads, there was considerable public controversy regarding the safety of two underwater tests that were planned. One was a shallow underwater test (Baker), and the other was a deep underwater test (Charlie, which was ultimately cancelled for other reasons). These underwater shots appear to have generated similar fears about fusion ignition, this time of the oceans.

The Harvard physicist Percy Bridgman, whose high-pressure lab had generated data on plutonium’s equation of state during in the Manhattan Project, and who had received the Nobel Prize in Physics for 1946, was so alarmed by the idea that he wrote to Bethe on the subject, and his fears were forward to General Leslie Groves:

What is worrying to me is the possibility that if the bomb is exploded in the ocean the hydrogen may be converted to helium with an astronomical release of energy. […] If the history of physics teaches any one thing, it is that long-range extrapolations are hazardous. Even the best human intellect has not imagination enough to envisage what might happen when we push far into new territory. […] To an outsider the tactics of the argument which would justify running even the slightest risk of such a colossal catastrophe appears exceedingly weak. […]

If I am right in thinking that the tactics of the argument are weak, then it would be wrong to drop the bomb whether or not the ocean explodes. Suppose the bomb is dropped as [is] at present planned, the ocean does not explode, and later it should become known to the general public that the argument had been weak and that the scientists had permitted the taking of a stupendous chance without doing everything in their power to safeguard all possibilities. There might be a reaction against science in general which would result in suppression of all scientific freedom and the destruction of science itself.

This appears to me as cause for greater concern than the blowing up of the ocean, which after all would not very much affect a world of dead men.

That it was forwarded to Groves points to some concern about it. It is possible that someone like Groves might have been worried less about the reality of Bridgman’s fears, but about the way that having such a thing in writing could, inadvertently, bring Bridgman’s fears about later accountability into reality. How “weak” was the argument, and how would they be later judged if someone — including Bridgman himself — made this point publicly?

The immediate outcome from Bridgman’s letter appears to have been a “debunking” article by Hans Bethe, which appeared in the March 15, 1946 issue of The Bulletin of the Atomic Scientists. By this point, Bethe notes, there had been “much public discussion” about the possibility of atmospheric or oceanic ignition. Bethe says that the concern is “very well taken,” but is thoroughly discouraging of the idea. His basic argument is that because an atomic bomb contains a lot of non-reacting material — everything beyond the part that is fissioning — that will necessarily lower the temperatures that are transmitted to the atmosphere, and that the temperatures needed for fusion are very high indeed. He concludes that while there were unknowns, safety seems guaranteed by wide margins.7

It is an interesting and different argument than any of the other ones given here. He also, per his expertise, discourses at length on the conditions inside of stars, and how different they are from those on Earth, even those places that are near exploding atomic bombs. It is perhaps the best evidence we have for how Bethe himself would have thought about it, which is interesting because it is not how, say, Teller seems to have considered the question. Bethe’s answer from 1946 is far more categorical than any of the other ones offered up so far.

“The dangerous temperature”

The concerns about Crossroads might provide some context for the formal writing up of a paper tackling the atmospheric ignition problem by Teller, Emil Konopinski, and Cloyd Marvin, Jr., titled “Ignition of the Atmosphere with Nuclear Bombs” (LA-602), dated to August 1946.8 It might be, for example, the “due diligence” desired to show that the “argument” about ignition was, in fact, not weak. But it only concerns itself with atmospheric ignition (not oceanic ignition), so perhaps not. But it is otherwise not entirely clear why this was written up nearly a year after the Trinity test, particularly if there is an earlier paper from 1943 on the same subject.

The Konopinski–Marvin–Teller paper is extremely technical but not too difficult to follow if one is familiar with the basic terms. The question they try to tackle is, what would the easiest possible atmospheric ignition reaction be, and are there conditions under which it could ever occur, based on what was known at the time?

They assume that a nitrogen–nitrogen fusion reaction would be “easiest” fusion reaction to induce in the Earth’s atmosphere. They did not know how easy it would be to induce, because they did not know the probability of that fusion reaction under different conditions of temperature or pressure (they did not know its cross-section, in technical terms). Given these uncertainties, they advocated a risk-adverse philosophy: “because of the uncertainties in the knowledge of these processes, the policy should be adopted of exaggerating the dangers at any point which appears at all questionable.”

So they made some conservative guesses as to what seemed reasonable based on what they did know, and then then reasoned about how this reaction might propagate in the atmosphere if it started, balancing the processes that would produce heat, and those that would dissipate it.9

All of their results were that no reaction would propagate. A major question of interest for the paper, though, was how confident they were in that answer. This came in the form of a “safety factor,” the ratio of energy losses to gains. If that ratio of losses to gains is >1, then the reaction is a net loss — no propagation, no global thermonuclear apocalypse. If it ever gets close to 1, however, that is a problem. They performed their equations with different assumptions about how much energy was being generated by the bomb, and found that for all reasonable numbers, the “safety factor” was very safe indeed.

But they found that if the initial energy was high-enough, it would get to a “safety factor” that was only around 1.6. That was still “safe,” but was close-enough to 1 to indicate that any major errors in their understanding could be quite dangerous indeed. This corresponded to an initial temperature of 10 MeV, which was very hot indeed. They explained that this would be the total energy of a weapon on the order of 27,000 megatons was entirely pumped into the air as heat. In reality, they explained, most of that energy would not be converted into heat, so even this was unrealistic situation — a weapon over a million times more powerful than the weapons of World War II — was not very plausible.10

They continued, though, and suggested that if somehow the energy of 3 tons of fissioning material was dumped into a very small volume of air — a sphere with a 7 meter radius — that might get closer to the “dangerous” temperature of 10 MeV. This is still not that plausible: the weapons of World War II each fissioned only about 1 kilogram of material, so we are still talking about something thousand times more explosive, on the other of 50 megatons.11

But this much lower number, they noted, does start to get one into the mindset of Teller’s contemplate Super bomb, which was assumed to be possible of being thousands of times more powerful than the weapons used against Japan, and would also release more of its energy in a “useful” form for heating the air than the fission bombs. Thus a fusion reaction in “a sphere of liquid deuterium only 1 to 1.5 meter in radius may suffice,” which in the range of the future multi-megaton Super bombs being contemplated during the war were imagined to be.12

Still, even in this situation, the authors have faith that the cooling effects would win out. The abstract of the paper is categorical: “The energy losses to radiation always overcompensate the gains due to the reactions.” But the actual conclusion of the paper couches its work in more uncertainty than this:

There remains the distant possibility that some other less simple mode of burning may maintain itself in the atmosphere.

Even if the reaction is stopped within a sphere of a few hundred meters radius, the resultant earth-shock and the radioactive contamination of the atmosphere might become catastrophic on a world-wide scale.

One may conclude that the arguments of this paper make it unreasonable to expect that the N+N reaction could propagate. An unlimited propagation is even less likely. However, the complexity of the argument and the absence of satisfactory experimental foundations makes further work on the subject highly desirable.

Now, again, this particular paper was finished after the Trinity test, so its conclusion that the atmosphere would not be set aflame by Trinity was presumably considered pretty valid. In his memoirs, Teller says that he, Konopinski, and Marvin, “demonstrated” with their equations that “such a cataclysm could not occur,” but that the famed physicist Enrico Fermi was not satisfied:13

But when we had completed our proof, Fermi insisted that we go one step further. All we had proved, he pointed out, was that such an explosion could not occur according to known laws of physics. But what presently undiscovered phenomena might exist that, under the novel conditions of extreme heat, might magnify the consequences and lead to an explosion? Considering that question occupied my time immediately before the test. I explained the possibilities I had considered and asked for suggestions from everyone who would listen to me.

Fermi’s question is a bold one, and perhaps is the reason for the expressions of doubt, and desire for future research, at the end of the 1946 paper. After all, if all of this is new physics, what is to say that they’ve actually got it all figured out? They knew they didn’t even know the exact cross-section of the reaction they are talking about — what about the things they didn’t know they didn’t know?

In October 1945, Teller wrote up a short paper for Fermi about the possibility of the Super, as part of the postwar planning purposes. In it, he mentioned atmospheric ignition in the context of an even bigger weapon:14

Careful considerations and calculations have shown that there is not the remotest possibility of such an event [ignition of the atmosphere]. The concentration of energy encountered in the super bomb is not greater than that of the atomic bomb. In my opinion the risks were greater when the first atomic bomb was tested, because our conclusions were based at that time on longer extrapolations from known facts. The danger of the super bomb does not lie in physical nature but in human behavior.

Which is an interesting data point in Teller’s own thinking on the matter, and the question of the unknown unknowns.

“Disastrous consequences”

The matter was not, for Teller, apparently resolved completely. In 1951, as work on the hydrogen bomb was moving forward to the point of a planned test in late 1952, he asked the physicist Gregory Breit, then at the Yale University, to independently re-consider all of the calculations with fresh eyes. Teller apparently told Breit that “the considerations of the wartime period might have been too hurried,” and that the plans to test a Super “have increased the importance of the question.”15

By this point, much more was known about fusion reactions, and the fact that they were in fact much harder to start and propagate than Teller had imagined in the 1940s. Breit and his “group” of postdocs and graduate students tackled the problem and had answers by the fall of 1952, prior to the first megaton-range thermonuclear test, Ivy Mike.

The Breit group’s recalculation was able to substitute much more definite data than had been available in 1943-1946. They concluded that, as before, there was a “dangerous temperature” in which the possibility of energy gains might overcome energy losses, and even revised it downward from the original — they agreed with 10 MeV as a dangerous temperature, but thought it could possibly go as low as 5 MeV. But they noted that the energy of the bomb being contemplated (on the order of 10 megatons) were 500 to 1,000 times too low (only 10-15 keV). “In the fall tests,” they concluded, “there is no danger that one can see.” They considered as well that “although one does not know all the factors with certainty it is quite unreasonable to expect them to be off to a degree which would matter.”

Even here, though, they felt the need to add a bit of critique. The original paper, and perhaps their own, suffered from the possibility of unknown unknowns:

The theory of nuclear forces is still in a rudimentary stage and while it is generally agreed that very large effects of this kind are improbable it would be unfortunate if one were mistaken and if one were the cause of disastrous consequences through overconfidence in present day nuclear theory. Much of the latter [the report] is based on incomplete evidence. The attitude regarding what constitutes certainty in question under discussion Is necessarily different from that taken in deciding what constitutes a good Ph.D. dissertation problem or even what constitutes a suitable subject for inclusion in an ordinary bomb test.

We are back, again, at Ellsberg’s question: how much risk is allowable if you are talking about the possibility of global annihilation? Should “normal” scientific attitudes towards risk and uncertainty prevail, or are they inadequate when the risk of being wrong is so large? Can one know, in advance, how wrong the unknown unknowns could be? It is interesting to consider that in only a few years, the second American thermonuclear test, Castle Bravo, would become a major radiological accident in part because of the inability of scientists to predict the physics involved in the test. Even the Trinity test itself was several times more explosive than had been anticipated ahead of time.

“The necessary conditions”

In the 1979, two Livermore scientists, Thomas A. Weaver and Lowell Wood, published a paper that re-studied the atmospheric and oceanic ignition problems with the benefit of many decades’ of data and advanced computing capabilities. Instead of asking whether kiloton- or megaton-range nuclear detonations could set the atmosphere of Earth on fire — they already knew it could not, plainly — they instead asked what the “necessary conditions” would be if you were to set off a runaway nuclear chain reaction in an atmosphere.16

They concluded, as I have written about in the past, that the “easiest” way for such a thing to happen would involve increasing the deuterium content of the ocean by twenty times, at which point it could be ignited by a 200 teraton (20 million megaton) explosion. Which is a very precise way to say “impossible.”

By this point, in 1979, I think it is fair to say that no scientists took seriously the possibility of atmospheric ignition. If anything, they were perhaps all too quick to dismiss it even retrospectively. By the 1970s, the idea of the paternal scientific expert giving wise counsel was coming under widespread attack as a result of the deep linkages that had formed between the scientific community, the weapons industry, and the perceived role that they played in enabling and perpetuating the Vietnam War.17

But it is worth returning to Ellsberg’s question, which, as we have seen, is echoed in many of the earlier discussions by the historical actors themselves: What level of risk is tolerable in this situation, particularly when these risks are being taken by agents who are not being subject to even the oversight of scientific peer review, much less democratic oversight?

Let’s say that we decide that Compton’s criteria of anything less than a 3 in a million chance is too much, when we are talking about killing everyone on Earth. One might still find that to be a dangerously risky proposition given the unknowns and the potential consequences — again, one-in-a-million odds events do occur, sometimes.

But let us imagine that we are at least somewhat comfortable to imagine that, under the circumstances of World War II or the early Cold War — where these bombs are hardly being developed and tested for mere interest but as part of a major global confrontation — that we might imagine that we could agree that these are somewhere on the order of magnitude of odds we might accept. One in a million. If the odds of atmospheric ignition were imagined by experts to be, say, one in a hundred, would the Trinity test have been vastly irresponsible? I think undoubtedly.

But here’s where someone like Ellsberg wants us to flip things around. It’s not just about atmospheric ignition. Ask yourself: what do you think the odds are for a nuclear war occurring in the next century? My sense is that most people who study this question seriously would consider them to be much more likely than one-in-a-million, and that the historical data we have — with a few very “close calls” within the first eight decades of the nuclear age — suggest that it might be very high indeed.

The fears of atmospheric ignition, in retrospect, seem perhaps quaint, and attackable by both mathematics and experience. But the underlying problem, the underlying question of how much risk is acceptable, seems to be as potent as ever.

His talk was on a very important question, one made all the more important by the rhetoric of two weeks ago: how do we evaluate various kinds of “nuclear threats” that are made or alleged to have been made? Perkovich prefers to call these kinds of things “nuclear manipulations,” because not all of them are genuine threats (e.g., if one is just “reminding” the world that you have nuclear weapons, that is different from, say, drawing a red line and saying, “if this event does or doesn’t happen, I will use nuclear weapons,” even if the former still can be rather “threatening” if made during a crisis). Perkovich’s paper on this, which you can read online here, is a careful attempt to think about the various factors that would go into evaluating such “manipulations” in real-time, with a very nice “decision tree” for weighing the relative urgency and seriousness of a given “threat.”

I find this kind of thing quite interesting and stimulating. I have an unabashed passion for taxonomic approaches to the world (make it orderly!) and of course my “historical brain” loves to think about examples from the past that could be fodder for complicating such a system. (And I recognize that complicating the system is often deliberately not the intent, nor necessarily helpful, but still, it is my impulse.) So, for example, where most of Perkovich’s criteria are simple yes–no binaries, I am always tempted to create little spectrums and then think about how they would be weighed. For example, “Did top leader make or affirm the statement?” for me instantly splits into a little hierarchy of possible “voices” for a threat, e.g., the “top leader,” a top military leader, someone in the cabinet, someone who is assumed to be a mouthpiece for the leadership even if they aren’t technically in it, and so on down the line. And instead of yes–no–unknown for “did it involve nuclear warheads?,” I want to add something like, “it was kept deliberately ambiguous,” and think about how I would want to chart that between “yes” and “no.”

All of this is a heuristic, of course, a way to practically and roughly think about how to distinguish between different kinds of situations and scenarios. It’s a little model, one aimed at utility rather than representation of the world as it is. I love little models, particularly because I am not that invested in whether they represent “reality” in a deep sense (plenty of models don’t, but are still useful).

Perkovich also mentioned that he has a book coming out from Adelphi Books later this year on Nuclear Threats in the Ukraine War, which I am very keen to see. There are bits and pieces of that story that have been reported on (some of which is in the paper linked above), but there’s quite a bit that is not generally known that I think is rather important to get out there, so I am hoping it will lead to a broader discussion of this issue.

After teaching on Wednesday I got onto a plane and flew to Rome, Italy, and from there took a shuttle bus to Perugia, for the International Journalism Festival. I had been asked to be on a panel about the nuclear imagination (and its limits) with Beatrice Fihn (formerly of ICAN), who I have enjoyed meeting and talking to over the years, and moderated/hosted by Rob Elder (of Outrider), who I have worked with many times (I am on Outrider’s Advisory Board). I find speaking for “only” 15 minutes almost intolerable (even when I lecture for hours I feel like I barely scratch the surface of these things), but I gave a little talk about the creation and limits of various kinds of “nuclear imagination.”

The main thing I wanted to try and get across is the ways in which “imaginations” about nuclear weapons have been created (a sort of complicated and dynamic interaction between the audiences — public and official — and the various “voices,” including experts, journalists, and “cultural” creators like artists and writers and directors), and talk a bit about the dangers of limited imaginations and ones that are based in essentially incorrect ideas. For the (mostly) journalists in the audience, my point was really to get them to see that they have a rather important role in this kind of thing (even if they are not the only “player” in it), and that a lot of things in the “nuclear world” that are taken for granted are in fact quite constructed and debatable.

You can watch the whole panel here if you are interested, but it not required! I felt like it went OK for a short talk, but it’s hard for me to know whether that is true or not, given that I am not really a member of the target audience (journalists) and I’m also not really sure what the “impact” goal of such a talk even is. But I usually wonder the same thing about talks at more academic conferences as well, and suspect most of why these things exist is as a form of networking (“here I am, this is what I’m working on, get in touch if you’re interested” could be the “ultimate” goal of pretty much all academic talks).

Perugia was interesting. I had never been to Italy before, and I am aware that it is much more than this one city, and that even Perugia must be more than the small piece of it that was hosting the conference. The conference was mostly in various venues (hotels, theaters, etc.) in the “old” part of the city, which was on a craggy hill and seems to have architecture dating from the 14th-17th centuries or so. If you like very old-seeming, somewhat decaying, very atmospheric kinds of spaces, well, they’ve got those by the spadeful. I hope you like getting lost and walking up steep hills, though, because the place is a complete maze and as mentioned built on a fairly impressive hill.

I will admit that Perugia’s charms wore out on me fairly quickly. The first few picturesque streets and cobblestone corridors were quite charming. But then they all started to blend together, and my frustration with the fact that I kept getting lost (Google Maps is just not built for this kind of environment, and it shows) made me long for the Paris’ 19th- and early 20th-century “baroque modernity.” Which was a preference/distinction I didn’t quite know I had, until I was walking around something that felt much more medieval and early modern.

There is also a way in which Perugia felt embalmed, probably because the only local economy near the “old town” was tourism, and the streets I walked in even outside of it were essentially devoid of pedestrians and the kind of pedestrian-dependent or pedestrian-catering spaces that I have started to take for granted. The whole place felt very anti-pedestrian in general, with far too many tiny Italian cars driving disturbingly fast around blind corners.

So while I found Perugia interesting to visit, and I enjoyed meeting some of the people at the conference (and I attended a few other panels, about nuclear weapons and AI weapons), but I was pretty ready to go home by the end of it. At some time in the future I will do a more “traditional” trip to Italy — Rome, Venice, etc. — because I know that I really only got a taste of it here.

While at the conference, I did manage to get a blog post out the door for Doomsday Machines, on the trope of “atomic bombs” prior to the discovery of nuclear fission:

I’ve wanted to write something up on this for awhile, because it’s a favorite of mine to lecture on, and it often surprises people. It is also the answer to a common question that I get, which is about how people conceptualized the meaning of “atomic bomb” in 1945. The answer is that they already had a conceptual framework for it, because the idea predated the actual realization of the weapon by several decades, and it is also one of the clearest cases we have where you can show that cultural products (like the novels of H.G. Wells) had an impact on the later attitudes of both scientists (Leo Szilard) and policymakers (Winston Churchill) when they grappled with an “emerging technology.”

Read more

“Ready!” said the steersman.

The gaunt face hardened to grimness, and with both hands the bomb-thrower lifted the big atomic bomb from the box and steadied it against the side. It was a black sphere two feet in diameter. Between its handles was a little celluloid stud, and to this he bent his head until his lips touched it. Then he had to bite in order to let the air in upon the inducive. Sure of its accessibility, he craned his neck over the side of the aeroplane and judged his pace and distance. Then very quickly he bent forward, bit the stud, and hoisted the bomb over the side.

“Round,” he whispered inaudibly.

The bomb flashed blinding scarlet in mid-air, and fell, a descending column of blaze eddying spirally in the midst of a whirlwind. Both the aeroplanes were tossed like shuttlecocks, hurled high and sideways and the steersman, with gleaming eyes and set teeth, fought in great banking curves for a balance. The gaunt man clung tight with hand and knees; his nostrils dilated, his teeth biting his lips. He was firmly strapped. . . .

When he could look down again it was like looking down upon the crater of a small volcano. In the open garden before the Imperial castle a shuddering star of evil splendour spurted and poured up smoke and flame towards them like an accusation. They were too high to distinguish people clearly, or mark the bomb’s effect upon the building until suddenly the façade tottered and crumbled before the flare as sugar dissolves in water. The man stared for a moment, showed all his long teeth, and then staggered into the cramped standing position his straps permitted, hoisted out and bit another bomb, and sent it down after its fellow.

Wells’ “atomic bombs” were not like the fission or fusion explosives that would come in the future. It is somewhat difficult to imagine exactly what he had in mind, as the text is not always entirely clear on it.

The section quoted above mentions an explosion with the appearance of “the crater of a small volcano.” He describes the mechanism in a bit more detail shortly thereafter:

Never before in the history of warfare had there been a continuing explosive; indeed, up to the middle of the twentieth century the only explosives known were combustibles whose explosiveness was due entirely to their instantaneousness; and these atomic bombs which science burst upon the world that night were strange even to the men who used them. Those used by the Allies were lumps of pure Carolinum, painted on the outside with unoxidised cydonator inducive enclosed hermetically in a case of membranium. A little celluloid stud between the handles by which the bomb was lifted was arranged so as to be easily torn off and admit air to the inducive, which at once became active and set up radio-activity in the outer layer of the Carolinum sphere. This liberated fresh inducive, and so in a few minutes the whole bomb was a blazing continual explosion. […]

[…]once [Carolinum’s] degenerative process had been induced, continued a furious radiation of energy and nothing could arrest it. Of all Hyslop’s artificial elements, Carolinum was the most heavily stored with energy and the most dangerous to make and handle. To this day it remains the most potent degenerator known. What the earlier twentieth-century chemists called its half period was seventeen days; that is to say, it poured out half of the huge store of energy in its great molecules in the space of seventeen days, the next seventeen days’ emission was a half of that first period’s outpouring, and so on. As with all radio-active substances this Carolinum, though every seventeen days its power is halved, though constantly it diminishes towards the imperceptible, is never entirely exhausted, and to this day the battle-fields and bomb fields of that frantic time in human history are sprinkled with radiant matter, and so centres of inconvenient rays. . . . […]

Once launched, the bomb was absolutely unapproachable and uncontrollable until its forces were nearly exhausted, and from the crater that burst open above it, puffs of heavy incandescent vapour and fragments of viciously punitive rock and mud, saturated with Carolinum, and each a centre of scorching and blistering energy, were flung high and far.

And so on, with his faux-technical terms, describing something that is both radioactive and a “continuing explosive.” To me it invokes a mash-up between a radiological weapon (contaminating with radioactivity), an open and erupting volcano, inextinguishable incendiary warfare, and the Sun.

Wells’ ideas about the atomic bomb and atomic energy are traceable to the writings of Frederick Soddy (1877-1956), the easily-overlooked British chemist who was awarded the Nobel Prize in Chemistry in 1921 for his work on nuclear transmutation with the (much more famous) physicist Ernest Rutherford. Soddy was the first major English-language popularizer of the idea of “atomic energy,” the notion that the newly-discovered phenomena of radioactivity implied that all matter contained huge amounts of latent energy locked into its structure. Whereas radioactive elements tend to leak away their energy steadily, and uncontrollably, Soddy, in popular lectures published as The Interpretation of Radium (1909), imagined a future in which that energy could be released on demand:

This bottle contains about one pound of uranium oxide, and therefore about fourteen ounces of uranium. Its value is about £1. Is it not wonderful to reflect that in this little bottle there lies asleep and waiting to be evolved the energy of about nine hundred tons of coal? The energy in a ton of uranium would be sufficient to light London for a year. The store of energy in uranium would be worth a thousand times as much as the uranium itself, if only it were under our control and could be harnessed to do the world’s work in the same way as the stored energy in coal has been harnessed and controlled. […]

As we have seen, we cannot yet artificially accelerate or influence the rate of disintegration of an element, and therefore the energy in uranium, which requires a thousand million years to be evolved, is practically valueless. On the other hand, to increase the natural rate, and to break down uranium or any other element artificially, is simply transmutation. If we could accomplish the one so we could the other. These two great problems, at once the oldest and the newest in science, are one. Transmutation of the elements carries with it the power to unlock the internal energy of matter, and the unlocking of the internal stores of energy in matter would, strangely enough, be infinitely the most important and valuable consequence of transmutation.

Soddy’s book did not speak of bombs, but its allusions to immense the power contained in the nuclei of atoms were frequent, and many of his examples of such, like the one above regarding the lighting of London, would quickly become standard clichés.1

He describes individual radioactive decay reactions as “explosive” in character, and does at one point make an unusual analogy about how much is unknown about the atomic realm at that point, comparing it to an architect being suddenly told that his everyday bricks could “could with effect be employed as an explosive incomparably more powerful in its activities than dynamite,” but he never warns of radioactivity or atomic energy as a fodder for weaponry. The jump from a new, inexhaustible source of energy to a dread weapon seems to have been Wells’ own.2

In the summer of 1922, various news syndication agencies ran a story that consisted of essentially one long quote from “one of the eminent English scientists” warning of similarly dire possibilities:

We are on the eve of scientific discoveries of so sensational and so far-reaching a character as to render Einstein’s theory, by comparison, child’s play. I predict that within the next five, ten, or at the outside twenty years, men will be able to say, “I have harnessed the atom.” […] The energy stored in such elements as thorium and uranium is stupendous. […] It only needs the knowledge of how to ignite it — how to cause the atoms to break up when we desire — to make this power immediately available. […]

It is conceivable, too, that such terrific force might eventually be liberated as to blow up the world.

Then consider the possibilities of war. The first nation to discover the secret will be in a position to wipe out all the other nations, literally, in a quarter of an hour. It could send over an aeroplane with a 2,000-pound bomb which would have as devastating an effect as that of a million aeroplanes carrying the 2,000-pound bombs in use today. […]

The question is: Which will that nation be?

This article circulated globally, at times accompanied by dramatic headlines, like “EARTH AS A BOMB.”3 Some of the accounts say it was an “eminent Oxford scientist” who provided the story. But who was the “eminent” scientist? It is quite unclear. That the article ran anonymously suggests that whomever the source was, they didn’t want their name publicly attached to the idea. To my knowledge, no one ever came forward later to claim their prescience as the source after the fact.4

In interwar September 1924, Winston Churchill tried his hand at writing about new technological futures in an article with the odd title, “Shall We All Commit Suicide?” Influenced by both H.G. Wells — Churchill claimed to have read all of Wells’ books since The Time Machine immediately as they came out, and multiple times — and his close scientific friend and advisor Frederick Lindemann (Lord Cherwell, Oxford professor), Churchill laid out a his grim sense of coming conflict and the role that science and technology might play in it:

The story of the human race is War. Except for brief and precarious interludes, there has never been peace in the world; and before history began, murderous strife was universal and unending. But up to the present time the means of destruction at the disposal of man have not kept pace with his ferocity. Reciprocal extermination was impossible in the Stone Age. One cannot do much with a clumsy club. […] It was not until the dawn of the twentieth century of the Christian era that War really began to enter into its kingdom as the potential destroyer of the human race. The organization of mankind into great States and Empires and the rise of nations to full collective consciousness enabled enterprises of slaughter to be planned and executed upon a scale and with a perseverance never before imagined.

Churchill warned that the peace of the Great War was likely a temporary one, as “two mighty branches of the European family will never rest content with their existing situation” — Russia and Germany. And just as surely as war will return, Churchill wrote, the technology of mass destruction will continue to evolve:

“Wars,” said a distinguished American to me some years ago, “are fought with Steel: weapons may change, but Steel remains the core of all modern warfare. France has got the Steel of Europe, and Germany has lost it. Here, at any rate, is an element of permanency.” “Are you sure,” I asked, “that the wars of the future will be fought with Steel?” A few weeks later I talked with a German. “What about Aluminium?” he replied. “Some think,” he said, “that the next war will be fought with Electricity.” And on this a vista opens out of electrical rays which could paralyse the engines of a motor-car, could claw down aeroplanes from the sky, and conceivably be made destructive of human life or human vision. Then there are Explosives. Have we reached the end? Has Science turned its last page on them? May there not be methods of using explosive energy incomparably more intense than anything heretofore discovered? Might not a bomb no bigger than an orange be found to possess a secret power to destroy a whole block of buildings — nay, to concentrate the force of a thousand tons of cordite and blast a township at a stroke? Could not explosives even of the existing type be guided automatically in flying machines by wireless or other rays, without a human pilot, in ceaseless procession upon a hostile city, arsenal, camp, or dockyard?

Churchill further worried that such weapons could allow “a base, degenerate, immoral race” of people to “make an enemy far above them in quality” their slaves on the basis of their possessing and being willing to use novel technological weapons.5

I don’t believe that Churchill’s article had a major public impact on the thinking about “atomic bombs,” but it doesn’t really need to have: it reflects that this discourse about science fiction weaponry had made it to what would very soon be the highest echelons of public policy in at least the United Kingdom (Churchill was on the political “outs” in 1924, but would soon come back), and joined into security debates. Churchill, Franklin Roosevelt, and Adolf Hitler in particular shared a common belief in the value of “wonder weapons” produced by scientific research, a shared “lesson” of the Great War for all of them. It is part of the context of why each of these nations were willing, to differing degrees, entertain the possibilities of fantastical wartime research, including atomic research.

There were other minor references to “atomic bombs” across the 1920s and 1930s, prior to the discovery of nuclear fission. The most important and perhaps famous inspiration was that of the Hungarian physicist Leo Szilard, who famously credited H.G. Wells for his development of the concept of the nuclear chain reaction in 1933, over five years before the discovery of nuclear fission. He later described it this way:6

In 1932, while I was still in Berlin, I read a book by H. G. Wells. It was called The World Set Free. This book was written in 1913, one year before the World War, and in it H.G. Wells describes the discovery of artificial radioactivity and puts it in the year of 1933, the year in which it actually occurred. He then proceeds to describe the liberation of atomic energy on a large scale for industrial purposes, the development of atomic bombs, and a world war which was apparently fought by an alliance of England, France, and perhaps including America, against Germany and Austria, the powers located in the central part of Europe. […] The book made a very great impression on me, but I didn’t regard it as anything but fiction.

I was no longer thinking about [H. G. Well’s book] until I found myself in London about the time of the British Association [meeting] in September 1933. I read in the newspapers a speech by Lord [Ernest] Rutherford. He was quoted as saying that he who talks about the liberation of atomic energy on an industrial scale is talking moonshine. This sort of set me pondering as I was walking the streets of London, and I remember that I stopped for a red light at the intersection of Southampton Row. As I was waiting for the light to change and as the light changed to green and I crossed the street, it suddenly occurred to me that if we could find an element which is split by neutrons and which would emit two neutrons when it absorbed one neutron, such an element, if assembled in sufficiently large a mass, could sustain a nuclear chain reaction. I didn’t see at the moment just how one would go about finding such an element, or what experiments would be needed, but the idea never left me. In certain circumstances it might become possible to set up a nuclear chain reaction, liberate energy on an industrial scale, and construct atomic bombs. The thought that this might in fact be possible became sort of an obsession with me. […]

In the spring of 1934 I had applied for a patent which described the laws governing such a chain reaction. This was the first time, I think, that the concept of critical mass was developed and that a chain reaction was seriously discussed. Knowing what this would mean—and I knew it because I had read H. G. Wells—I did not want this patent to become public.

It isn’t that Wells’ book gave Szilard any strong technical inspiration — Carolinum, it isn’t. Szilard’s neutron-based chain reaction is his own. But Wells did clearly provoke Szilard’s imagination about atomic bombs, and helped frame how he thought they might matter if they were possible. And thus Szilard’s first impulse was towards secrecy, towards believing that such a chain reaction would be vastly important development for humankind. And that impulse set off its own historical chain reaction, leading to Hiroshima, the H-bomb, and beyond.7

Szilard didn’t have a candidate reaction for his concept of the chain reaction, and while he did some research to search for one, he ended up essentially shelving the question for several years. Nuclear fission was discovered by Otto Hahn, Lise Meitner, Fritz Strassman, and Otto Frisch in late 1938, and became global news in early 1939. The moment that Szilard heard about the discovery, he was ready to understand what it might mean:8

[The physicist Eugene] Wigner told me of Hahn’s discovery. Hahn found that uranium breaks into two parts when it absorbs a neutron; this it the process which we call fission. When I heard this I saw immediately that these fragments, being heavier than corresponds to their charge, must emit neutrons, and if enough neutrons are emitted in this fission process, then it should be, of course, possible to sustain a chain reaction. All the things which H. G. Wells predicted appeared suddenly real to me.

This is also one of the clearest examples we have of the complicated relationship between scientific research (radioactivity, transmutation), scientific popularization (Soddy), science fiction (Wells), a feedback loop back into scientific inspiration (Szilard), and broader policy decisions (Churchill).

The pre-fission history of atomic bombs is important, not just for Szilard, but for a vast number of other people, including scientists, statesmen, and the everyday public. It created a context for understanding what “atomic bomb” even meant prior to their invention, and towards thinking about the idea of a vastly powerful science-borne weapon as a possible concrete reality. It is why, in 1945, when the United States announced that it had invented an “atomic bomb,” people understood immediately that this meant a vastly powerful, possibly game-changing weapon, and not, say, a very tiny explosive.

The atomic bomb in Wells’ book may have been “strange even to the men who used them,” but Wells’ book made the concept considerably less strange than it might have otherwise been.

This wasn’t quite Russia–Ukraine 2022 levels of usage, thank goodness, but it’s still a pretty significant spike. It is also interesting to me how quickly it went down to baseline afterwards — a graphical rendition, perhaps, of how people react to nuclear crises in general, where there is a sudden burst of attention and then a return to not worrying much about it at all.

NUKEMAP did pretty well in that period, although there was some performance degradation for the parts of it that require Amazon EC2 and RDS instances to function (casualty calculation and loading some dynamically-modified configuration data). It has given me some ideas as changed to make so that NUKEMAP can still load its basic functionality even if the Amazon assets are knocked out by too much traffic, which I’ll implement when I get the chance.

And how did you do? I found the whole thing very alarming. Not so much because I thought the US was about to use nuclear weapons — although the inability to rule that kind of thing out was troubling — but because I did think that there was a real danger that the US was about to commit a very terrible massacre of civilians. Trump’s language about killing “a whole civilization,” described as “grave” in some headlines where the word “genocidal” should have been used, is unprecedented for a sitting president. It is reprehensible, a war crime, and an impeachable offense by itself: whatever his intent, the language was meant to invoke a genocidal threat.

I was asked about this by a New York Times reporter, and had some rather strong things to say about it. They only quoted one of my least strong statements, which perhaps is for the best:

Alex Wellerstein, a historian who studies nuclear conflicts, said that even if Mr. Trump does not carry out the extent of his threat, the president’s violent rhetoric damages his credibility as a negotiator and his country’s standing in the world.

“You’re talking about a world that largely increasingly sees the United States as unhinged and dangerous, and not a reliable partner,” he said, “where all of the countries that typically align with democracy and freedom are on the other side of the United States.”

The more substantive thing that I said to them was that while one can point to instances of sitting presidents making nuclear threats, the truly (and intentionally) aggressive ones have been done in private, not in the public sphere. And the reason for this is that such threats a) terrify one’s own populace, b) terrify one’s allies, and c) there is a good chance your enemy will call your bluff. All of which were on display in this case.

Was Trump’s rhetoric meant as a nuclear threat? I don’t know. I did also tell the Times reporter that a) if the words had any intentionality behind them at all, they were meant to invoke specter, and b) it didn’t really matter what the intentionality was since any reasonable audience would (and did) consider that to be a plausible interpretation of them.

Intentionality does matter a bit, particularly when we are talking about words perhaps taken out of context, or goaded out of someone. When Truman made comments about the atomic bomb being under “active consideration” at a news conference during the Korean War, that was a case of a news story being manufactured out of nothing. That news story did succeed in terrifying both the US population and its allies (the UK immediately requested an audience with Truman), but it also led to an immediate clarification and rebuttal by the White House. (If you are curious about this incident, I discuss it at some length in The Most Awful Responsibility. Even at the time it was clear that this was a “slow news day” problem.)

In the case of Trump, there is no such excuse. That does not mean that I credit his intentionality all that highly, because I think his grasp on language is pretty poor, but it is hard to imagine a world in which he (or his allies) would interpret such language differently if it was coming from the Iranians.

Well, now that can has been kicked down the road some more. Coincidentally, my post on Doomsday Machines this week was about looking at what US nuclear war plans were in the late 1950s, through the lens of The Power of Decision, an in-house production for the USAF about how its bomber forces would coordinate to destroy Communism from 1958:

I’ve been looking at The Power of Decision for a long time, particularly at “the Big Board” and all that it implies. The restored version by Peter Kuran made finalizing my thoughts on it much easier, as there were a number of very basic things (like “what does the legend on the board say?”) that were unanswerable without better resolution than was previously available to me.

This upcoming week I will be out of town for part of the week, so I’m not sure what I’ll get posted for Doomsday Machines, but I have a few things in a “partially finished” state that I’ll try to turn around before I head on the road. Here’s hoping for that, among other miracles!

Read more

Just this week, the filmmaker/special effects artist Peter Kuran, of Trinity and Beyond fame, released a restored version of the film that is such an improvement over the muddy, pixellated version that had been made available to the NSA. You can view it in its entirety on his YouTube channel:

There is so much one can say about it, but for the moment I want to focus on one aspect of it that I have been obsessed with for some time: its war plan. There are three war plans discussed in The Power of Decision: WAR DANCE, which is a described as “a typical war plan for our heavy units”; QUICK STRIKE, which is the dramatized war plan shown in the bulk of the film, and the one shown in the most close-up detail; and FOLLOW THRU, which is an operation that was meant to be running directly after QUICK STRIKE. Because it is the most detailed and most carefully discussed, I am going to focus entirely on QUICK STRIKE.1

I have been pouring over the details of QUICK STRIKE for some years now. I put off writing this post until after Kuran’s restored version was posted (I had gotten wind of it a bit earlier) because I thought it would help me resolve a lot of unanswered questions to be able to see the “Big Board” — yes, that is what it is called — more clearly. Kuran quite generously also allowed met to access a version without a watermark for this post.

What has made QUICK STRIKE a fascination of mine for so long is that it is devilishly hard to get information on “real” war plans for any period, much less this bomber-centric one, in a period in which the US had vast forces to deploy. We have had previous releases that gave a sense of targets, and we know where many of the weapons were located, and we know what some of the planned “movements” were, and we know what the general philosophy of these planes were… but it’s another thing to see something that is meant to be a “realistic” nuclear war plan, made for internal, classified consumption. It is, we should note on the onset, not an actual nuclear war plan — as I will discuss, many of the details are clearly fuzzed/fabricated — and it is, even within its own context, meant to be a future war plan, but I believe it does reflect much of what we know war planning in this period looked like, and vividly so.

I’ve created a mosaic out of all of the screenshots of the Big Board showing QUICK STRIKE in the film. This was done with over a dozen carefully selected images of the restored film and trial-and-error with Photoshop’s Photomerge feature, careful hand-masking of layers to get the highest resolution possible, and then an overall straightening of the piece. Here it is in its “raw” form:

This map does not read itself. There’s a lot going on in it. The black spots are not redaction marks — they are just areas that do not appear in any particular frame, or if they do appear, it is at such a distance (and they have so little information in them) that it just distracts from the overall image (e.g., that small area in the Pacific, which you can see in the first film screenshot above — fitting in that extreme angle into this was not within Photomerge’s capabilities, and it did not seem worth it). I have not “filled in” any missing areas for the sake of aesthetics.

I have also made an annotated version of the above mosaic, which I will use to help “decode” and analyze the plan below. I will explain the various annotations in turn.

First, our orientation: this is a polar projection, centered on the North Pole, looking “down” at the planet. This is common for Cold War “face off” visualizations for reasons that are somewhat geo-strategically obvious: a “side-on” project (e.g. Mercator) projection obscures the distances quite a bit when talking about travel over or near the poles (especially Mercator, which dramatically distorts near the poles). So to get your bearing, take note of the North Pole (where the thick vertical black line and a thin horizontal meridian meet), and look at North America in the lower right, and Eurasia in the upper left.

There a few different pieces of information shown on this map. First is, of course, the map itself, with its representation of geographic locations. Note that the Soviet Union, the People’s Republic of China, and the Warsaw Pact states (Eastern Europe) are essentially undifferentiated and treated as one large, pink-tinged entity. We know that the war plans of this period did not allow for “de-selecting” China or other countries — it was assumed that all of Communism was the target, and the SAC planners actually resented and feared the idea that a civilian authority (like the president) would ever want to have a more “limited” plan, as Daniel Ellsberg writes disturbingly about in The Doomsday Machine.2

There are two additional sets of borders added to the “base” map: early warning radar lines (or “nets” as they say in the film). The black line with red and white borders around Eurasia is the Soviet warning network, and the yellow-on-red line around North America is its American equivalent (i.e., the Distant Early Warning Line and other radar stations).

Let’s zoom in on a screenshot of the United States:

Dotting the Allied countries are red and black dots. The black dots are “launch bases”: bases that American attacks would emanate from. The red dots are “recovery bases”: friendly airfields that any plane that has survived its attacking run, or any damaged plane, could land at after their attack, potentially refueling and then participating in a future attack wave (e.g. FOLLOW THRU). In the above, you can see that all of the bases in the continental United States are “launch bases,” and that the only “recovery” bases visible are in Alaska and Canada. In the larger mosaic, you can see lots of “recovery” bases abroad (more on bases abroad in a moment).

The lines that emanate from the black dots and make their way to the Soviet Union are the attack themselves. They come in six colors: black, blue, red, green, orange, and yellow. Some of the lines converge at large colored circles with shapes on them; these are aerial refueling points, and each of them have their own code names: SUN SPOT, ROBIN HOOD, RAIN DROP (shown over northern Canada above), HOT ROD, HEN PECKED, and LIL’ LADY. Each of the different “plans” has different code names for its refueling points. These are where outgoing long-range bombers would rendezvous with aerial refueling planes, like the Boeing KC-135 Stratotanker, and take on more fuel before making their bombing run.

One important thing: the bases marked in the United States are mostly nonsense. The location of SAC bases in the USA is not (and was never) secret, and we also know (from declassified reports) where SAC stored nuclear weapons in 1958. A few of those “launch bases” are correct-enough — e.g., Amarillo AFB, March AFB, and Biggs AFB are in cities marked as bases — but a lot of actual bases are missing, and a lot of the labeled bases are just totally wrong/nonexistent. The base locations in The Power of Decision are bogus. This is something I can say with much more authority now that we have the restored version of the film, as it was harder before to see really where everything was located. Again, this is meant to be evocative of an actual war plan, and I think it is that, but it should not be mistaken for an actual war plan.

What do the colors of the lines mean? To the left of the big board, behind our narrator, is a table of lines of the same colors showing their flight times to targets. It’s not easy to see the whole thing in detail, but you can see some of it in the above screenshot. Each line in the table is labeled with one of three acronyms: BW, GMS, and SRW, each of which also has either a number (e.g. 203 SRW) or letter (e.g. “E” GMS) in front of it. So the group of lines at his elbow, above, are 491 BW (blue), 581 SRW (black, very long), “E” GMS (short yellow), and 10 BW (blue).

In Air Force lingo, a BW is a Bomb Wing, a SRW is a Strategic Reconnaissance Wing, and a GMS is a Guided Missile Squadron. The coloring of the BW and SRW lines appears to be inconsistent and without any obvious rhyme or reason; I am not sure how you would correlate the table on the left with the map on the right at all. The GMS lines are always yellow, and the short timing on the table is pretty consistent with them being surface-to-surface missiles of the sort that were just going online in the late 1950s, like the AGM-28 Hound Dog. The yellow lines are also the only ones that never branch.3

The thicknesses of the lines also seem rather arbitrary. Sometimes a thick line branches into many smaller lines, implying the thickness correlates with the number of planes, but a lot of the times thin lines also branch to similarly thin lines, and some thick lines do not branch at all. Two massive red lines that emerge from the eastern USA intersect at refueling point HEN PECKED, combine, and then just as abruptly end at the Soviet early warning line. Only red, orange, and green lines are ever thick — blue lines never are. Some of the red, orange, blue, and black lines are spindly thin. What does it mean? Perhaps nothing, in this case, but it would make sense if these lines did have meaning in an “actual” war plan, and not this “evocative” version of one.

Above is a detailed screenshot of Western Europe, which I have flipped on the vertical axis so it is oriented with north being up. I want to use it to talk about both the foreign bases, and the targets.

First, on foreign bases: as you can see, there are launch bases and attack lines emanating out of bases outside of the United States. In Europe, these are set in the United Kingdom, Spain, Morocco, and Libya. As with the US detail, there are some things about this that are correct, some things that appear to be erroneous. The UK sites do not align with actual SAC bases at the time very well. The Spanish lines appear to emanate from Torrejón, but are missing the SAC bases at Morón and Zargoza. The Morrocan lines appear to emanate from a place far to the east of the actual bases (Sidi Slimane, Ben Guerir, and Nouasseur); the left-most red dot (a return base) could be Ben Guerir, but the black dots look like nonsense. Wheelus air Base, in Libya was an important coordinating base for SAC, but did not host nuclear weapons per its own internal (now declassified) lists.

One could try to make excuses for this: the verbal description of the plan in the film says that aircraft have already been deployed to forward bases, and that can mean that weapons have been moved around prior to attack. But I think it is far simpler and more likely to just assume that the map locations were intentionally obfuscated by people who wanted the attack to resemble an actual strike (with plausible countries and locations) but were not going to give actual detailed target or base information to the film crew if they didn’t need to.

Similarly, the orange lines coming out of the Azores (another detail that had confounded me until the restored film was available) are in a similar category as Wheelus: as far as I know, the Azores did not store nuclear weapons, but do have important US air bases in them.

The launch bases in Okinawa, Guam, and Greenland (Thule) do seem accurate. It is less clear with (mainland) Japan what is going on — are the bombers rendezvousing there and then continuing on, or are they meant to be actually based in Japan? Are they Strategic Reconnaissance or bombers? Japan did not (willingly) host nuclear weapons during the Cold War, although it is clear that the US did host them at Okinawa and some other US bases at times. But definitely not, say, in Tokyo, as indicated on the map.

For the sake of comparison, I’ve created a quick little map, below, that approximates the same map projection as the screenshot of the United States above, with actual SAC bases, including nuclear-armed ones, shown as dots, for around the period of the film’s creation and its near-term projected future. This is not a complete map of SAC deployments at the time: the projection necessarily leaves out many, such as the weapons deployed in French Morocco (Ben Guerir, Nouasseur, and Sidi Slimane), Greenland (at Thule), Guam, Okinawa, Spain (Morón, Torrejón, and Zaragoza), and the United Kingdom (at Brize Norton, Fairford, Greenham Common, Lakenheath, and Mildenhall).4 The point here is that one can readily see that these maps do not have much in common — the data does not line up, except in a few places that might just be random.

The targets of the attack similarly look somewhat vaguely accurate, but also somewhat bogus. Yes, Moscow was heavily targeted — no secret there! As are a few other major areas, like Leningrad (St. Petersburg), Tomsk, and what looks like areas in East Germany, Poland, and Soviet Ukraine. But there are a lot of more specific targets that are conspicuously missing. The scale and scope of US targets against the Soviet Union in the late 1950s is known to have been vast.

It is also difficult, I would note, to judge exactly what is targeted. The lines of the planes sometimes end in “target icons,” but more often than not they just end in some part of the map without anything indicated. The rhetorical stance of The Power of Decision is that most of the targets are defensive or counterforce, with not much discussion of the targeting of cities or even industry. Magnitogorsk, a major site of Soviet industry, is not indicated with any target labels (you can see its label on the map), but there are some bombing routes that terminate near it. I do not suggest, again, that these bombing routes or labels are accurate, but just suggest that the plan as displayed seems to be far more extensive than the target labels would let on, presuming those planes aren’t just going to those unlabeled spots on the map and simply turning around to go home again.

The labels for the target types are mostly pretty straightforward — and primarily counterforce. “SAM” (large black circles) are surface-to-air missiles. “F & T Control” (yellow circles) are sites that coordinate fighters and tactical responses, and possibly early warning radars stations. “Control” (maroon squares) appears to be “government control” (see below) which suggests command and control sites. “A & TN weapons sites” (black squares) are nuclear weapons storage sites. “LRAA HQ” (purple triangles) are Long Range Air Army headquarters, where LRAA is essentially the Soviet equivalent of SAC (at least in SAC documents from the time). “Liquid fuel” (large gray circles) storage sites and “transportation centers” (large red cirlces) speak for themselves. The only strange one, not explained at any point in the film, is “M Center” (green circles). I had wondered if it was “military centers” but based on their placement — along the Soviet/Chinese borders, except to the south — my guess is that the M standards for “missile.”

Later in the film, they do a “preliminary strike report” (above) with categories that mostly overlap with the map categories. You can see that a single category, “Missile Sites,” has perhaps absorbed both “SAM” and “M Center.” We also have a clarification that “Control” is probably “Government Control.” We also have two categories not indicated explicitly on the map: “air supp[ort] industries” and “war sust[aining] industries.” Again, all of the rhetoric of these targets is about counterforce targeting.

Lastly, let’s talk about the timing of this attack. The film explains that QUICK STRIKE is meant to elapse over 20 hours or so — QUICK by the standards of world war, to be sure, but slow by the standards of the missile age. Those concentric black lines going into the interior of the USSR, with H+0, H+1, etc. indicated on my annotated version, are estimates of a bomber’s flight time once it has hit the Soviet early warning net. So in principle a bomber could make it all the way to the heart of the USSR in 4 hours. In practice, many of them would not be trying to do that, and many would probably be shot down by enemy defenses. But, of course, they’ve planned for that. As our narrator explains:

Remember, this is a “bomb as you go” concept. Targets are bombed progressively, shallow targets first, with the strike continuing to deeper targets. This enhances the penetration capability of succeeding waves. This pattern is essential because we are playing the Soviet’s game. He picked the time of day, he knows we’re coming, we must use every trick to overwhelm him.

So if you bomb one target and then get shot down, that’s one less target for the bomber behind you to worry about.

There is no mention in the film of targeting Soviet cities, and no attempt made to estimate enemy casualties. The only casualty estimates are American ones: 60 million “personnel casualties,” 20 million wounded and 40 million dead. The 1958 US population was around 175 million, so that is a third of the country rendered dead or injured.

As a point of comparison, in the introduction to The Doomsday Machine, Daniel Ellsberg described an official report he received in 1961 that he received which suggested that the deaths in the Soviet Union and China from an all-out thermonuclear attack by the United States would start at around 275 million at the moment of war, and rise to 325 million over the next six months. Including deaths in Eastern Europe, as well as collateral deaths from fallout from just American attacks in other nations, would raise the total deaths to around 600 million, “a hundred Holocausts,” as Ellsberg puts it.5

So what is the take away of all of this? The specifics of the QUICK STRIKE war plan shown in The Power of Decision are certainly made up for the purposes of the film. But the general feel of the war plan is probably correct: in the late 1950s, the United States could, within a few short hours, mobilize a bomber force so vast, and with so much firepower, that it could exterminate the populations of multiple nations. And, in fact, this system was so rigidly constructed that they really did not have any options to limit that destruction, even if someone (say, the President) had wanted to.

The rhetorical stance of the film is meant to one of responsibility and constraint: in the film, the Soviets have started World War III, but the United States is going to end it. The actors — and again, they are all actors — do not appear to ever relish their role in the war, but they also appear satisfied to grimly carry it out. There is no hand-wringing. There is no, “my god, what have we done?” moments, as they contemplate a slaughter that would out-do the Holocaust by an order of magnitude. “Sir, we have failed in our primary mission: deterrence,” a colonel briefing on the outcome of QUICK STRIKE begins, with a dark tone to his voice. “However,” he continues, brightening, “our second mission, of destroying the enemy, is apparently fulfilled.” It’s more than a little chilling.

I’ve been teaching with the diagrams of QUICK STRIKE for over a decade now. I think there’s real value in showing how the abstractions of deterrence theory were rendered into a horrific, tangible, effective killing machine at the height of the Cold War. I sometimes have students who (adorably, naively) ask: “but would they actually have used the bomb in this way?” And being able to point to this proud, in-house, top-secret, self-congratulatory film is one easy way to answer that question. It sure looks like they would have.

There is something, too, I think, in the fact that these 1950s plans were all bomber-based: that human beings, at every step of the way, would need to carry out this horror, this atrocity. It was not a “push button” affair; while the system’s wheels were well-oiled, as the film makes clear, it was hardly automatic. You can’t blame the logic of machines for this one.

When I call this kind of thing an atrocity, I am not saying that the people who designed it, or served in it, were themselves evil. I understand full well the logic of deterrence: build the system, and perhaps it will never have to be used. But if you have, indeed, failed at your mission of deterrence — and the film is explicitly based on the premise that deterrence has failed — then what have you got? What moral shelter can one take, as one sets into motion the largest act of civilian slaughter in human history?

But I’m not complaining. I am glad I could make the trip. Joshua Tree was quite beautiful, and (believe it or not) I had never been there before. (I am from Central California, and while I have spent some time visiting Southern California over the years, there is much I haven’t seen there.) The vistas reminded me of photos of the Nevada Test Site — scrubby desert broken up by rocky mountains. Quite beautiful in its own, austere way.

I saw a coyote, and several chubby rodents (which may be White-tailed Antelope Squirrels), and many birds (but no roadrunners). And there were flowering cacti in abundance. I found myself wondering how this landscape would look to people crossing it in ye olden days — would it be as beautiful if one didn’t have easy roads in and out of it, or would it simply look intimidating?

The family event was lovely (I will not be going into detail on my blog, thank you very much!), and despite the wear-and-tear imposed by this kind of travel I am glad I could make it work out. I have not driven a car since coming to France, and I believe I more than made up for it on this trip, spending some 7 or 8 hours behind the wheel total on Los Angeles freeways.

I have to admit that I felt that the vision of Los Angeles I saw on this trip did not compare very favorably to Paris — roads, roads, and more roads, moderately (for LA) busy, with most people driving monstrously large vehicles, and an endless stream of garish LED billboards advertising primarily the services of accident lawyers and online gambling. These things do not exist in Paris (billboards, for example, seem to be very heavily regulated there, in both size and content). I found myself thinking about how a born-and-bred Parisian would think they’d landed in some kind of cyberpunk wasteland dystopia. But I am aware that Los Angeles is more than just this, and that staying in a motel near LAX for two nights did not really allow the city to put its best foot forward.

And while this was not a work trip, I did squeeze in one bit of “work” while I was out here — I gave a talk at the RAND Corporation in Santa Monica, where my longtime friend and fellow nuclear historian Edward Geist works. I had never been to visit RAND before, and so the experience alone was an valuable one just for the taste of it. I was highly amused, for example, to see that the city of Santa Monica had erected a statue of a mushroom cloud, made out of chains, across the street from RAND’s campus (above).

RAND was quite interesting. The people I met were about what I expected: earnest and smart, perhaps a slight overrepresentation of people on the spectrum, with good questions about the material. And they were also quite interested in talking about NUKEMAP, which is always fun for me, particularly because so much of the work I do on that is relatively “invisible” to the everyday user (“behind the scenes” stuff). I was also able to take a look at at their library, which included a lot of interesting early works by RAND luminaries.

The talk I gave on the Castle Bravo accident, essentially similar to the one that I mentioned some weeks ago. I’ve been writing it up as a formal academic paper, and have found some new documents that further my argument a bit more. The most amusing of which is that I believe I have traced the erroneous, pre-Bravo theory of “stratospheric trapping” — the idea that megaton-range detonations would not have any significant local fallout because their mushroom clouds would rise to such a high altitude, which was one of the two erroneous approaches to thinking about fallout that led to Bravo accident — to a possible initial author. And, dear reader, you will be as shcoked as I was to find it may have originated as early as 1950 from none other than your friend and mine, Edward Teller.

I was asked: how did I find such a thing? The answer is just that I have many, many, many reports and documents on my computer, and occasionally I do things like search all of them for the word “stratosphere” and go over the results (in this case, I have about 1,600 PDFs with that word in them, but by glancing at the file names I can rule out a lot of them). That might sound laborious but it’s a lot easier than wading through archival boxes, let me tell you. And I have enough success with this method that it feels a bit like a game, like I’m mining for treasure. And the worst-case scenario is that I find something unrelated-but-also-interesting!

I love the serendipity of this kind of research. Once one has found something to focus on, sure, it is good to dig into it, but wading through lots of different things can turn up so many unexpected results. Which is what happened with this week’s Doomsday Machines (posted on Saturday, because my jet-lagged brain just didn’t have the bandwidth on Friday to wrap it up), which is on the role of organized religion in US Civil Defense planning:

That this is coming out on Eastern weekend is, I assure you, a total coincidence (I am pretty unaware of religious holidays, and would not have even known it was Easter if someone hadn’t mentioned it). I stumbled across the two pamphlets discussed, “The Clergy in Civil Defense” and “The Church and Civil Defense,” while looking for information about something else (the “something else” will be a future post). Then I started digging into the personages involved, particularly this Fred W. Kern, and was quite delighted to find that he had made Civil Defense into a holy ideological struggle (what fun) and also that he was FCDA Administrator Val Peterson’s brother-in-law (which was not obvious). The latter I was tipped to by the fact that in Val Peterson’s post-gubernatorial papers, Kern’s correspondence (which I haven’t seen) is included in the “Peterson family” collection, and with some digging I was able to find that Kern had been married to his sister. Not a major research accomplishment, but satisfying nonetheless.

Read more

The first, from October 1951, was titled “The Clergy in Civil Defense” (AG-25-1) and prepared under the supervision of a Religious Advisory Committee of the FCDA.1 Its use of the term “clergy” is, it clarifies, meant to be pan-religious — “the clergy of all faiths” — although the committee was composed of four Christians (of different sects) and one rabbi.

The role of the clergy (thus defined) is divided into two phases: “pre-attack” and “post-attack.” Pre-attack is perhaps the most interesting: they are charged with the task of reinforcing a particular ideological stance:

(a) Maintenance of firm resolution and composure of the people through preaching and worship.

(b) Through pastoral care of families and individuals, creation of a sense of responsibility and spiritual security among the people to meet the possible exigencies of attack on a scale never experienced before in the United States.

(c) Maintenance of courage and spiritual well-being of children through regular programs of religious education (especially in respect to parent groups and teachers).

(d) Encouragement of the enlistment of volunteers for the civil defense services.

(e) Acting as chaplains to the various civil defense services such as warden, rescue, and fire fighting, during training and in actual operation.

(f) Planning, organizing, and preparing services of the clergy and other religious activities in time of attack or other emergency.

(g) Planning and arranging the part of the clergy and members of religious organizations in mutual aid and mobile support programs.

(h) Planning for religious ministry in evacuation reception areas.

(i) Planning with chaplains of adjacent military installations for cooperation and mutual assistance in case of attack.

The first four in this list strike me as quite different from the rest. It is one thing to conclude that pre-attack (or pre-disaster) preparation means thinking about the organizational responses to the attack (or disaster). But the first three are about creating some kind of spiritual or psychological sentiment in their congregations, which seems to go well outside the normal role for the federal government to recommend. An encroachment of “turf,” one might say. And the fourth, actively encouraging participation in the Civil Defense program, assumes that said program is in fact in alignment with the beliefs of these different organizations — something, perhaps, that they could take for granted in 1951, but, as we will see, becomes a trickier assumption as the decade continued.

In the post-attack phase, the role of the clergy shifts to being of “great help in sustaining and restoring the morale of the community.” This is pretty much what you would expect: they are to administer religious rites for “for the seriously injured, dying, and dead,” to conduct burial services (“including possible mass burials”), comfort the survivors, lead ministry and worship “where appropriate,” attend to the religious needs of civil defense workers, and participate in “mutual aid” activities as is possible and appropriate. All of which strikes me as quite practical by comparison to the pre-attack directives.

In 1954, when “The Clergy in Civil Defense” was issued, Civil Defense programs were not yet especially controversial. There are a variety of reasons one could point to for this — the thermonuclear age had not yet arrived, the threat of atomic war against the continental United States still felt somewhat remote to most Americans, and Americans could by and large simply ignore Civil Defense if they thought it was not for them. The amount being expended on it was never large, and it largely did not intrude into American lives one way or the other with the exception of some mild indoctrination for children like “Duck and Cover” drills.

Starting in 1954, however, this situation changed significantly. The Castle Bravo nuclear test accident made it quite evident that the thermonuclear age was at hand and it meant that the destructive capacity for nuclear war had increased dramatically. Both the US and the Soviets began speaking publicly, in much more dramatic terms than before, about what nuclear war would mean. And the FCDA and local Civil Defense authorities began a series of national drills, known as “Operation Alert,” in which they mandated that everyday citizens participate in mock Civil Defense exercises.

The latter point is probably the most important, from the perspective of public opinion about Civil Defense. When your city shuts down for a few hours so that you can pretend you have been obliterated, it forces one to take stock of the situation. Political scientists would say that it increases the salience of the risk: it goes from something abstract to something tangible.

And while most people went along with the exercises, one finds the emergence of a small but significant group of people who began objecting to them on both practical and moral grounds. When the 1955 “Operation Alert” exercises came around, Dorothy Day and the Catholic Workers of America refused to participate, instead handing out pamphlets:

We will not obey this order to pretend, to evacuate, to hide. In view of the certain knowledge the administration of this country has that there is no defense in atomic warfare, we know this drill to be a military act in a cold war to instill fear, to prepare the collective mind for war. We refuse to cooperate.

Over two dozen people were arrested for their non-participation in the drill. While the anti-Civil Defense arguments would broaden over the years — by 1960, over a thousand people would be arrested in NYC for protesting the drill — the fact that this initial argument against Civil Defense was rooted in a religious pacifism, and a specifically Catholic organization, is what strikes me as being interesting in this context. The 1951 pamphlet and its authors assumed, perhaps correctly for its moment, that Civil Defense was religiously uncontroversial. Three years later, however, serious religious objections to Civil Defense had emerged, and consensus could not be taken for granted.

Which is interesting context to consider for the revised version of pamphlet AG-25-1 that was released in October 1956, now re-titled as “The Church and Civil Defense” (AG-25-1 Revised).2 Gone, apparently, is the council of religious advisors — in fact, no author is listed whatsoever. Instead, it appears that the FCDA Administrator since 1953, former Nebraska Governor Val Peterson, created a Religious Affairs Office which consolidated the functions of the previous council and made them a more integrated part of the FCDA. Heading this office, since 1954, was one Dr. Fred W. Kern, a Lutheran reverend — and Peterson’s brother-in-law — who ran the office for the next seven and a half years.3

What was Kern’s agenda? Kern gave a talk at a Civil Defense conference in 1956 describing in vivid his approach to the job, and the world:4

When I came with civil defense, no one knew exactly what I should do and I didn't either. I think there was in the minds of some people the idea that the church is such a big institution and has so many millions of people that somehow we ought to get them behind us — we ought to get the church to support civil defense . Now, how are we going to do that? We know very well that everyone in any public office knows that you are not going to go to the church and tell them the government wants them to do so and so, the government says that this is what you ought to do If you ever tried that approach, you know from very clear experience that the opposite may have happened, because the church in this country is free, and that is one of our cherished freedoms. We don't have the government interfere in any way with what we do in our religious life, so long as it isn't contrary to the interests of our government or our citizenship.

I had to do some studying on this business of civil defense because I knew very little about it. I started reading books on civil defense and on communism. Naturally when we think about the question, “Why do we have civil defense," we conclude, "Well, we have to protect ourselves, we have to be ready to live, to preserve our life, we have to be ready to do something for the preservation of our country, our properties. "And, of course, you have to ask yourself, Against whom? "Well,” you say, "only one source, and that is communism.

This is quite an interesting jump — that the role of the church in civil defense is rooted in a deep anti-Communist sentiment. This is clearly what animated Kern, beyond the more prosaic functions that religious organizations could serve after a major disaster. Kern’s talk frames the entire issue in theological–ideological terms:

It has been my contention that we have civil defense because of an ideological friction and conflict in the world. […] The cause of all the friction and conflict and the struggles between East and West and democracy and communism rest with the idea or the concept that we have concerning the true nature of a human being. What is man? It’s a moral issue, because it is a moral judgment applied to the nature of human life and human beings. And there’s only one institution in the whole world and certainly in our country that is dedicated to this problem of the moral judgment of life, to the nature of man, to his origin, the value of his life and his destiny. And that is the church. […]

In the Soviet Union there is no God. It is strange how philosophy can even presume to be logical and rational without the acknowledgment of a Creator. The Russians believe there is nothing spiritual in the world, there is no God, no spirit; there is nothing spiritual about a man — he is material.

And so on, and so on. “Civil defense,” he concluded, “should include spiritual civil defense.“ Because even after a nuclear war, Communism would still exist, even if “every living Communist” was killed. Because Communism was an idea, and as such, “you can’t kill it with a bullet. The only way to meet it is on its own grounds, and that is basically an ideological basis, ideological counteroffensive.”

It seems likely to me that Kern’s expansive mandate was a reaction to exactly the kind of skepticism of Civil Defense that had been developing in the intervening years. Whereas “The Clergy in Civil Defense” could take for granted that Civil Defense was essentially seen as a positive activity, Kern is trying to sell it as an essentially holy war.

The pamphlet, “The Church in Civil Defense,” is not nearly as fiery as Kern’s own speeches, but one sees his beliefs in it:

The very existence of Western civilization is endangered. Our present state of culture, developed over centuries, is threatened with extinction. Not only democratic government, the highest form of political order yet evolved, which offers the greatest degree of individual and group freedom while guarding the rights of all, but religion itself, is challenged by a godless, totalitarian tyranny. The church cannot do its work effectively in a state which stifles it If civilization is to continue to make progress and if democratic government is to exist, then the church must be free to continue its mission.

Communism is a substitute religion. At its deepest level the conflict between Soviet communism and the free world is a religious conflict. We are not dealing with an inconsequential perversion of philosophy and religion. Communism tries to interest mankind in religious terms in spite of its avowed godlessness. It has its substitute god and utilizes all the facets of religious motivation The Party tries to usurp the place of the church and to give the writings of Marx and Lenin the status of sacred scriptures. The kingdom of God is promised in the classless and stateless, that is, governmentless society . […] The old Marxian charge that religion is an opiate used to oppress people can be dispelled only by an alert and positive church, a church motivated by its tenets of love of God and love of man, and therefore, a church actively interested in any program of human welfare. Civil defense is a program of human welfare devoted to saving and sustaining human life.

Other than the introductory notes, the actual recommendations are not all that different from the 1951 pamphlet. The only notable difference that jumped out to me was that each clergyman involved with Civil Defense was entitled to a helmet and special armband that would so identify them in time of disaster:

There is also a fascinating organizational chart that divides the work into three functions: the spiritual, self-protection, and mutual aid. I have to admit that having “mass burial” and “panic prevention” being put next to one another raised an eyebrow.

There are a smattering of records of Kern’s work beyond this pamphlet. He was apparently still in the role when the FCDA was consolidated into the Office of Civil and Defense Mobilization in 1958, and stayed in the role until the early 1960s. He gave talks to many religious groups and organizations, and was the one who led the opening “invocation” at official FCDA meetings. His office facilitated the donation of thousands of religious books for post-apocalyptic use, as shown in the photo below.

The official accounts of Kern’s talks to religious groups tend to be pretty positive. A PhD thesis from Matthew Justin Hall at the University of Kentucky on Southern Baptists in the Cold War suggests that the main concerns that at least that group had with Civil Defense were rooted in whether the post-nuclear state would impose too many requirements on religious organizations, weakening separation of Church and State. Kern, in both his pamphlets and talks, urged that this was not the case, that the plans were only to strengthen existing religious activities.5

Aside from the fact that “clergy” and “church” in this time was limited exclusively to Christian and Jewish faiths (with even latter’s inclusion feeling a bit tokenish), I would also just note that like all Civil Defense at this point in time, it is extremely focused on white audiences. Which is not all that interesting in the context of Eisenhower-era Civil Defense, except inasmuch as neglecting African-Americans when targeting religious communities is, well, excluding a potential audience that would possibly have been quite engaged. There is undoubtedly more that could be said on this point.

Ultimately, both of these pamphlets are interesting artifacts from Cold War America, and the shifting ways in which the efforts at Civil Defense were framed in both the Truman and Eisenhower administrations. The push towards an increasingly ideological interpretation of the question of what the role of religious institutions ought to be in preparing for, and recovering from, a nuclear attack, is likely both a reflection of the changing strategic context, as well as possibly a reaction to the growing skepticism towards Civil Defense within the United States. This axis of national security, ideology, and religion is of course quite a potent one in the United States, particularly during certain administrations — Eisenhower, Reagan, Trump — that have sought to explicitly imbue religious sentiment into their ideological pronouncements.

A strange thing: France is on Central European Time (GMT+1), not Greenwich Mean Time (GMT+0), despite being much closer to the latitude of the United Kingdom than, say, Germany. The consequence of this is that “solar noon” is considerably diverged from “local noon” here: we’re an hour or more later than the Sun. So instead of the Sun being overhead at noon, it is overhead at 1pm (13h) or even 2pm (14h).

Why is this? I found a blog post on this from a few years back; I have not tried to independent verify it. It says, in brief: Paris was on Greenwich Mean Time from 1911 until 1940, when the Germans put it on Central European Time — Berlin time — during their occupation of the country. This was not uncommon during the war, as a simplifying measure (the Potsdam Conference was on Moscow time, as it was in Soviet-occupied East Germany). Actually, it was even a little weirder than this: in the early part of the occupation, only the northern part of France was under direct control of Germany (“Occupied France”) and the southern part was autonomous under a pro-German puppet state (“Vichy France”). Only the northern part switched to CET, so the country had two time zones split by their political status. When the Germans officially occupied Vichy France in 1944, then it also switched to CET.

When the war ended, France maintained CET (year-round; it did not have summer time until the 1970s). Apparently there were some “patriots” who referred to switch back even during the war.

All of this is very odd to me. Your sworn enemy occupies your country, changes the time zone for their convenience to one that does not align well with the Sun, and after they have been evicted, you keep the “off” time? This is famously-independent, famously-standardized France we are talking about, too.

It isn’t just France that’s on the “wrong” time zone, it’s also Spain, which is tremendously “off” from solar time as it is also on CET. They also adopted CET during World War II, by choice, and have kept it ever since.

Apparently the total coordination of time zones and adoption of DST was only done in the 1980s and 1990s. I would not be surprised to learn that this was a question of European unity — the idea that many things would be facilitated if one knew that the continent (minus Portugal!) was on the same time zone. But I haven’t dug into that at all; it is just a hunch.

Now, many people here (including in my household) like the fact that Paris is on CET, because you get a much longer day, at the expense of a darker morning. I’m not necessarily objecting to that.

But I still find it strange that the French maintained German time after the war. I suppose there’s a deeper idea buried in there: some times when things change, they can never quite go back to what they were.

This week, among other things, I’ve been trying to fix up some things with MISSILEMAP. The main motivation is financial — every once in awhile it gets enough traffic that it costs me money to maintain it (thankfully paid by a grant), because it is still on Mapbox, which is very expensive, and I would like to switch it over to the Protomaps system I developed for NUKEMAP. And the excitement about missiles earlier this month drove it into territory that is going to cost my grant a little bit — which is grant money I’d rather spend elsewhere, to be honest.

But that’s not as easy as just flipping a switch, because MISSILEMAP was coded up quite some time ago. And while I am in there, “under the hood,” I see lots of other things to improve. One of them is to use a more robust Circular Error Probable function that will allow advanced users to do things like have non-circular CEPs (ignore the apparent linguistic contradiction). And I’ve been looking at implementing a time-of-flight estimator as well (which, to do well, is much harder than you might imagine). I’m hoping to have a little update with these things (and a few rough edges polished off) in the next week, with any luck.

For Doomsday Machines, I wrote up a short piece on “un-inventing the bomb”:

It’s a topic I mentioned on here a few weeks back, as it came up during my teaching, and as I was mulling it over I realized I could tie it in with A Canticle for Leibowitz, so I just wrote it up. And I do love the unreality of photos of aerogel (above). I had another blog post almost ready to go, an interview, but I am delaying it for another week or more, due to some evolving circumstances… so what I put up for next week might be a surprise to all of us. And there is a chance I won’t get anything up at all, as I will be briefly traveling to the United States for a family event, but I’m trying to get something in the hopper before I head out. We’ll see…

Read more

For post-nuclear apocalypses, the trope frequently takes the form of the survivors having only the dimmest or confused memories of what a nuclear weapon actually is. In Beneath the Planet of the Apes (1970), this manifests as a hidden, subterranean religion carried on by mutant human survivors who worship a massive Doomsday bomb emblazoned with the symbols for Alpha and Omega. In Russell Hoban’s Riddley Walker (1980), the memory persists linguistically as the “Bad Time.”

One of the works that engages with this question of forgetting in a sustained way is Walker M. Miller, Jr.’s, A Canticle for Leibowitz (1959). The novel is a compilation of three short stories (written independently but then “fixed up” into a singular story) that looks at three distinct far-future periods after a nuclear war. As with the Roman Empire, some institutions have survived the fall: notably, the Catholic Church, that enduring institution.2 In the novel, the Church once again becomes a repository for old knowledge, which, over the course of many centuries, leads to the reinvention of a technological civilization — and to yet another nuclear war.

This cyclical (and perhaps cynical) message is at some intersection of a statement about human nature and technological inevitabilities: once invented, the bomb cannot be un-invented, and once they have the bomb, humans will inevitably use it again, it is just a matter of time. It would be an interesting variant to imagine the same story but without the Church’s role; is the problem that the knowledge was preserved, or would that rediscovery have been inevitable eventually anyway?

A relevant and provocative argument from the sociology of nuclear technology can be found in Donald MacKenzie and Graham Spinardi’s “Tacit Knowledge, Weapons Design, and the Uninvention of Nuclear Weapons,” published in the American Journal of Sociology in 1995.3 The argument that MacKenzie and Spinardi make is about the nature of knowledge and forgetting, and its possible relevance to nuclear weapons.

The core of the argument revolves around an idea that was by then quite common in the sociology of science and technology known a tacit knowledge. They explain the distinction early on in the article:

Explicit knowledge is information or instructions that can be formulated in words or symbols and, therefore, can be stored, copied, and transferred by impersonal means, such as in written documents or computer files. Tacit knowledge, on the other hand, is knowledge that has not been (and perhaps cannot be) formulated explicitly and, therefore, cannot effectively be stored or transferred entirely by impersonal means. Motor skills supply a set of paradigmatic examples of tacit knowledge in everyday life. Most of us, for example, know perfectly well how to ride a bicycle yet would find it impossible to put into words how we do so. There are (to our knowledge) no textbooks of bicycle riding, and when children are taught to ride, they are not given long lists of written or verbal instructions. Instead, someone demonstrates what to do and encourages them in the inevitably slow and error-ridden process of learning for themselves.

Anything you can communicate by writing down is explicit: the Pythagorean theorem and instructions on how to use it, for example. Or a nuclear weapons blueprint. But not all knowledge is of this form. “Tacit” means wordless — it is knowledge that you feel in your hands, in your body, in your experience. Riding a bicycle, as above, is the paradigmatic example.

One could write down how to ride a bicycle, but reading that description would not make one capable of doing it. Indeed, the best text about how to ride a bicycle is essentially one that gives instruction as to how to teach oneself how to ride a bicycle:

1. Acquire a bicycle.

2. Understand that the goal will be to continually propel it forward while balancing on top of it and turning the pedals with your feet, in a forward “running” motion. Your hands will need to turn the front wheel to change direction. Observing someone else doing this first would help. You will be continually shifting your balance to maintain stability.

3. Attempt to do this over and over again until your brain figures out how to coordinate your movements and balance in a way that gives you control over the bicycle. You will fail frequently in the beginning. Wear a helmet.

This isn’t really a recipe that you can just follow, it is a program for education, and the actual acquisition of the understanding will come when some nebulous connections “click” in the brain. And like so many things involving tacit knowledge, you can’t really forget it either: if you go 10 years without riding a bicycle, you can still probably do it, you will not need to re-learn anything.

I think the above description gets at both the benefits and difficulties of this term. It does seem to describe certain kinds of knowledge well, although the barriers between it an explicit knowledge are not as iron-clad as they may at first seem. Just because something is tacit does not mean it cannot be learned or taught, it just means that the written word is insufficient for doing it by itself. A lot of what we call experience and know-how is tacit knowledge: the kinds of embodied understandings that one gets over time with close familiarity with any subject that goes well beyond what one can find in a textbook. And this is why, incidentally, “experience” matters so much more than formal knowledge in most contexts: because we recognize there is a difference between “mere book-learning” and “hands-on work.”

Ask anyone who has gotten a post-graduate higher degree and they will tell you that most of the “work” of that experience is essentially an apprenticeship where one “learns by doing.” This doesn’t negate the value of book learning — you need that, too, for a lot of things — but it emphasizes that true mastery isn’t just the book learning, it’s the tacit knowledge, too.

So what is the relevance to nuclear weapons? Nuclear weapons are complex technical artifacts that require even more complex technical artifacts to produce the components that make them work. As a whole, “nuclear weapons” aren’t even just the weapons themselves, but the entire large technical system that is required for them to exist and be sustained. Obviously many of the technical details of these systems can be, and are, written down. But MacKenzie and Spinardi argue that much of their production must also be in the form of tacit knowledge, locked up in the heads and hands of the people who build and maintain them.

The article, written in that immediate post-Cold War glow, asks the question: what if nuclear weapons are never tested again? What if, after decades of not making them, the tacit knowledge is lost? It is not fanciful, as an idea. Indeed, MacKenzie and Spinardi base their argument itself on interviews with weapons laboratory scientists at the time who voiced exactly this kind of concern, fearing that the future generations of scientists would become increasingly distant from the hard-earned experience needed to even maintain nuclear weapons capabilities, much less imagine innovate in them.

Are there any examples of this happening? The most concrete example of this “forgetting” actually happening is with a mysterious substance known as Fogbank. Fogbank was a classified component used inside certain US thermonuclear weapons. As one weapons worker put it: “The material is classified. Its composition is classified. Its use in the weapon is classified, and the process itself is classified.” It was later indicated that it is an “interstage material,” used in the transferring of energy from the fission primary to the thermonuclear secondary inside a multi-stage nuclear weapon. It is also clear that it is fairly exotic: it is not something used in many other contexts, or produced regularly, and has apparently rather unusual properties.

Whatever it is, the point of bringing up Fogbank is because the US government “forgot” how to produce it. As a Los Alamos publication put it:4

Produced by skilled hands during the 1980s, Fogbank is an essential material in the W76 warhead. During the mid-1990s, Fogbank production ceased and the manufacturing facility was dismantled. As time passed, the precise techniques used to manufacture Fogbank were forgotten.

When it came time to refurbish the W76, Fogbank had to be remanufactured or replaced. In 2000, [the National Nuclear Security Agency] decided to reestablish the manufacture of Fogbank. Officials chose to manufacture Fogbank instead of replacing it with an alternate material because Fogbank had been successfully manufactured and historical records of the production process were available. Moreover, Los Alamos computer simulations at that time were not sophisticated enough to determine conclusively that an alternate material would function as effectively as Fogbank.

Although Fogbank is a difficult material to manufacture, scientists soon discovered that restoring the manufacturing capability would prove an even greater challenge. Scientists faced two major challenges:

• most personnel involved with the original production process were no longer available, and

• a new facility had to be constructed, one that met modern health and safety requirements.

Despite efforts to ensure the new facility was equivalent to the original one, the resultant equipment and processing methods failed to produce equivalent Fogbank. The final product simply did not meet quality requirements.

The full article explains (in ever-elliptical descriptions) that the scientists went back into the archives to look at the historical data on Fogbank’s production, discovering that there were indications of similar problems in the past. They found that maintaining a specific level of “impurity” was partially responsible for the quality of the final product, but that the historical records gave no indication about how that was accomplished.

They ultimately discovered that it was actually their newer processes that were causing the issue: they were doing too good a job of removing impurities that would otherwise have been in the original material. The later scientists ultimately discovered that the earlier scientists who knew how to make Fogbank were not themselves totally aware of what it was that was contributing to its success: “The historical Fogbank production process was unknowingly based on this essential chemical being present in the feed material.” Which is an interesting wrinkle on the question of forgetting: the original “knowers” likely have incomplete knowledge themselves!

So can one un-invent the bomb? MacKenzie and Spinardi say, perhaps: one could deliberately un-invent the bomb, by eliminating them and destroying the records of how to make them, but one could also accidentally un-invent the bomb, by simply not making more of them and maintaining that store of tacit knowledge to adequate levels.

The authors try not to overstate the case, though. They acknowledge that there more-or-less simpler ways to make nuclear weapons than the most advanced varieties that make up the bulk of the warheads currently in the world, that there is considerable overlap in some aspects of weapons knowledge with other fields of knowledge, and would certainly not argue that even an “un-invented” technology could not be re-invented. We may not know exactly what the ancient recipe for Greek Fire was, but that has not stopped people from developing other kinds of incendiary weapons.

Still, the argument an interesting and provocative one, even as it raises some obvious objections. There are ways to reduce dependence on tacit knowledge, for instance: that is exactly what many instruments and machines are invented to do, it is the nature of automation.

As a very simple example, a 3D printer is a machine that reduces the tacit knowledge required to manufacture certain kinds of small components. Instead of learning how to sculpt a small figure out of plastic, I can instead take a data file (explicit knowledge) and command the printer to make me the figure. Even this will have some tacit component — I gather than using 3D printers well requires some hard-earned experience — but now we’ve shifted the tacit component from one domain (sculpting plastic) to another (maintaining the machine), and probably reduced it (becoming a competent sculptor takes years of specialized experience; learning to keep a printer functioning probably does not).

Are we locked into these technologies forever? Even after, as A Canticle for Leibowitz suggests, the technologies may have destroyed the civilization that produces them? It is a grim thought. One hopes not. The idea is what my colleague Benoît Pelopidas calls nuclear eternity, and is implicitly built into many discussions about nuclear weapons, even by those who argue against them, without being interrogated.5

One can imagine ways out of the bind. Despite widespread believe that technology has its own driving momentum, technologies do not exist for their own sake. Their creation and maintenance requires an immense about of work and effort. If you have doubt about that, you don’t have to trust the historians or sociologists, just look at the budget requests for “modernization” of American nuclear weapons: if the US does pledge billions, perhaps trillions, for ongoing maintenance and improvements, then its arsenal will wither away to nothingness. While presented as a problem to be (expensively) solved, it is actually a rather encouraging idea, restoring a much-needed sense of human agency to these conversations. As abstract as they seem, nuclear weapons are still just machines.

A satirical, probably vibe-coded app was circulating, and my main issue with it is that the technical aspects are so dramatically wrong that it actually ends up supporting the opposite conclusion than its author desires. That is, it makes digging nuclear canals look too easy.

As I have mentioned on here before (and written up at Restricted Data in more detail), I am in the process of totally overhauling the NUKEMAP code, including the effects code, and so do enjoy the opportunities to look into such things again with fresh eyes. The current NUKEMAP “crater” function is pretty limited, and so I took this nuclear-canal discourse to update my codes (not yet on the NUKEMAP website) for cratering.

I posted a thread about it on Bluesky, which I will just link to above rather than reproduce here. The long and short of it is that nukes are very good at destroying cities but actually a lot less impressive than you might think at digging holes. Even the full-sized, 100 megaton Tsar Bomba, detonated at the optimal depth for digging a hole, can “only” dig a hole the width of Manhattan island. Which, for a weapon that can set the entire New York City metro area on fire, is pretty unimpressive.

The long and short of it is that in order to dig a canal near the Strait of Hormuz, even ignoring the fact that the area is inhabited and has a mountain range you’d need to go through, you’d need hundreds of nuclear detonations. Which is just to say, again, that separate from the political difficulties and issues with nuclear fallout, it is just not a practical idea — it is just not on the table that the US would be able to, on any short-term timescale, set up such a thing.

But I didn’t mind digging up the equations and graphs for it. I find that kind of thing very enjoyable. It combines a few different faculties that I like: I feel like I’m learning and understanding more, I enjoy transferring things from the difficult-to-use world of printed graphs and equations into the much-easier-to-use world of my new code library (which is set up to just be dramatically easy to use once you get everything into it correctly), and I like feeling like the NUKEMAP upgrade project is making progress (which it is).

I’ve been doing this with other aspects of the effects code, bit by bit, over the last few weeks. The blast overpressure code now goes down to much lower levels of pressure (0.25 psi minimum, instead of 1 psi) which, aside from just giving a bit more sense of the areas affected, also adds some more fine-grained fidelity for how casualties are be calculated. The model for initial (acute) nuclear radiation allows for much more tweaking by weapon type, as different weapons have slightly different gamma versus neutron outputs, and those can be summed up independently. The cloud size model will have many more bells and whistles to it, and be somewhat more accurate to the original data.

In the older NUKEMAP code, I tended to prefer finding polynomials for the various effects functions so that I could extend them a bit beyond the ranges of the original graphs, for example, and in the new version I tend to use the graph data verbatim for the range of yields they cover, and to only use small exponential functions, based on the data at the extreme ends of the data curves, for going beyond the “known” data. It’s a strategy that turns up to be easier to implement, as well, because finding a complex equation that fits the entire curve is much harder than just getting a simple equation for the ends of it.

I don’t know if anyone else reading this does this kind of thing, but if you do, out of all of the software options out there for tracing graph data, PlotDigitizer has become my favorite. It’s very easy to set the scales, and the “zoomed in” view that it allows makes it very easy to fine-tune the plot. It shows you the data as you do it, and you can just copy and paste it from there. There is a free and paid version; I have just used the free version so far, and it works great, but I can see some possible useful features in the paid version… who knows. I used to do my pixel digitization by just manually measuring and recording with Photoshop and Excel and, well, that’s obviously the least easy way to do it.

Using the PlotDigitizer, I can digitize a graph within a few minutes in a way that I find very reliable (I do not trust automatic digitizers, and please do not recommend just feeding them into an AI bot…), plug the data into Excel, come up with the extrapolation equations using Excel’s trend line settings (its “power” equation type usually works best for nuke effects; if I really need to do complicated curve fitting I use FindCurves.com, which will just throw every equation form in the book at data looking for a good fit), and then I have some pretty straightforward means of moving the data into my code framework. I do a quite a bit of testing with any data I import, because it is easy for errors or weirdness to creep in, and just so I make sure that I know that it is doing what it is supposed to do. It is very satisfying.

In case you missed it — which you very well may have, given that I posted it on Saturday rather than Thursday or Friday (I ran out of time) — for Doomsday Machines this week I wrote about a healthy sampling of the photographs that exist of the Hiroshima mushroom and smoke clouds:

I was inspired to do this by a photo of the Nagasaki mushroom cloud that I’ve seen circulating periodically, and some of the confusion that is caused by the difficulty of making sense of its scale. I had thought about doing a single post about both Hiroshima and Nagasaki, but as I began writing about Hiroshima I realized that it was, by itself, quite enough of a subject, so I will defer the treatment of Nagasaki for a future time! I had never quite tried to make a survey of what photographs exist of the Hiroshima clouds from below — nor tried to really make complete sense of the cloud structures visible in them (particularly trying to differentiate if they were the mushroom cloud, the mushroom stem, or the pyrocumulus smoke cloud caused by the firestorm itself, so that was a useful exercise.

For Doomsday Machines next week, I will post an interview that I did last week with my friend Benjamin Wilson, a historian of science whose new book, Strange Stability: How Cold War Scientists Set Out to Control the Arms Race and Ended Up Serving the Military-Industrial Complex (Harvard University Press, 2025), has been causing waves and controversy in the arms control community. We had a really great conversation about the history of arms control and its trickier, more hidden aspects. Don’t miss it!

Read more

The most familiar of these photos was taken by S/Sgt. George Robert (“Bob”) Caron, the tail gunner on the Enola Gay, the B-29 which dropped the Little Boy atomic bomb over Hiroshima:

This photograph is surely cropped in the darkroom from a larger negative; it has the gauzy appearance of too little resolution being blown up to too large a size. It is, from a technical perspective, not that great a photograph — the lighting is poor, the focus is poor. This is not an artistic critique, as it was essentially an amateur snapshot taken by a soldier in a combat zone. It does highlight the separation between the “head” of the mushroom cloud and the “stem” of it, a consequence of the weapon’s height of burst.

There is another photograph by Caron that gives us a sense of how much the previous one was cropped. This one is far less commonly used, but I prefer it greatly:

I feel that whatever the cropped version gains in terms of getting a better view of the cloud, it loses its sense of scale and immediacy. The wider shot allows one to better imagine what it would look like to be seeing this in real life, from the air, and gives a better sense of scale — especially since most of us today are very familiar with the view from outside of an airplane.

Is the first photograph simply the cropped version of the second? I think not. If you superimpose them, as I have done below, you can see that there are some key differences:

Even barring my own alignment issues, the cropped photograph (581589 A.C.) seems to have been taken a bit earlier than the second one, as the “head” and “stem” separation is a bit different, the “head” is reaching a bit more upward in the second photograph, and, importantly, the entire cloud is rotated slightly along the axis, presumably because of the movement of the plane. It is difficult to tell but one suspect that the cropped photograph was used (and cropped) because it had slightly more definition in the final cloud than the second one.

There is another photo of Hiroshima after the bombing that is often used to illustrate its cloud:

It is an impressive sight. Taken at a much greater distance, the cloud here is much higher than the previous ones, and casting quite a shadow. But this does not actually show the mushroom cloud from the atomic bomb at all; this was taken several hours later, well after the initial mushroom cloud had stabilized and been blown away in the wind. Rather, this is a pyrocumulus smoke cloud caused by the firestorm in the city. For the sake of simplicity I will call this the “smoke cloud,” to differentiate it from the “mushroom cloud” created by the immediate detonation of the atomic bomb.

The photo above was taken by a reconnaissance plane several hours after the attack. From the perspective of the Army Air Forces, these were not successful attempts, because the smoke from the fires obscured their attempts to photograph the ground for at least another day. Here is another photograph, from much closer, of a later reconnaissance attempt, where the cloud is not really meant to be the subject, but is obscuring the subject — the city:

It would not be until August 8, 1945, that they would succeed in getting a completely “clean” (smoke-free) photograph of Hiroshima.

There is one more photograph that I have seen listed as a view from above, supposedly from the collection of a member of the 9th Photographic Technical Squadron, and only “rediscovered” fairly recently:

The above photograph also shows a pyrocumulus smoke cloud. Is it of the Hiroshima smoke cloud, or some other attack? I can’t really tell visually; the landscape looks like a lot of Japan does from the air at that time, and I’m not versed enough in aerial reconnaissance to really pass judgment. It seems plausible, anyway, but given as the owners of the photograph were listing it as the centerpiece of an eBay auction with a ridiculous price tag, I am wary of taking self-interested claims at face value without some confirmation. It does resemble the other smoke cloud photograph geometry, but perhaps many firestorm clouds look somewhat similar. (I have no idea what price was ultimately achieved on the auction, if any.)

Lastly, there is also 16 mm film footage of the Hiroshima cloud, shot from the instrument aircraft, The Great Artiste, by the physicist Harold Agnew. A copy has been digitized by the Hoover Institution Library and Archives, which houses the original:

The technical quality is, again, understandably poor, and the movement of the plane dominates it. The two sequences show different times of the clouds evolution. Both look to be somewhat after Caron’s photographs, as the cloud head in the first one is quite broader and looks more stabilized. In the second one, it appears that the cloud head is starting to break apart in the winds. The separation between the head and the stem is quite visible in the first one, as is the fact that there is some significant horizontal displacement between the two; this is present in Caron’s photographs but less obvious.

These are, to my knowledge, the main photographs of the cloud itself from “above”: the view of the bomber, or the reconnaissance plane. It is a view of destruction from several miles up, quite different than the view from the ground. There are a few photographs from the latter perspective, which I will write about at a later time. For now, I just want to highlight a few other photographs of the cloud and smoke itself: these ones taken by the Japanese. The Hiroshima Peace Museum said in 2013 that it knew of 21 confirmed photographs of the mushroom cloud from the ground. I have only seen half a dozen or so, shown below.

The first set are a sequence of three photographs taken by the photographer Mitsuo Matsushige, who was at home some 7 km / 4.3 miles north of ground zero. He appears to have rushed outside and photographed the cloud within a minute or two of the detonation:

The above photograph is really perhaps the most amazing photograph of the cloud that I have seen. The immensity is obvious. The houses give it an immediacy and some sense of scale. Anyone who has tried to photograph large clouds or buildings knows that it is very difficult to really capture the lived sense of scale, so if this is what the photo looks like, the reality must have been even more awesome. We are, I think, looking up at the mushroom stem, with the separated mushroom head much further above and harder to make out.

At this distance from ground zero, Matsushige was far-enough away that blast, acute radiation, or thermal radiation were not a threat. He may have been in an area that would be exposed to “black rain” — precipitation caused by the aforementioned smoke clouds — which contained some amount of radioactivity. The question about whether the black rain from Hiroshima measurably increased cancer rates is a controversial and tricky one.

Matsushige has another photographs of the cloud from the ground, presumably taken shortly after the first one:

Again, it is a striking photograph, and again, I think this is only the “stem.” The “poor” quality of the negative — my photography teacher in high school would have rejected this for its spots and problems — adds to the reality-effect in this case, given the context.

Mitsushige also took one more photograph, at what looks like a later time, and from a different vantage point:

This photograph was apparently taken around noon, several hours after the attack, and as such is a complement to the aerial photographs of the smoke fire.

There are a few other such photos from the ground, such as this one taken from Kure Naval Arsenal, some 20 km / 13 miles to the southeast of Hiroshima:

This looks either like a quite late mushroom cloud, or an early stage of the smoke cloud. I suspect the latter, just based on its particular size and geometry, but it is hard to tell. Whatever it is, it is not very mushroom-like in its appearance at this point; mushroom clouds do look more like “columns” after they stabilize, so it is not impossible, although I suspect it is the smoke cloud. The Library of Congress lists it as an “unusual photo of the Hiroshima bomb explosion,” which is an odd choice of words. Do not all of the photographs qualify, in some deep way, as “unusual”?

There is also this photograph that is listed on Wikipedia as being “found in Honkawa Elementary School in 2013 of the Hiroshima atom bomb cloud, believed to have been taken about 30 minutes after detonation of about 10km (6 miles) east of the hypocentre”:

That is definitely the Hiroshima mushroom cloud, with a clear separation of the “head’ and “stem.” I do not think this is 30 minutes after the attack; this looks to be only a few minutes after the bombing, on par with the cloud shown in the Agnew film. My guess is that it is still before the mushroom head had reached its stabilized altitude (in which is temperature and density matched the air around it, which stops its ascent), so that would put it at 10 minutes or less after detonation. Apparently the Hiroshima Peace Museum put the time at 20-30 minutes, but even 20 minutes feels a bit late to me. The photographer is unknown.

The last one I will put here is from a very remarkable perspective:

It is a remarkable photograph — a truly literal “view from below.” The above image comes from the Chugoku Shimbum newspaper’s Hiroshima Peace Media Center, who say (per Google Translate) that it was taken by Seizo Yamada, an employee of the Chugoku Shimbun, and that he was some 6.5 km / 4 mi from ground zero at the time, visiting a friend. Yamada described the fireball as “dark red” when he took the picture. The Chugoku Shimbun says that if he had been at work as usual that day, he would have been more definitively under the cloud. The perspective gained by being just a bit further out than Matsushige’s photographs is striking: close-enough to look like it is right on top of the photographer.

The scale of the Hiroshima mushroom cloud, and the smoke cloud, is immense, and hard to fathom. In fact, it is hard to know the exact height: while later equations were developed for generalizing the stabilized heights of mushroom clouds as a function of weapon yield, the “raw data” of these, derived from nuclear testing, shows considerable messiness. But from a human perspective, they are huge — many multiples of our tallest buildings. The next time you are at the base of a building like One World Trade Center in New York City, consider that the Hiroshima cloud was well over 10 times taller than it.

In a future post, I will talk about the similar cloud photography from Nagasaki — which has many interesting differences in terms of appearance and documentation — as well as other kinds of “on the ground” accounts of the atomic bombings.

My pal Cheryl Rofer has a good post up at Lawyers, Guns, and Money this week which hits at something I have felt, and been saying privately, especially this last week:

This week, I’ve talked to or messaged with a number of reporters. Rob has been doing a crazy number of interviews. That adds to the personal distractions. The wearing-down means I no longer have patience for certain types of foolishness, including the media’s willingness to consider Trump a rational actor. Others are giving up that pretense as well.

I have also been queried by a number of reporters this week (sometimes I get back to them, sometimes I don’t, I am trying to take the attitude of “it is not on me to be the expert on every single aspect of this world,” and I do not claim to be any kind of Iran expert), and I’ve felt a similar weariness.

There is a journalistic assumption that somehow we should be parsing Trump & Co.’s actions and phrases as if they were made with some intent at conveying meaning, truth, or rationality to them. I am convinced, for example, that Trump does not know what “unconditional surrender” means as anything other than a clichéd war-related phrase, and that Hegseth does not know what “give no quarter” actually literally means (much less its legality under American or international law).

I do not think there is any reason whatsoever to trust the Trump & Co. are being truthful in any statement. I do not think there is any reason to think that they have some kind of real plan. I do not think there is any reason to assume they have earned any benefit of the doubt, or that they have either the interests of the United States or the greater world in mind.

My sense is that the American journalists I know are all too aware of all of this, yet feel obligated to go through the motions of pretending. It’s has always been tiresome but it also increasingly feels dangerous and deluded. The media continues to “sane-wash,” to assume good faith and rationality, to phrase bald-faced statements of racism and idiocy as if they are perhaps “controversial.” Can we be done with it, already? When will the time come that we can just say, plainly, that the Emperor has no clothes, and not feel like we are saying something controversial?

Cheryl made another point that was on my mind:

As the stories came out about the White House deliberations, though, the military (presumably) did feel stung enough to leak that it wasn’t them who ignored this Gulf of Hormuz thing. They specifically pointed to the President, but as long as they are carrying on this improv war, they’re culpable too.

They are culpable, as is everyone keeping this show on the road, whether they agree with every aspect of it or not, whether they believe they are holding back some bulwark of even worse ideas. (I will admit that even paying taxes right now feels like collaboration, to a degree. I guess I’m feeling very Henry Thoreau about the whole thing at the moment.)

The whole thing makes me furious. Not “angry,” in the sense of a quick emotional sensation, the kind that everyone is trying to monetize one way or the other (including the endless text fundraising spam from Democratic candidates — like, are you kidding me? Are you for real right now?). I mean really furious. It is all so morally and ethically and civically offensive. On a level that mocks any safe and empty phrases about the desire to “resist,” much less the pathetic letters of disapproval registered by “concerned” politicians.

One of the things I write about in The Most Awful Responsibility is how the Korean War was the real “Cold War turning point” for Truman, the event that really and truly caused him to be a “Cold Warrior.” What I mean by this is that Truman’s view of the world situation, and the Soviet Union, prior to the summer of 1950 was about the tenseness of it, but he essentially thought of it as some kind of misunderstanding. The Soviets he seemed to regard as an essentially petulant teenager of a world power, testing boundaries and acting out, paranoid and problematic, but not really an existential risk to world peace and stability.

Even well after most of his cabinet and advisors had moved to be more anti-Soviet, Truman seemed to think that it was a salvageable situation — that eventually the Soviets would see that the United States was not really going to be all that aggressive towards them or their interests, so long as the Soviets didn’t push things too far. In one of the more bizarre moments, in the spring of 1949, Truman told people (including Oppenheimer) that he believed that by 1951, the world would probably agree to nuclear disarmament — which Oppenheimer plainly thought was nuts. As it happened, the Soviets detonated their first atomic bomb a few months later.

But the invasion of South Korea by the North, which Truman (and those around him) credited to the Soviet Union (rightly or wrongly; we now know it’s a bit more complicated than that), really made him furious. It offended Truman’s sense of moral and civic virtue in a truly deep sense. Why? Because Truman was, in fact, quite personally committed to the idea that the proper response to the horrors of World War II was to create some kind of lasting peace. And here the Soviets were, not 5 years later, starting a new war of choice. One gets this sense that this is really just too much for him. Sure, play games in Berlin and Greece — none of that is great, but it’s a bit of jostling inside the constraints of war. But to embrace war itself, just to do it, just to see if you can get away with it — this disgusted him.

That’s the level of disgust and fury I’m at right now. It’s not that the many other horrific, immoral, disgusting policies pursued by Trump & Co. didn’t deeply offend me. They all did. But the Iran war situation is one that has been war-gamed and discussed and — quite rightly — avoided by multiple presidents, of all parties, for a long time now. Because all of the scenarios about its outcome have consistently suggested that it would be very risky, that the costs would not be worth the benefits, and that there were going to be better options on the table. And yet, here we are, and Trump & Co. didn’t even feel the need to try and actually “sell” it to the American people or American allies.

We’re all going to get stuck holding the bag on this one. Even if one imagines that Trump & Co. will someday get their comeuppance — one can dream — it won’t actually remove the costs that they have imposed on everyone else. And there will be high costs, of different forms. We’re well past pretending otherwise.

I don’t usually write that much about politics on here, both because it is not my métier and because I really don’t want to spend all my time thinking about them — I find thinking about nuclear weapons and nuclear war far less depressing, whatever that says about me. I figure you have plenty of sources of that in your life right now, if that’s what you want to read about. But one can’t escape it.

Well. In other news, I was asked by some reporters whether NUKEMAP had seen increased usage over the last week or so, and indeed it did. Here is just a quick snapshot of my “unique visitors” according to Cloudflare, which is just one (easy to access) way to see the change over time, even if I don’t really trust the exact numbers (web analytics are an impressively large sham, and only are useful in relative terms):

This is still quite low compared to the Russian invasion of Ukraine in 2022, which is probably a good thing for the world, and my servers and their bills. I’m happy to say that the latter are at least much more robustly done since 2022, however.

For Doomsday Machines this week, I wrote about “PLAN A,” a visualization/simulation that I worked on some years ago with researchers at Princeton University:

I will admit that this was a bit of a punt — I had a busy week and did not feel I had the time to do something more substantive, and talking about my own software is very easy for me to do, even if (as in this case) I had to “reconstruct” how things worked because I didn’t really take good notes on it at the time. I have something more substantive planned for next week.

Read more

Building up a plausible “scenario” for such a war is hard. To do that, you need a sense of targets, of attack capabilities, and targeting philosophies — as well as some sense as how these might be adjusted as things unfold. These matters are, understandably, highly classified. We have some sense of what all of these things looked like in the past, but even then there are some pretty murky areas and strong differences of opinion. And given how an actual full-scale nuclear war has never taken place — which is a very good thing! — we have a dearth of data on how these things might actually happen in the real-world.

But that doesn’t mean we can’t think about it — or that we shouldn’t. Thinking about these kinds of “scenarios” is important if we are to have a grounded, tangible, and concrete understanding of what these weapons are and what they imply about the world. And, as I think is pretty clear by now, I am one of those strange individuals who actually finds thinking about the “end of the world” quite interesting.

“PLAN A” was a project that I worked on with three researchers at the Princeton Science and Global Security Program — Alex Glaser, Moritz Bastian Kütt, and Tamara Patton — in 2017. I have only vague memories of the earliest conversations and meetings we had about it (my e-mails from the time about this start in media res, and so that suggests that we first had conversations and meetings about it prior to writing anything down about it), but my basic recollection is that there was going to be an art exhibition at Princeton about nuclear weapons, and Alex Glaser, a physicist who I have known for many years now (and written about a bit) had reached out to me about an idea they had for visualizing a nuclear exchange as part of the exhibition.

Here is the final video output, which is useful to watch before I talk about how it works “under the hood”:

At the time, I was doing a lot of programming and teaching with D3.js, a Javascript library for data visualization that has a very good support for geographic data.1 So what I believe I suggested (again, my memory is hazy on the details nearly 10 years later) is that it would be relatively straightforward to build a quick-and-dirty browser-based Javascript application that would allow one to “plug in” some data files about your nuclear attack “scenario,” and then it would allow you to visualize what it would look like. Then they could make screen captures of the Javascript application and integrate those into the final piece of video art.

I also ran the final list of “detonations” through NUKEMAP’s casualty calculator to come up with some rough numbers as to how deadly the attack would be, at least as a result of the initial effects (NUKEMAP’s casualty calculator does not, at present time, estimate possible fallout casualties).

A few years later, well after the exhibit had ended, Alex G. uploaded the output to YouTube, where it has enjoyed an amount of exposure (several million views) that nobody really expected. Because Alex G. generously (?) listed me first on the Princeton website about the video, I get e-mails periodically from people — journalists, scholars, random people — wanting to know details about the simulation. I generally send them to Alex G., because that wasn’t really my part in the whole thing.

My understanding is that Alex, Moritz, and Tamara, along with some input from the late Bruce Blair and the thankfully-still-with-us Sharon Weiner and Zia Mian, came up with the “scenario,” which unfolds over several “phases.” First the had to define a list of “locations,” each of which corresponded to icons on the map, and could be the origins or destinations for attacks. Here is an example of what the “locations” file looked like:

I (very unhelpfully to future me) did not document what these fields meant, but I can reconstruct most of them. “State” was either “NATO” or “Russia,” and was mostly for internal record-keeping (I don’t think it changes anything about the output). “Type” specified categories that would correspond with how the data appeared on the visualization (there was a separate file that defined what each of these appeared like, if they appeared at all — one kind of “type” was basically an invisible waypoint that planes could fly to).

Each “location” clearly had a unique ID (e.g. “saratov_1”). The latitude and longitude were its initial location, obviously. “no_of_dv” defined the number of delivery vehicles (which could be missiles or planes), “wh_per_dv” was how many nuclear warheads were contained on each delivery vehicle, and “subtotals” looks like it is just the total warheads per location if it had warheads (which I think was just for our own record-keeping; it is not used in the code). “yield_per_wh” is the explosive yield per warhead and does not seem to have been used (that would only have been used for the casualty calculation and I don’t think we used these files for that). I think “mirv_target_group” was something used for planning purposes because it doesn’t show up in the visualizer code anywhere.

Then we have the “phases,” which are the attacks themselves. These were scripted also using data files. Here’s a selection from phase 2:

These described how and when different attacks would be taking place. “attackid” is a unique identifier, and “type” is the kind of thing that was launching the attack (e.g., an ICBM silo, a submarine, a mobile ICBM launcher, a tactical fighter, a tactical missile). The “launch_id” is the location (from the previous “locations” file) sending out this attack.

The “target_id” is a list of locations that are being aimed at by the weapon. In some cases there would be a single location, but in some, as above, you can tell they are MIRVed, and a single missile is meant to hit several different targets. “trigger_at” was a timestamp (in the above, they are set to begin immediately) at which this part of the attack would take place.

These lists were all made by hand, which is definitely not the easiest way to do it, and I’m sure it was very time-consuming! I am glad I did not have to do this part — the programming was much more fun than the tedious compilation of lists.

The code could read these data files and then “step” through the time in regular intervals, triggering, say, a missile launch, and then locating (and visualizing) where that missile ought to be at any given time after the attack. We used some very simple assumptions for the speed of everything, reasoning (if I recall) that the fine-grained differences a) would not likely be visible at this scale of the map anyway, and b) didn’t matter for the evocative purposes of this visualization (which was not meant to be any kind of “perfect” simulation). For MIRVs, the code treated all of them as the same missile until about 50% through the trajectory, in which it then split them into multiple warheads. This was, again, somewhat arbitrary.

Here is what the “visualizer” looked like in its most basic state:

As you can see, it is a little different than the final product. It allows one to load any given “phase” datafile, and rearrange the visual display of the map, and then “start” the attack. The default time factor for development, as you can see, is 100X realtime, because realtime is slow.

How slow? Below a little screen capture I made showing the first minute of phase 2 at 1X speed.

I actually sort of love it in realtime, because while it is slow, it is not imperceptibly slow. It is disturbing to imagine that those are actual missiles, and that their slow crawl would mean that they would find targets in 20-30 minutes or so. There’s a Weeping Angels quality to it, where it barely seems to be moving at all, but if you look away and look back, you perceive the jump.

The visuals of “PLAN A” were quite deliberately and obviously inspired by the “Big Board” shown in the 1983 film WarGames:

Alex G. is fairly blatant about his nostalgia for 1980s digital culture (he snuck a Space Invader-inspired sprite into a paper in Nature, which takes some chutzpah), and I’m not opposed to it either.

I also appreciate that by doing that kind of retro visual riff, you are signaling to the savvy viewer that this is not pretending to be some kind of perfect modern simulation. I am deeply suspicion of anything that really pretends to be a perfect simulation, and with NUKEMAP and MISSILEMAP have gone to some lengths to make sure that it is visually clear that they are meant to be evocative and back-of-the-envelope, not perfect simulations.

I don’t recall it coming up in active discussions, but the game DEFCON (2006) probably informed some of my aesthetic thinking on it, too, particularly with the planes, submarines, and other little icons, which are not part of the original WarGames maps (they just have little white dots for targets). DEFCON, of course, also took its aesthetic cues from WarGames.

Another thing that I am sure we had in mind for the overall inspiration was Isao Hashimoto’s 2003 video art project of nuclear explosions, titled “1945-1998.” I am sure that at the time I said something like, “this is the kind of art that would have been pretty hard to make still in 2003 but could be basically programmed in a day a decade later,” because I am prone to say things like that.

I like Hashimoto’s piece a lot, and often show it to students in class. The almost unbearable slowness of its early periods is offset but the sudden, musical cacophony of the years of heavy testing. When things like the test moratoriums occur, their sudden silence is also quite profound, and leads to interesting conversations.

“PLAN A” is, thankfully, a work of fiction. It is a work of creativity and imagination for all involved — no less so for the programmer/designer (me) and the people coming up with the target lists and sequences (the Princeton crew). That does not mean it is not plausible or grounded. Plenty of fiction can be plausible and grounded, and plenty of even implausible or ungrounded fiction can and does reflect some kind of fundamental truths about reality. All simulations are fictions, really, and that does not mean they cannot be useful. Is it “accurate”? I don’t really know what exactly that means in this context (or how we would know).

Again, the terms I personally prefer for talking about it are evocative and plausible. At its best, it is meant to give a view of modern nuclear war that is more grounded than “everybody would die” (lots of people would die, but not everybody, for both better and worse), and is probably within some order of magnitude estimate as to what the immediate effects would be of a war of this sort, and is a not-at-all-impossible scenario that could be imagined occurring (although it is not the only possible scenario that one could imagine occurring). Evocative and plausible are really is about as ambitious as I personally would be for its predictive power, but I’m also not a scientist of any sort (“just” a historian), and so not under any obligation to aspire to much more than that.

“PLAN A” has definitely been a lot more successful, in terms of audience exposure, than anyone involved expected it to be: it was designed to be a very localized kind of art, part of a specific in-person exhibit. I think the exhibit “toured” a bit, but any art exhibit’s audience is necessarily pretty small. Its second-life on YouTube increased its reach exponentially. To what end, I have really no idea. Does watching “PLAN A” alter or affect the viewer’s understanding of nuclear war? If so, how? In what directions? Would it impact how they thought about different kinds of nuclear policies? (Does asking such a thing signal that I am thinking too functionally about art? Is it art, or some kind of design, or what? Does it matter?)

I don’t have the slightest clue about such things, to be honest — and I’m not the target audience — but I like asking the questions, and I like to believe that things like “PLAN A” might cause others to ask questions, as well.

Aside from the content of The Fog of War, which is probing and provocative, I also delight in talking to students about The Interrotron, the unique camera setup that Morris used for the latter film to impressive effect at capturing his conversations with Robert McNamara in a shockingly intimate fashion. The Interrotron is a system of two cameras and two teleprompters looped together so that the interviewer and interviewee can make constant eye-contact. It leads to a very different experience for the interviewee, as well as capturing that sense of eye-contact in the final filmed product. I have always thought it is was a painfully, agonizingly clever and simple solution to a major deficit in most interviews.

All of which is to say that being reviewed by Morris is quite a thing for me by itself, as he is an artist I have long admired, and whose work has been formative in my own thinking about the possibilities of documentaries, and epistemologically-informed history. (Morris also was a graduate student of Thomas Kuhn’s, many years ago, and they had a spectacular falling out, which Morris has written about, so there is a history/philosophy of science connection there, as well.)

Morris’ review of the book is, as one colleague of mine put it, “as good as a review can be.” My colleague continued: “Morris gets you, Alex.” I find it difficult to read reviews of my own work (even positive ones), much less send them to others, but I do feel seen in this review, as Morris very much understands the kinds of epistemological sensibilities I have regarding the writing of a history like this, and believes (rightly or wrongly!) that I accomplished my goal with it. Says Morris: “It’s a truly great book, perhaps a masterpiece, that gets us as close as we can to understanding what went through Harry S. Truman’s head when he presided over the bombing of Hiroshima and Nagasaki.” Ah, who could ask for more in a reader!?

You can read his review here, if you are inclined, and if the paywall is getting in your way (and you are still interested), I will post a PDF copy below, just for you.

Last week I was in London again, this time for a workshop at the Science Museum that was dedicated to the questions of preserving JET, the Joint European Torus, a fusion research facility located in the United Kingdom that has served out its life and is in the process of being decommissioned. It was an interesting discussion.

My main contribution was a) giving a talk about the classification aspects of “peaceful” nuclear fusion over the years (both magnetic, like JET, and inertial confinement), much of which was derived from my first book, and b) an attempt, at the end, to encourage the “there’s no money for anything anymore” Brits that they should try to think a bit bigger than just “can we saw off a piece of the machine and put it in a museum and put the data in an archive somewhere,” all of which is good and fine but really seems like the bare minimum. I know there’s no money for anything, etc., but one might as well try to start by thinking bigger. But these are the moments in which I feel the most American in European contexts, when I don’t start with the assumption that there can’t be any money found for anything…

It was just a day-trip this time (no coffin hotel), and because of Eurostar shenanigans I had to get there a few hours earlier than I had intended, so I went for a walk in Hyde Park and Kensington Gardens (which are both near the Science Museum), which was very nice.

A sign there told me that in the daytime, one could frequently see tawny or small owls in trees near their favorite holes, and that sounded quite fun to me, so I kept an eye out. Every time I saw a bird in a tree, though, it always turned out to be a pigeon or, improbably, a duck. I felt a little like the sign was, as the Brits say, “taking the piss…”

The weather has been lovely in Paris — basically springtime. I really did not expect the winter to be so short and relatively easy here. I guess I expected that because we are so far north — Paris has the same latitude as Newfoundland — that it would be more like Boston than New York. But it has been easier than even New York, I think. It is 60ºF / 16ºC today, and sunny. There’s lots to complain about the world right now, but the weather here isn’t one of them…

For Doomsday Machines this week, I returned to a document that I wrote about for Restricted Data some years back: the earliest formal American study I have seen, from August-September 1945, of what a nuclear attack on the Soviet Union might look like. It’s not a “war plan,” per se, but it gives insights into what the military thinking behind an atomic “war plan” of that period would look like:

I hemmed and hawed a bit about whether I should “re-use” a document or not, and decided that given the different scope of Doomsday Machines from Restricted Data, I could probably make an argument for it, and that I was looking at it again with “fresh” eyes, ones that have now looked at many more “war plans” (or at least, “war planning documents”) than I had at the time. I also have much better microfilm scans than I did at the time, so that’s something. For many readers, I suspect the truly terrifying and depressing thing about this document is not its contents, but the fact that the last time I wrote about it was 2012, which was 14 years ago… yikes!!!

Read more

The “problem” tackled by the Top Secret–Limited study on “Atomic Bomb Production” was simple: “To determine the United States requirements for atomic bomb stocks in the interim post-war era.” The phrase “interim post-war era” is an interesting one — what, at that moment, are they in between?2

The study makes four assumptions for tackling the “problem”:

a. The United States must be prepared to conduct offensive operations against any other world power or combination of powers.

b. The United States will maintain sufficient bases and air forces capable of attacking the strategic heart of any potential enemy.

c. The immediate destruction of the enemy’s will and capacity to resist is the primary objective of the United States Army Strategic Air Forces.

d. Extensive research regarding the strategic vulnerability of all major powers will be conducted later and will permit a more complete analysis of bomb requirements.

These are both bland and interesting at the same time. Should it really be US policy, in the waning days of World War II, to assume that it needs to be able to wage war against “any other world power or combination of powers”? That is quite a requirement — to take on the entire world. The issue of bases is similarly grand in ambitions: at this time, the US does not really have “intercontinental” capabilities, so having sufficient bases to strike “any potential enemy” is quite a lot. And the goal of “the immediate destruction of the enemy’s will and capacity to resist” is also a serious assumption, particularly with the requirement of immediacy (US strategic capabilities in World War II were not “immediate” and took years to develop and deploy).

All of which is to say that this document, so far, makes perfect sense from the context of the strategic bombing of World War II, and perfect sense from the context of a later Cold War, but is actually quite audacious in its assumptions from the perspective of the immediate end of World War II and the beginning of the postwar period. It reflects no sensibility about war being over, or that the goal of US policy should be to prevent future world wars.

It then launches into its analysis of how the atomic bomb, which it acknowledges is not “just another bomb,” would be used in such a war in the future:

It is assumed that the United States may be required to conduct military operations against any other nation or combination of nations in the world, and that, finding herself at war with these powers, the United States would be desirous of immediately crippling the ability of the enemy to wage war. It is to be noted that the requirements established in this paper contemplate an M-Day force capable of being employed immediately upon initiation of hostilities and the estimated quantities of bombs required must be available at that time. There has been no attempt to estimate the quantity of atomic bombs which would be required to conduct a prolonged war of attrition. Therefore, the assumption was made that the initial mission of the air force units allocated for preparation, transportation, and delivery of these atomic bombs should be the immediate destruction of the enemy centers of industry, transportation, and population.

Again, this is certainly not the only way one could imagine that atomic bombs would be used in a future war, although it is unsurprising that the strategic bombing analysts would approach it in this way. This “M-Day” force would be tasked with an immediate destruction of a hypothetical enemy, a nearly overnight thing.

And who is the hypothetical enemy? The study claims agnosticism about what the future might be, but, for the sake of argument, settles into the obvious one:

During the period 1945 to 1955 it is probable that at the beginning of any war, bombs will still be delivered by the conventional airplane. It is also obvious that during this period Russia and the United States will be the outstanding military powers. For the purpose of this study the destruction of the Russian capability to wage war has therefore been used as a basis upon which to predicate the United States atomic bomb requirements. It is to be noted also from a geographical aspect alone, Russia is in the most favorable strategic position of any major power.

Perhaps I am cynical, but this feels like a bit of a fig leaf — “we’re only looking at Russia (the USSR) because it’s going to be very powerful and we have to pick someone.” One somewhat just wishes they came out with it: “we are assuming that in a future war, the Soviet Union would be the possible aggressor, and certainly the one worth planning for.”

To tackle the issue, the stockpile estimate relies on the data of Intelligence Research Project No. 2532, “A Strategic Chart of Certain Russian and Manchurian Urban Areas,” dated August 30, 1945. This was produced by analysts in the Air Plans Section of Operations Division of the War Department, and consisted of tables of major cities in the Soviet Union and Soviet-controlled Manchuria. Each of these cities were listed by population, approximate area, and their “strategic importance” as measured by their industry, oil, and transportation infrastructure. A fantastic map of the target areas was produced:3

This intelligence study did not restrict itself purely to targets, though. What good is a target if you cannot reach it?

The atomic bombs of World War II were delivered by then state-of-the-art B-29 aircraft, but the range of the B-29 was still quite limited relative to the expanse of the Soviet Union and its distance from the United States. Indeed, the report contains a graphic showing what those limitations were, if one were assuming the continental United States as the launchpad:

The same oceans that provided the United States with a great sense of security from foreign attack, in this case, worked against it: the range between the United States and the Soviet Union was just very far.

There are answers to this problem, of course. One is to create the means for longer-range delivery of bombs. Another is to station bombers closer to the potential target. The analysts embraced both ideas: the next generation of bomb, the B-36, was planned to have much greater range than the B-29, and certainly in the meantime you could put B-29s much closer to Soviet soil. This, again, they illustrated with a map:

Of note is that this plan requires continual American access to bases in a number of countries, as well as expanding coverage to other possible bases that the US did not have a current hold in.

This is a bombers’ view of the problem; the planners were aware that there were “unlimited possible applications of the fundamental atomic energy in conjunction with future developments of rockets and guided missiles, both in their propulsion and in their explosive characteristics,” and limited themselves only to the next ten years as a way of reducing the analysis to something feasible.

So they had targets, and they had the means of hitting them (hypothetically). To figure out how many bombs they needed, they would need to make some assumptions about the damage-potential of the weapons, as well as the reliability and accuracy of the weapons. They had very little data on either of these fronts.

For the damage-potential, they had a preliminary survey by the Joint Target Survey, dated September 1, 1945, of the specific damage done at Hiroshima and Nagasaki, baed entirely on aerial photographs of city. (Only after the formal Japanese surrender, the next day, would US troops and scientists be allowed to actually visit the cities and conduct a more thorough survey.) They concluded that within about 6,000 feet (1.3 mi / 1.8 km) of ground zero, the number of buildings damaged was 100%. As you moved outward from ground zero, the percentage dropped off, until around 18,000 feet (3.4 mi / 5.5 km), at which it became negligible.

Based on this, and other analysis of the Japanese attacks, they came up with an estimate of how many bombs they would need to do devastating damage to each city in the Soviet Union and Manchuria, assuming the weapons available were the same as the ones used against Japan. They ended up with a table of “Estimated Bomb Requirements for Destruction of Russian Strategic Areas” for 66 Soviet cities, over 900 square miles of which to destroy, requiring 204 atomic bombs.

The biggest cities — Moscow, Leningrad (St. Petersburg), Tashkent, Novosibirsk, Kiev (Kyiv), Kharkov (Kharkiv), Riga, Kaliningrad, Odessa, and Ulan-Ude — required 6 atomic bombs. Seven more cities required 5 bombs, and another seven required 4 bombs. Twelve required 3 bombs, and fifteen required 2. The final fifteen took just 1 atomic bomb to destroy. None of the Manchurian cities were assigned bomb counts; the report reasoned that they weren’t as important as the Soviet ones. The number of bombs assigned appears to have been largely a function of size, although there seems to be some variance or judgment there.4

Even just attacking the top 15 priority cities would put the Soviet Union into a terrible position, as they contained “the bulk of all major industries” in the country. “It is obvious that the immediate destruction of the complete list of 66 cities would have an even more devastating effect on Russia,” the report notes. “Therefore, an optimum requirement for atomic bomb stocks would be the number necessary to obliterate all of these cities.”

From this kind of reasoning, the report then moves on to set some idealized numbers. To destroy the first 15 high priority targets would require (by their logic) 39 atomic bombs. Add 10 more for “neutralization of possible enemy bases in the Western Hemisphere,” another 10 for “strategic isolation of the battlefield” (tactical use) and you end up with 59 desired bombs. But, they reasoned, the weapons will not be all accurately delivered. They assume that through enemy action and inaccuracy, only 48% of the bombs launched might reach their targets. So you divide 59 by 48% and end up with a minimum need for 123 atomic bombs in the American stockpile.

If you do the same math, but assume the full 66 city list, with its requirement for 204 bombs, plus the 20 “other” bombs and the 48% effectiveness factor, then the optimum number of weapons would be 466 atomic bombs.

466 atomic bombs is a small number by the later standards of the Cold War, when thousands of bombs were produced per year. It is an absolutely massive number for 1945, however, when the maximum output of the Manhattan Project infrastructure was planned around the idea of a few bombs per month. As it turned out, the US stockpile grew very slowly in the first years after World War II, for a variety of reasons, and the US did not reach these kinds of numbers until 1949 or so. Interestingly, the US likely had enough fissile material for the “minimum” requirement by around early 1947, but its weapons assembly system was so anemic that it had essentially zero bombs “ready to go” at that time, and the parts for maybe a dozen weapons.

A copy of the overall report and recommendation was sent to General Leslie Groves, head of the Manhattan Project. He viewed it with not a small amount of professional offense — he felt the requirements were unrealistically high because the analysts had underestimated the power of the atomic bombs (he felt they were twice as effective as the report claimed) and overestimated how much destruction was actually required. “It is not essential to get total destruction of a city in order to destroy its effectiveness,” he wrote back. “Hiroshima no longer exists as a city even though the area of total destruction is considerably less than total.”

This wasn’t really a “war plan,” although it was a plan made in preparation for a war. The purpose of the document was to think through the questions that might come up for a future atomic war, without getting into the real details of it, and with very little knowledge about the weapons themselves. It would not be for several years until more realistic “war plans” were developed, and even these started more as “studies” than “plans”: exercises in thought that could guide policy decisions, not actual operations that could be put into practice, because they didn’t have the bombs, the bombers, or the bases yet. These are still works of fiction and fantasy, however rooted in analysis they attempt to be, and there is some difference between such fictional scenarios and, say, the reality of a ready-to-go war plan that could actually be executed.

Still, it seems telling that one of the first things the American strategic bombing staff did once their war with Japan was over was contemplate how best to kill an entire country with their new weapon. They make no attempt to estimate casualties, of course. Let us imagine that their population statistics were correct, and that their attacks were calibrated to kill an approximate equal percentage of those cities as the Hiroshima bomb did for Hiroshima. The “minimum” (15 city) list was estimated to contain 10.1 million people, and the “maximum” (66 city) list was estimated to contain 21.8 million people. At Hiroshima, around 28% of the city was killed within the first few weeks. Apply that number above, and we end up with a range between 2.8 million (minimum) and 6.1 million (maximum) dead.

By the standards of the later Cold War, where estimates in the hundreds of “megadeaths” were routine as a consequence of thermonuclear assaults and fallout plumes, 3-6 million doesn’t sound like that much. But it is sobering, and perhaps depressing — whatever one thinks of its necessity or appropriateness — to consider that the fires from World War II had hardly been extinguished before at least one of the armies of its victors, deep within the labyrinth of official secrecy, was contemplating another Holocaust’s worth of civilian dead.

I had initially planned for the LSE trip to be a day trip — train over in the morning, train back in the evening — which is tiresome but doable (the London–Paris train is about 3 hours or so). But I realized somewhat late in the planning that the return trip left a bit too early for that to work with the time of my talk, and so I decided, at the last minute, to stay the night and come back in the morning.

But where to stay? London is expensive, especially at the last minute, and I was not doing this as a holiday. When I looked online, I saw that there were some very cheap options in central London: “capsule” or “coffin” hotels, of the Japanese style. I thought, hey, let’s give it a try! For £34, it might be one night of bad sleep, but to be honest probably any option was going to be that (I don’t sleep that well in hotels, either), so I might as well see how it is. And the location was good; hard to beat that price for any bed in Piccadilly Circus, which was not far from where I would need to be for my train.

This place calls the little rooms “cocoons,” which is a better name than capsule or coffin, but they are still little coffins. I’ve read about these kinds of things and always wondered how they’d be in practice. There’s something very appealing to me about a little private space and a low price — I don’t mind a somewhat spartan existence for these kinds of overnight stays.

This one also leveraged modern technology in useful ways: there are multiple floors, and access to each floor is only allowed by a key card. Once on the floor, there are multiple rooms (“dorms”) with maybe a dozen or so capsules in them, and the key card again is necessary to get into your “dorm.” The actual cocoon does not have any kind of outside lock on it — you can bring or rent a padlock if you are going to leave stuff in it for the day — but you can lock it from the inside. There is a large (co-ed) shared wash/shower/bathroom area for each floor. There are CCTV cameras (it’s the UK, after all) in all common areas.

The cocoons are not that large. About the length of a couch. I am definitively average American male height and I could fit in the bed, but even I found it a bit cramped. Anyone taller than average is not going be able to stretch out all the way.

There is no storage space at all; there are a couple of hooks that you can attach a backpack or duffle bag to, but you’re basically just sharing the bed space with your luggage, clothes, shoes, etc. There is a light and a somewhat noisy fan, an electrical outlet, and a USB charging port.

The head clearance is pretty low, so you have to think about how you are maneuvering. You can sit up a bit. Definitely not for the claustrophobic (I am not). The walls are not thick so you can hear people talking or coughing; people were mostly pretty quiet, though. The noisy fan helps a bit as it acts like a white noise machine.

The biggest difficulties of the setup were that the bed was not quite firm enough for me, and that tends to make my legs ache a bit after awhile, and there aren’t a lot of options for repositioning them in such a small space. More importantly, the pod is pretty warm — the little fan can’t possibly deal with the ventilation of a (living) human being in a coffin, much less 20 of them stacked next to and on top of each other. It wasn’t a deal-breaker, but it wasn’t great.

All together, I would say that it was not a great night’s sleep — but again, I don’t sleep all that well in hotel beds anyway, so that is often the outcome for me. I wasn’t the worst. It wasn’t as enjoyable as the fantasy of it. Would I do it again? I don’t know — maybe in a real pinch. Better than sleeping on a street bench, I guess.

I think perhaps the real comparison would be between this and a “normal” hostel. I haven’t slept in one of those, although I’ve at times considered it for trips like this. What has kept me away from them is the shared dorm experience — I want my privacy, I don’t want to worry about other people snoring or partying or talking or thieving. And that is really the “problem” (real or perceived) that the cocoon is trying to solve: it is a shared dorm experience, but with some privacy. The problem is that the privacy comes at the expense of comfort, both in terms of reduced space and the lack of adequate ventilation. And you still are at the mercy of the others with you regarding noise, although in my case the others were pretty respectful (and perhaps everyone being locked up in their little cocoons reduces the will to socialize, or perhaps it is just self-selecting… who knows).

Could it have been better? Sure — if the cocoons were just a little longer, you could imagine them holding luggage a bit better and not feeling quite so cramped. If they were each a little taller and with some more engineering attention given to airflow and ventilation, you could imagine the warmth issue not mattering so much. Or so it seems to me.

The night was very quiet, though, so that is something. In the morning everyone basically got up and emerged at about the same time — probably by necessity, as anyone getting up produces a good deal of thunking of heads into pod ceilings. I washed up and checked out and wandered off to find some breakfast.

After that, I had a few hours to kill before my train, so I went for a walk and ended up in Regent’s Park, which was nice. I was amused by the presence of not one, but two statues along the theme of “ugly babies attacking birds.” While walking down the path, this amusing little Egyptian goose walked down past me, with great determination, like he had somewhere to be and had no time to talk. Perhaps he was worried about the advertised ugly babies getting him. Perhaps he was late for a very important date. I also saw a massive mute swan flying around the park at high speed, which was impressive, but I failed to get a good photograph of it in flight, alas.

I don’t have any novel thoughts on the news of the day, particularly the US and Israeli attacks on Iran. What will they bode, in the short or longer term? What kind of war is this? I don’t have any clever insights, just inquiet — “worried” in French, one of those little words that somehow works even better in English than in French. Every week is some kind of new fresh horror, danger, and spiraling uncertainty, with people at the helm who you wouldn’t trust to watch your dog (much less your children).

I do appreciate that being abroad allows me some sense of distance, some slightly removed sense of culpability, but it is not as much as one might desire. At least the weather has been nice here, I suppose — people are already speaking of springtime, something my experiences of living on the US east coast have caused me to be superstitious of, but it would be nice if it were true. Last week, there were days that were 60ºF / 16ºC here — truly warm! So I am grateful for that, at least.

Last week, I watched the made-for-TV nuke thriller By Dawn’s Early Light (1990), which I found to be amusing but not that great. I am aware that many people like it (there is little that such a statement cannot be said for), particularly people who saw it when they were young. Look, I liked Robot Jox (1989) when I was a kid, and I can still like it today, but that doesn’t make it, you know, great cinema. I wrote up some thoughts about it for Doomsday Machines while I was on my Eurostar train back to Paris on Friday, taking one of more of silly quotes in the film for the post title:

If you want to watch the entire thing easily, it is currently on YouTube, because apparently even HBO doesn’t really care about its copyright status these days:

Read more

There are many variations of this conundrum that have been explored in well-known and often well-loved films. Some of these films have themselves come to embody these problems expertly: Dr. Strangelove (1964) evokes the mad generals and the Doomsday Machine; WarGames (1983) raises the problems of automation, cyberattacks, and artificial intelligence; it is too soon to know if House of Dynamite (2025) will have such a cultural relevance, but if it does, perhaps it will be about the problems of attribution, confusion, and the short time-scales for nuclear decision-making.

By Dawn’s Early Light (1990) is not a well-known or well-loved film. It was an HBO “original movie” — read: made-for-television film, before HBO got good at such things — about the conduct of nuclear war. In almost every respect it is a very Cold War film, having been adapted from William Prochnau’s Trinity’s Child (1983), with a little bit of Glastnost window-dressing adding to keep it slightly topical.

I recently tried to explain the plot to someone briefly, and he seemed outright skeptical that I was relating it correctly, as it is a bit over the top. The basics are thus: the US detects a nuclear missile launch from Turkey to the Soviet Union which it has not ordered. The Soviets launch a number of their own missiles in retaliation at the USA. The Soviet president tells the American president that they determined, belatedly, that the Turkish missiles were launched by anti-Soviet separatists, but that automated Soviet defensive systems assumed they were in a war with the United States. The President is encouraged to stand down, but the Soviets will also agree to accepting a retaliation in kind, and end the conflict thence. Should the United States engage a larger counterattack, then the Soviet Union will escalate as well.

Thus far, this is something of a reverse Fail-Safe (1964) situation, which is not a terrible concept. Things start to go off the rails soon after. The President (played by Martin Landau) is apparently incapable of basic self-preservation and allows himself to get both partially nuked at the White House and later apparently dies when his helicopter encounters another Soviet nuclear detonation.

This causes a search for the next legal successor, who is an utterly unqualified Secretary of the Interior, who is given the code-name “Condor.” Having assumed the President’s duties onboard TACAMO — a presidential airplane equipped for nuclear war fighting — Condor quickly shows himself to be obsessed with “winning” the nuclear war and with little interest in deescalation.

A belligerent Colonel Fargo (Rip Torne), who is clearly regarded as an extremist by his military contemporaries, feeds Condor’s paranoia and desire for ultimate justice, encouraging him to issue a strike on Soviet command and control. The head of the Strategic Air Command, code-named “Alice” (James Earl Jones), himself on board Looking Glass (another command and control airplane), believes this is madness, but passes on the orders anyway.

A large part of the film is focused on the crew of a particular B-52 bomber, code named Polar Bear 1, that is scrambled in the early part of the crisis. Its crew are surprisingly dysfunctional, apparently having not prepared much for the job of thermonuclear war, or unable to deal with the stress of it. The pilot (Powers Boothe) and his co-pilot (Rebecca De Mornay) are engaged in an illicit sexual relationship (“fraternization”).

Over the course of their part of the film, they are attacked by Soviet MiGs (which they defeat and evade) and penetrate Soviet airspace. They are given the orders to conduct the strike on Soviet command and control. The co-pilot refuses to carry them out, claiming that they are madness — if you kill their leaders, who will be able to stop the war? The pilot demands compliance, to the point of offering the co-pilot a cyanide tablet, but ultimately backs down and agrees to turn the bomber back.

This, in turn, leads to one of the crew members who had previously exhibited some mental instability to attempt to take over the plane (calling them cowards), a situation that resolves itself when he uses an ejector-seat to (presumably) commit suicide, which ejects the other crew as well. And thus the pilot and co-pilot find themselves the only people on the plane.

Meanwhile, it turns out that the original President has not died, but was just injured, and was taken to a FEMA bunker. He regains consciousness (but has lost his eyesight), and attempts to wrestle back control of the situation. Having seen Polar Bear 1 turn back, the Soviets turned back some of their own bombers as well, as a sign of mutual desire for deescalation. In several hours, however, US submarines will surface to receive their own launch orders, which the Secretary of the Interior has given.

The President contacts the Secretary of the Interior, but cannot convince him to relinquish control — because the Secretary of the Interior doesn’t want to, we understand — and as the President’s copy of the authentication codes was lost in the attack that took his eyesight he cannot simply override him. Ultimately. the President contrives a way to issue his own orders to the submarines, but if the submarines receive two sets of contradictory orders, they will default to orders for war. Still with me?

To resolve this, the SAC commander, “Alice,” agrees to ram Looking Glass into TACAMO. This, however, appears to be impossible, as the Secretary of the Interior’s aircraft is too far away for Looking Glass to catch up to it before the final deadline, but the pilot on TACAMO aircraft realizes what is happening and turns his plane so that Looking Glass can collide with it, giving a final salute to “Alice.”

Having destroyed the Secretary of the Interior, the President’s orders to cease hostilities (apparently, somehow) go into effect.

Meanwhile, Polar Bear 1 was confronted by US forces who had been ordered (by the Secretary of the Interior) to shoot it out of the sky, but circumstances conspire to spare them. They fly eastward into the sunrise.

As a film, this is not that great. It probably works better as a mass-market paperback of the Tom Clancy variety; it certainly has that feel to it, and, at least according to Wikipedia, it is a relatively faithful adaptation. The acting is adequate. The overall feel is fairly cheap: most of the action takes place in a few cramped rooms (a very limited number of sets), with the action shots being obviously model planes or limited to computer screen readouts that are narrated by the characters. There is one nuclear mushroom cloud, but it is clearly a matte painting with a model plane in front of it. It definitely feels like a made-for-TV movie of this era.

The question to ask, I think, is not “is this a good film” — no, not really — but rather, what are the anxieties that this particular take on nuclear war represent? And in this department, if you get away from the rather silly aspects (planes ramming other planes, etc.), it does end up probing some interesting issues.

The first and most obvious one is the initiating attack: a separatist group (non-state actor) somehow hijacking an existing nuclear weapon and using it to provoke a nuclear war between major powers. Similar plots exist, although I primarily associate these with the post-Cold War (e.g., Clancy’s Sum of All Fears, 1991). So neither the 1983 source material nor the 1990 HBO adaptation are really behind the times on that one. This particular plot combines the non-state-actor provocation with an attribution problem and an automated defense system, so that is wrapping up a few anxieties into one package.

But that isn’t really what the film is about; that is just what kicks it off, and it doesn’t get talked about much after it happens. What the film is really about is the psychological state of people involved with nuclear weapons choices. I appreciate that it focused on both a “high” and “low” aspect of this question. The “high” involves the people at the very top of the chain of the command, like the President, the Secretary of the Interior, the top generals and colonels. The “low” are the people in charge of actually dropping the bombs, the B-52 crew.

On the “high” end of the chain of command, the problem one of appropriate judgment (prudence), madness, and the problem of succession during time of war. A Secretary of the Interior is not chosen for that role on the basis of their ability to conduct a nuclear crisis, even though they are indeed eighth in the line of succession. It is bad enough, perhaps, that a single person — the President — has this responsibility (for any President, really), but to then contemplate how eighteen other people might handle suddenly being put into that situation is, well, rather heavy.

In the case of the film, jostling for authority among the high end of the chain of command provides much of the film’s dramatic material. And who is to say how anyone would react to such a situation? In the film, many millions of Americans have already been killed by the war in practically an instant — how would any of us react to that, given these terrible options?

On the “low” end of the chain of command, the problem is multiple. How will a bomber crew react to learning that its home base has been eliminated, and any family it had behind are likely dead? Will they continue to function as a cohesive unit? How will they react to opening up their orders and finding them to be something they judge as madness? And how will the personal relationships between these crewmembers affect these discussions, if they happen at all?

Wikipedia says that in the original book, the pilot and co-pilot, though of mixed sex, are not in a relationship, and that this was an alteration made for the film. It is an interesting choice, as it evokes another common anxiety about how men and women would interact in military situations generally. In this case, it is the female co-pilot who refuses to go through with the orders that she believes will lead to madness, and the male pilot who initially insists upon following them, but relents when faced with the possibility of his lover’s individual death. These are, of course, pretty stock gender roles, but the question they are trying to get at — that individual relations and states of mind may play a role in how these sorts of things work in practice — are interesting ones.

Ultimately, the plot feels too contrived, and the bomber drama feels, well, unrealistic. The bomber crew in particular do not feel much like actual military personnel to me: they seem barely prepared, at least mentally, for their mission, and I just do not find it very believable that they would refuse to follow out these kinds of orders. Millions of Americans have died by this point; they know this. As far as they know, the escalation has already occurred. At that point, even one of their own crewmembers has died from the fighting. What are the odds that they are going to get moralizing? Much less turn back? What are they even turning back to? Even they do not know.

Still, the value of centering that part of the “action” around a bomber crew is that they do have some time to deliberate, their actions are not wholly automatic, and they could, in principle, rebel. It’s much harder to imagine ICBM launch commanders doing that sort of thing. There is a decent exchange to this point between the pilot and the copilot: “You would have turned a Minuteman key without thinking!” “They gave me time to think.”

Could one imagine some kind of struggle for authority playing out at the higher levels, in the event of a contested succession? Could we imagine someone utterly unqualified for the job aligning him or herself with escalatory elements, and ignoring the suggestions from other military officers? Could we imagine those senior generals following such orders anyway? These are trickier questions, and perhaps the more interesting ones, ultimately. They are rendered a little ridiculous in the film by making both “Condor” and Fargo so over-the-top — you’re never really tempted to agree with them at all. When Fargo pushes to “cut off the head of the Soviet chicken,” you are not tempted to see that as the product of a rational mind.

By Dawn’s Early Light isn’t in the running to be considered a great nuclear crisis film, but it’s not the worst possible nuclear crisis film, either. It raises a few interesting issues and has its moments, if you are willing to overlook the plot contrivances and the very obvious budget limitations (which necessarily means that no sense of what is happening “on the ground” is possible).

The film was surprisingly well-received at the time (although the reviews seem to be cutting it quite a lot of slack given its relatively humble made-for-TV status), although it doesn’t seem to have had much of any lasting impact on the cultural zeitgeist. It has not aged particularly well, and comes off as cheaply-made and fairly hokey. Still, some of the issues it gestures at — if you can focus on them and not the “thriller” contrivances — are interesting ones.

It makes for a curious pairing when watched alongside House of Dynamite (2025), Fail-Safe (1964), and other films that have tried to capture the psychological aspects of a failure of nuclear deterrence and the horrible choices that they may require.

If I were to speculate (and it wasn’t just a placebo effect), it feels like the experience of going to a formal class, sitting there with other students and an instructor, and being “on the spot” for being called on (I even had to go up to the chalkboard at one point, to conjugate the verb écouter, “to listen”), has caused some kind of switch to be thrown in my brain. I have read that a slight addition of stress improves language acquisition, which is why whenever one spends much time in any foreign country one starts to notice all sorts of language patterns even in languages you don’t speak at all (when I was in Japan, it “clicked” at one point that the character names of each of the cities we were in were on every license plate). And of course the context of education matters a lot, as every teacher knows.

Whatever the cause, it has been very fun to experience first-hand. Even now when I continue with my Duolingo I notice that my brain is much more effortlessly translating the French sentences I am giving, seeing them as “wholes” rather than “collections of words to be individually deciphered.” And going over even basic review in the classroom setting seems to be “cementing” a lot of basic concepts that I already basically knew but perhaps had to think about too much (and was thus prone to overthinking).

The classes are also fun just in general. The instructor is great — she speaks continuously in French but is always at a level of vocabulary and grammar that is appropriate for the class level, and articulates very clearly, and does all of this with an atmosphere of good humor that makes one feel comfortable, even though one’s brain is constantly a little bit stressed by the language. It makes the two hour class (twice a week; there were more “intensive” courses on offer but I was afraid they would really constrain my schedule too much) go by very fast.

My other classmates are also interesting and I hope to get to know them better over time. We are all étrangers — “foreigners,” but also “strangers,” which is an interesting linguistic lack of distinction — as one would expect, all adults, but the age range varies (I was for the first class clearly the oldest one there, but in my last class they had reshuffled some people from other classes and a few other members joined who were at least my age if not a bit older), as do the backgrounds and walks of life.

During the introductions I took some notes of the nationalities: American (at least one other aside from myself), Australian, Egyptian, Filipino, German, Indian, Iranian, Japanese, Peruvian, Polish, Russian, Tibetan, and Ukrainian. Their proficiencies seem to vary, too — some seem more comfortable talking but don’t seem to know the grammar very well, some seem very good on grammar but bad on talking, etc. Many have taken French classes before; my pure-Duolingo approach seems unusual. So I feel I am neither the best nor the worst prepared for the class, which seems about where I ought to be.

It has also been interesting because I haven’t been a student for some time now! It has been about 20 years (!) since I did actual coursework as a student, in graduate school, and since then almost all of my time in the classroom has been as a teacher. I am a much better student now than I ever was before. Perhaps that’s about motivation to some degree, but it’s also because as a teacher one can’t help but see students self-sabotaging all the time and think about how that’s not the way to learn anything. And I also have given a lot more thought to how the brain learns things over the years, which perhaps helps.

So now that the tables are turned, I am absorbing every word, thinking about every response, taking notes on everything even if I know it (because the act of taking notes is an act of synthesis), and so on. And I am also appreciative of the work of the teacher, perhaps because I can see better what her “craft” is having been in that position myself, and am perhaps tailoring my own approach to the class as to what I would desire of a student in my own classes (my phone is away during class). And, in fact, on Wednesdays I go from teaching my own course at Sciences Po to being a student in my French course.

So all of that is quite good and makes me feel like taking the course was even a better decision than I thought it would be when I signed up. The Duolingo approach (which I am still doing, every day) has been very useful and gotten me surprisingly far, but most of my gains with Duolingo have been quite slow except for vocabulary and reading (I can read at about a B1 level, which is enough to read a newspaper and understand most of it), and I am just not sure that an app can get one’s brain wired around speaking and listening correctly (Duolingo has an “AI chat” talking feature and it is pretty terrible and I think not helpful at all; certainly it is nothing like having actual conversations with actual humans).

I think combining something like Duolingo with an in-person class might round-out the deficits inherent to a formal class as well — only a few days per week, not a lot of instant feedback on things like homework, limited ability to adapt to individual needs, etc. Anyway, we shall see!

Last week I saw notice that the ethnographic documentary filmmaker Frederick Wiseman had passed away. This was hardly a surprise — he was 96! — but as someone who has been watching his films for a few decades now, and often assigns them in classes and recommends them to colleagues, I heard of it with a somewhat heavy heart.

Wiseman’s documentaries for me have always been the “gold standard” for what contemporary ethnographic examination of “first world” society looks like. I believe the first Wiseman documentary I saw was Near Death, which I saw as an undergraduate at UC Berkeley in a course on “the History of Death” (I always did seek out cheerful subjects), taught by Thomas Laqueur. In graduate school I saw Primate in a course on “Filming Science” taught by Peter Galison and Robb Moss. And independently of any course I have watched High School, Hospital, Meat, Essene (about a French monastery), Welfare, Boxing Gym, and probably some others that I am not recalling immediately.

These are not fun films to watch. They are studies. They require immense patience and concentration. Some of them are frankly shocking but not in a fun way. During Primate, one gets a truly intimate look at how primate study takes place in a scientific setting, and it is, well, not an easy thing if you are someone who cares about the mental and physical states of other animals, much less fairly intelligent ones. There is one scene in particular where, after carefully attending to a Rhesus monkey, the scientist in question neatly decapitates it and begins to remove its brain. It is done clinically and no doubt as a matter of complete routine, but one gets no notice that it is about it happen, and the viewer has already spent some time with this monkey.

But that is, of course, a window into how science is actually done in laboratories, and the very mundanity of that rendering of the living monkey into a monkey’s brain — the true object of study — says something profound about the scientific approach to the world. I say this without judgment of science for this; that is where its power of description comes from! But it is not how we like to talk about how science works, because it is in fact quite disturbing to see the living world rendered into data, coldly and clinically, and absent the narrative context that we usually wrap around this activity to make it more acceptable to the public.

That’s the value of an ethnographic approach, and that’s the value of Wiseman’s films. It is not so much that they “let you see things as they are” — Wiseman himself was clear that there was an editorial mind at work, that the films were necessarily incomplete and heavily manipulated records — but it does let you see things closer to how they operate in the mundane, day-to-day, on their own terms. It is a closer view, anyway, than these institutions typically present to the public, for good reason.

Anyway, in case you missed it, for Doomsday Machines last week I wrote up a discussion of Wiseman’s 1988 film, “Missile,” which is about the training of launch officers for Minuteman intercontinental ballistic missiles in the mid-1980s. The film, I emphasize, is tremendously boring, but tremendously important if you are someone interested in how these kinds of institutions function. I know of no other source that gives quite as good a view of what it means to be a nuclear launch officer:

A reader pointed out in the comments that the final simulated launch sequence in the film, with the two women (and from which I took the thumbnail above, which is an amazing subversion of the normal “nuclear keys” trope, if only for the fingernail and all that it signifies alone), is available on YouTube:

I find the analog 24-hours clocks they use in the silos — seen in the sequence and in the thumbnail above, and in a earlier portion of “Missile” the instructors emphasize how important it is for the officers to learn to tell time on them — a fascinating artifact in and of themselves. I have never really noticed them but apparently they are common in many “official” contexts, and George Orwell has the clocks “striking thirteen” in 1984 as a symbol of the militarization of his future society. Fascinating stuff.

OK, that’s a lot for one week — I was feeling guilty about the lack of Lyndon photos last week, so I tried to make up for it this week! Next week I will be taking the Eurostar again, as I am giving a talk about the new book at the London School of Economics next Thursday. I am getting to be an old hand at this Eurostar thing…

Read more

Wiseman’s documentaries are essentially a genre unto themselves, a fly-on-the-wall approach to institutional ethnography. Wiseman’s subjects are all social organizations, institutions, or spaces of widely varying types, and the titles are typically spare and descriptive: High School (1968), Hospital (1970), Juvenile Court (1973), Racetrack (1985), Central Park (1990), Boxing Gym (2010). Some of them are even more simple and declarative, such as Primate (1975), which focused on the Yerkes National Primate Research Center, or Meat (1976), about a slaughterhouse.1

Wiseman’s films typically eschew any added soundtrack, informative text overlays, or obvious “plot.” Somehow, over many, many decades, he talked himself and his crew into these often quite sensitive places and was allowed to just film, film, and film. Out of vast amounts quantities of filming — hundreds of hours per film, probably — the people who he films appear to just forget he is there, and he then edits the eventual footage into what could be described as a profile of his institutional subject.

They are not, of course, truly narrative-free. The films always tell some kind of “story,” although not often one with any kind of clear narrative resolution. They always tell you something about the kind of people who come together in whatever place he is profiling. Wiseman doesn’t tell you how a hospital works in Hospital, for example, but he shows you the kinds of spaces that hospitals are: collisions between people of all walks of life, under circumstances both mundane and exceptional. The identities of the characters in his film are usually unclear except in their relationship to others: the teacher and the student, the doctor and the patient, the researcher and the monkey.

Wiseman’s films can have an illusion of unvarnished objectivity, although Wiseman was clear that this was truly just an illusion:

What you choose to shoot, the way you shoot it, the way you edit it and the way you structure it... all of those things... represent subjective choices that you have to make. … The compression within a sequence represents choice and then the way the sequences are arranged in relationship to the other represents choice. All aspects of documentary filmmaking involve choice and are therefore manipulative. … I think what I do is make movies that are not accurate in any objective sense, but accurate in the sense that I think they’re a fair account of the experience I’ve had in making the movie.

One of Wiseman’s least-heralded films, in my experience, is one that came out in 1988 with the simple title Missile. Missile is about the training of Minuteman launch officers at Vandenberg Air Force Base, and appears to have been filmed in 1986 (at the end there is a reference to the recent Challenger accident). It is exceedingly rare to get this kind of “access” to missileer training, and over time it has also no doubt become something of a time capsule of that particular Cold War moment.

Like most of Wiseman’s films, Missile is interesting in what it shows, but the act of watching it is one of incredible boredom, and requires considerable patience. The film begins with what feels like the first class in a 14-week training sequence, and these classes are not meant to be exciting. Missiles are not meant to be exciting: if working in a nuclear weapons facility is exciting then something has probably gone very wrong. The boredom of it all is contrasted with the “awesomeness of this responsibility,” as the first instructor puts it, of the task at hand: being the ones who would, if called upon it, be turning the keys that launched thermonuclear-tipped ICBMs.2

The first 20 minutes of the film are just this class, in fact, and the focus of the class is on the ethics of the US nuclear weapons system. It is a meandering discussion, with the instructor asking the students what they knew about the My Lai massacre, and about the Nuremberg trials and the “just following orders” defense. The students were aware of these things, but the instructor never really (at least in Wiseman’s showing) hammered home any particular conclusion about them. If anything, he seemed to defend aspects of My Lai, emphasizing the civilian-appearance of the Viet Cong.

So what does it all add up to? The instructor never really resolves any of the ethical or even legal questions on film, other than to be very clear that while they do not want their launch officers to be “robots” (a word he pronounces almost like “ribbits,” an accent affectation that fiction could never capture), they also do not want any launch officers having any kind of second thoughts before they take the role. Anyone who has the slightest doubt, or ethical or moral hesitation, about following through on a properly-authenticated Emergency War Order (EWO, pronounced ee-woe), is explicitly told to leave before they waste too much of everyone's time.

Which is a bit awkward. One student — one of the few women — offers up the notion that the mere presence of the missiles deters nuclear war, and that this makes the mission a moral one. The instructor doesn’t really appear to take her up on this with much enthusiasm, even though it is a standard line about these jobs.

Perhaps he understands why this doesn’t really solve the issue he’s posing: the moral/ethical question is not would you accept maintaining deterrence, the moral/ethical question is, if deterrence fails, would you follow an order that could result in the deaths of millions of civilians. Which is to say, deterrence can perhaps create the moral “cover” for doing nuclear-weapons related work… but if you’re actually going to be turning those keys, you need a bit more than that to fall back on.

Or perhaps he doesn't care all that much. Because the goal does not appear to give the candidates that kind of moral absolution. Rather, the goal of this moral discussion appears — to me at least — to be about weeding out anyone who will be a problem. Which is to say that the goal doesn’t appear to create rigorous moral or ethical justification for actually turning the launch keys, but to simply make sure that nobody who is proceeding with the training actually requires that kind of justification.

Which is not necessarily a critique, although bringing up this issue in the context of My Lai and Nuremberg lends itself to a certain kind of critique, and that seems to be Wiseman’s point. The entire training, and its Southern California setting (which Wiseman occasionally shows us), is, well, banal. It’s not necessarily the banality of evil, but it’s banal nonetheless. There are a few high-level discussions about things like ethics, the domestic legal authority for using nuclear weapons, and even the goal of nuclear deterrence, but for the most part this feels tacked-on to the main part of the training, which is about technical procedures for making sure the missiles are ready to launch at all times, and then launching them when ordered to do so.

What is the mission of the Minuteman missile system? An instructor asks the candidates this question, and they dutifully refer to a manual in front of them where they read:

The mission of the Minuteman hardened and dispersed weapon system is to deliver thermonuclear warheads against strategic targets from hardened underground launchers in the continental United States.3

That’s the mission. Not deterrence. Not preserving freedom. Those things, to the degree that they are at play at all, are the more abstract justifications for the system. The system itself is an immensely complex sociotechnological arrangement of materials, electronics, regulations, and, at the heart of it, people. It is mundane, it is banal, it is tangible. This is what Wiseman is showing us — the banality of deterrence, perhaps.

And of course, as with all instruction, there is a bit of banter and humor, even about grim topics. Some of the candidates have previously worked with liquid-fueled Titan II missiles, and one of the instructors joked about how much easier the maintain the solid-fuel Minuteman missiles are:

Titan is, in a word, a volatile system. You drop a wrench, you can launch a missile — something like that. No, I’m just kidding.

He smiles and raises his eyebrows, and we can tell he’s only somewhat kidding. (In 1980, a dropped wrench set off a chain of events in a Titan II silo in Damascus, Arkansas, which resulted in the missile fuel detonating, ejecting its thermonuclear warhead, and killing one officer. Perhaps the influx of Titan candidates is because they began to decommission Titan II bases after that.)

As one of the instructors tells the candidates, the classroom work is abstract, and the real learning, and telling who has learned, takes place inside the simulated Launch Control Center where the candidates will drill on both the day-to-day aspects of being a missileer and the act that will hopefully never occur — the launch of the missile. Wiseman’s crew is allowed to see this in some detail, and we watch various candidates stumble through the procedures with varying degrees of confidence, observed all the while. “When you get into the EWO [Emergency War Order] part of the mission, that’s the job,” one observer lectures them. “Launching missiles is your job.”

Wiseman tacks between the classroom, the simulator, and the even more truly mundane activities on the base — a cookout and a softball game. The classroom is where abstract discussions take place about the nature of nuclear war. Like all people who work with these things on a daily basis (myself included) they lapse into an easy familiarity with horrible ideas. The ethical and legal issues appear to be reserved exclusively for day one, and further talk becomes about “the job” and is lack of individual responsibility to do anything other than to launch the missiles if asked to. As one instructor puts it:

The thing you want to do is get your [launch] key turn in when you’re supposed to. And to do that, it needs to be very rudimentized [sic], very structured, you don’t want to let anything bother you, just get the task done, and leave it at that. Get a successful launch and take care of any other problems that happen afterwards.

There is a lot of discussion, both in the classroom and in meetings between instructors (which, again, somehow Wiseman had access to) about making sure that weak candidates are either rejected or improved. We get to see one instructor giving a candidate test-taking strategies. If you have four options and don’t know which one is correct, choose the longest one, it is usually correct. If you have four numerical values and don’t have a clue which one it is, throw out the extremes and guess between the remaining ones. Perhaps these are decent test-taking strategies, but it feels a bit insane to imagine telling someone good strategies for guessing on a test about nuclear weapons use.

Wiseman’s eye is clearly drawn to the presence of women in the training, and the fact that it appears to be a novelty; some of the classes look like women make up roughly 25% or so of those present, while some seem to have none. All of the instructors and senior officers appear to be male. In one discussion, apparently with all men, the issue of “fraternization” is raised, with male candidates warned that while they do not need to shun friendship and camaraderie with the women, they should make sure that, “just don’t get it to the point where people start thinking that something’s going on.” In another class, where everyone appears to have put in for assignment at Whiteman Air Force Base (Missouri), the women there make the point that Whiteman was at that time the only Minuteman base that had women launch officers.

One of the final sequences in Missile is of two women candidates running through a full simulation of an Emergency War Order, launching the ten (simulated) Minuteman missiles assigned to them. The women do a great job; they have studied the procedure, they know what they are doing, even if it is clear they are still new at this (they both attempt to put the launch keys in upside down the first time, for example, but quickly resolve the issue). They are efficient, working together to prove, to anyone who might doubt, that they would be perfectly capable of fighting a nuclear war if called upon to do so.

Is Missile a good film? Yes. Is it an important film? Yes — everyone seriously interested in nuclear weapons ought to see it, in my mind, because it gives an insight into these “people of the bomb” that is otherwise hard to capture, and informs common questions even today like, “would the launch officers carry out nuclear launch orders under any circumstances?” (Answer: probably; that is plainly a large part of what they were selected for.)

Is Missile fun to watch? No. It’s boring. And that’s the point. It makes perfect sense that films about nuclear weapons that are trying to capture the interests of a disinterested public are often thrillers, but the everyday reality is just… reality. It’s just another facet of life. The part that’s not normal is that these activities, if taken down certain roads, lead to the death of thousands, millions, even possibly billions. It’s the juxtaposition, between the tedious checklists and acronyms on the one hand, and those terrible possible futures on the other, that makes everything about nuclear war feel a little surreal.

Wiseman’s Missile captures these juxtapositions beautifully. The best shot, for my money, is the one (above) of a female candidate turning her launch key in the simulator. It is a portentous image: if this had been a real launch, some 30 nuclear warheads would now be on their way to the Soviet Union, each capable of releasing around 20 times the energy as the Hiroshima bomb. Wiseman is zoomed in on her hand as she turns the key, and we can see her red nail polish, a very different kind of signifier. After the indicators signal that the missiles are out, and her job is completed, she concludes dutifully: “And that’s it. That’s all she wrote.”