The Doomsday Machine

by Daniel Ellsberg

  The instructions he briefed to the crews just before the raid were unique in the history of bombing up until that time. The enormous B-29s were designed to fly at very high altitude, very fast, and in a tight bomber stream to deal with fighters with their coordinated guns. The tactics he prescribed that night were ones the crews had never heard of before. They were not to convoy. They were not to go up to high altitude. They were not to circle around, using up fuel, until others got in place for an enormous stream that was to go high over the city. Instead they would crisscross the city from their bases by the most direct route. Therefore they would save a great weight of fuel, which would go into extra bombload.

  Most dramatically, they were to strip the planes of guns and ammunition, thus saving another ton and a half of weight for bombs. By these tactics, he counted on increasing the bombload of his 334 planes by over 50 percent. They could each go in with six to eight tons of bombs, mostly incendiaries.

  LeMay determined that he wouldn’t inform General Arnold of his plans—thus protecting his superior from blame if the mission were unsuccessful. The development of the B-29s was Hap Arnold’s pet project; he considered them the key to the future of the Air Force. But their development and production had cost more than the Manhattan Project, and they’d had technical problems that had kept them out of the war in Europe. Partly due to the weather over Japan—almost constant overcast, and a jet stream at high altitude of two hundred miles an hour that made accurate bombing impossible coming or going—they hadn’t shown much for their money.

  LeMay’s superiors in Washington desired above all to prove that the 29s could do a big job and keep strategic bombing in the war in the Pacific, which would make the case for getting an independent air force and keeping strategic bombing after the war. What LeMay was sparing his bosses from knowing was not the deliberate firebombing of civilians; he knew that was what they wanted, really what they had sent him there for. What he chose to conceal from them until the last moment was the radical tactics he was going to employ, potentially dangerous to costly aircraft and crews though possibly essential to the “results” they wanted. He planned to take personal responsibility for the tactics and their possible failure.

  In the Kantor-transcribed staccato flow of words (ellipses below as well as italics are from the original text), LeMay reflects:

  … Plenty of strategic targets right in the primary area170 I’m considering. All the people living around that Hattori factory where they make shell fuses. That’s the way they disperse their industry: little kids helping out, working all day, little bits of kids. I wonder if they still wear kimonos, like the girls used to do in Columbus in those Epworth League entertainments, when they pretended to be Geisha girls, with knitting needles and their grandmother’s old combs stuck in their hair.

  … Ninety percent of the structures made of wood. By golly, I believe Intelligence reports said ninety-five! And what do they call that other kind of cardboard stuff they use? Shoji. That’s it.

  … Each type of weapon has some good points as well as some bad points; but if I now had my choice, and had available an overwhelming quantity of any type of fire bomb which could be employed, I wouldn’t stick to one particular type. No. Of course magnesium makes the hottest fire, and it’ll get things going where probably the napalm might not. But the napalm will splatter farther, cover a great area. We’ve got to mix it up. We’re not only to run against inflammable wooden structures. We’re going to run against masonry too. That’s where the magnesium comes in handy.

  … No matter how you slice it, you’re going to kill an awful lot of civilians. Thousands and thousands. But if you don’t destroy the Japanese industry, we’re going to have to invade Japan. And how many Americans will be killed in an invasion of Japan? Five hundred thousand seems to be the lowest estimate. Some say a million.

  … We’re at war with Japan. We were attacked by Japan. Do you want to kill Japanese, or would you rather have Americans killed?

  … Crank her up. Let’s go.

  LeMay’s recollections in 1965 that civilian casualties were a regrettable, unavoidable side effect of attacks intended to destroy “home factories” were as disingenuous as the RAF euphemisms about housing and industry being the objective when German cities were blanketed with incendiaries. (Moreover, in reality, the home-factory system had been abandoned by the Japanese in late 1944.) Many years later, Roger Fisher, a long-term Harvard law professor—who had been a close friend and consultant to my boss John T. McNaughton when I was working on Vietnam in the Pentagon—mentioned to me that he had been General LeMay’s “weather officer” in Guam at the time of the Tokyo raid. That got my attention, and I asked him what he remembered of that night. He told me, “I briefed that day, as usual, on the weather to be expected over the target, and he asked me a question I’d never heard before. He asked, ‘How strong are the winds going to be at ground level?’ I started to tell him we could predict the winds at high altitudes, with reconnaissance flights, and even at intermediate altitudes if we dropped balloons, but we had no way of knowing what the ground winds would be. But he broke in and asked me, ‘How strong does the wind have to be so that people can’t get away from the flames? Will the wind be strong enough for that?’ ”

  “What did you tell him?”

  “I didn’t know what to say. I stammered something about how I didn’t know the answer to that, and I left and went to my quarters. I didn’t go near him again that night. I had my deputy deal with him. It was the first time it had entered my head that the purpose of our operation was to kill as many people as possible.”

  LeMay’s instructions were very frightening to the bomber pilots when they heard them at the briefing. Incredible. Going in at low altitude, almost naked of guns; they had never heard anything like this. He hated to send them in by themselves, he said, without him in the lead. But they went.

  Again, LeMay from Mission with LeMay:

  [Up]drafts from the Tokyo fires171 bounced our airplanes into the sky like ping-pong balls. A B-29 coming in after the flames were really on the tear would get caught in one of those searing updrafts. The bombers were staggered all the way from five to nine thousand feet, to begin with. But when the fires sent them soaring, they got knocked up to twelve and fifteen thousand feet.

  According to the Tokyo fire chief the situation was out of control within thirty minutes. It was like an explosive forest fire in dry pine woods. The racing flames quickly engulfed ninety-five fire engines and killed a hundred and twenty-five firemen.

  The airmen found the glow of the flames lighting the sky. The clouds, they said, looked like cotton wool dipped in blood from a hundred and fifty miles away. It was a false dawn over Japan.

  The Tokyo fire was not, by definition, a classic firestorm (though it’s usually described as such), drawing winds into a defined area from all directions. There was a ground wind blowing. If Fisher had been able to predict it, LeMay would have found the answer to his question very reassuring. The effects of the wind went far beyond his requirement. The Japanese called it a red wind, akakaze, whose speed got to be quite high, twenty-eight miles an hour. This meant that the blaze moved ahead of the wind and developed a kind of mass fire—akin to a firestorm—known as a sweep conflagration, a tidal wave of flame which planners had hoped to get before, but the wind conditions had to be exactly right. And on this night they were.

  This moving wall of flame rose hundreds of feet in the air. It projected radiant heat, invisible infrared rays, ahead of it that would knock people down and burn them before the flames even reached them. It had all the effects of the firestorms in Hamburg and Dresden, but winds acting as a bellows produced temperatures even more intense than in those conflagrations, eighteen hundred degrees Fahrenheit. People fleeing suffocation in the shelters took to the streets to escape and became blazing torches unable to move in the melting asphalt. Tokyo, like Venice, was covered with canals, to which mothers raced with their children to get away from the heat. The sm
aller canals began to boil, and families boiled to death by the thousands.

  Between eighty thousand and a hundred and twenty thousand people were killed that night. Many of the crews of the bombers had to put on their oxygen masks, at five thousand feet, a mile above the flames, to keep from vomiting from the sweet, sickening smell of burning flesh.

  LeMay continues:

  Contrary to supposition172 and cartoons and editorials of our enemies, I do not beam and gloat where human casualties are concerned.

  I’ll just quote AAFWW II [Army Air Force World War II] volume V, page 627, and let it go at that. “The physical destruction and loss of life at Tokyo exceeded that at Rome … or that of any of the great conflagrations of the western world—London, 1666 … Moscow, 1812 … Chicago, 1872 … San Francisco, 1906 … Only Japan itself, with earthquake and fire of 1923 at Tokyo and Yokohama, had suffered so terrible a disaster. No other air attack of the war, either in Japan or Europe, was so destructive of life and property.”

  The italics are my own. [LeMay’s.]

  General Arnold wired me, “Congratulations. This mission shows your crews have got the guts for anything.” It was a nice telegram but I couldn’t sit around preening myself on that. I wanted to get going, just as fast as was humanly possible.

  It would be possible, LeMay thought, “to knock out all of Japan’s major industrial cities during the next ten nights.” And he set out to burn the next most populous seventeen cities in succession. After that, the next fifty.

  * * *

  The new campaign was not a secret from the American public. Time, in its issue dated March 19, 1945 (released March 12, two days after the Tokyo firebombing), had an accurate account of the tactics, the incendiary bombloads, and the operation’s intent. The lead, under the heading “Firebirds’ Flight”:

  A dream came true last week for U.S. Army aviators: they got their chance to loose avalanches of fire bombs on Tokyo and Nagoya, and they proved that, properly kindled, Japanese cities will burn like autumn leaves.

  Giving LeMay’s estimate that fifteen square miles of the city had been totally destroyed, Time noted:

  Never before had there been an incendiary attack of comparable scale. The Luftwaffe’s “great fire raid” on the City of London (Dec. 29, 1940), made with a maximum of 200 tons of incendiaries, burned not more than one square mile. Major General Curtis E. LeMay’s Marianas firebirds were in another league.

  No estimates of Japanese casualties appeared in this story, but not because of American sensitivities to the deaths of their enemy. Another story in the same issue, describing the success of American troops in the Pacific digging Japanese troops out of their holes and bunkers with napalm and flame-throwers, was titled “Rodent Exterminators.”

  After further raids on Tokyo in May, the New York Times was reporting173 casualty estimates for Tokyo civilians that were actually exaggerated. Under a three-line headline claiming TOKYO ERASED, SAYS LEMAY was this headline on an independent article:

  51 Square Miles Burned Out

  In Six B-29 Attacks on Tokyo

  LeMay Backs Figures With Photos of Havoc

  —1,000,000 Japanese Are Believed

  to Have Perished in Fires

  John W. Dower notes that in the accompanying article,

  only in the eleventh paragraph,174 on an inside page, did it get to the astonishing estimate of fatalities—and suggest that the subhead may in fact have been restrained. “It is possible,” the Times reported, “that 1,000,000, or maybe even twice that number of the Emperor’s subjects, perished.” The remainder of the article focused on the dates of the six raids and number of B-29s lost.

  The fatality estimate for Tokyo was exaggerated by a factor of ten or twenty, but more suggestive in retrospect is how casually such a staggering number of projected Japanese civilian deaths could be reported, and tucked away, by this date. It did not even qualify as the lead story.

  When Truman later mentioned that neither the prospect nor the actual use of the atom bomb ever gave him a moment’s hesitation or a night’s troubled sleep,175 that seemed odd to many Americans, including myself when I first read it. After all, he might have said that it was a difficult, in fact anguishing, moral problem, a grave decision, but that there was just no way around it. How could it not be a moral challenge?

  But Truman sometimes went on to mention something that was scarcely clear to many Americans then, and still is not: that we had long been killing more people than that in the course of our non-nuclear firebombing attacks. And that was true—not only for Truman but also for FDR before him. For five solid months before August 1945, the U.S. Army Air Force had been deliberately killing as many Japanese civilians as it could.

  The atomic bomb simply did it more efficiently, one bomb doing what it took three hundred bombers to do in March. But we had three hundred bombers, and more, and they had been doing the same job, night after night, city after city, some sixty-seven of them before Hiroshima. The United States Strategic Bombing Survey reported, shortly after the war, “It is probable that more persons were killed in one six-hour period176 … than in any other recorded attack of any kind.”

  Contrary to Stimson’s highly influential but totally misleading account in Harper’s in February 1947, “The Decision to Use the Atom Bomb”—written for Stimson by McGeorge Bundy177 while he was in the Society of Fellows, and a successful propaganda counter to the impact of John Hersey’s New Yorker report “Hiroshima” in August 1946—there was no moral agonizing at all among Truman’s civilian or military advisors about the prospect of using the atom bomb on a city.† That moral threshold had been crossed long before. There was, in reality, no debate or even discussion whatever in official circles as to whether the bomb would or should be used, if it were ready in time before the war ended for other reasons.

  One such foreseeable reason for Japanese surrender before the bomb was dropped would be the announcement at Potsdam in July of the scheduled Soviet entry into the war against Japan on August 8. The Soviets wished but were not permitted to sign the Potsdam Declaration, which would have announced the end of their neutrality with Japan (and unavailability of the Soviets as a mediator with the United States, which—we knew from intercepted communications—the Japanese were counting on to get better surrender terms). Another possible ending might come—as recommended by the Combined Chiefs of Staff and by virtually all civilian advisors except Byrnes—from informing the Japanese (before the Soviets entered the war in August) that they would be permitted to keep the imperial institution and Hirohito as emperor, as the United States intended. Neither of these possibilities, both well known by high-level insiders, was mentioned in the Stimson article.

  Seventy years of public controversy about “the decision to drop the bomb” have been almost entirely misdirected. It has proceeded on the false supposition that there was or had to be any such decision. There was no new decision to be made in the spring of 1945 about burning a city’s worth of humans.

  The atom bomb did not start a new era of targeting or strategy or war making in the world. Annihilation of an urban civilian population by fire had already become the American way of war from the air, as it had been the British way since late 1940.

  Thus, there is an ironic undertone to the judgment on the atomic bombings by Admiral William D. Leahy, chief of staff to Presidents Franklin Roosevelt and Harry Truman, in his postwar memoir:

  It is my opinion178 that the use of this barbarous weapon at Hiroshima and Nagasaki was of no material assistance in our war against Japan. The Japanese were already defeated and ready to surrender because of the effective sea blockade and the successful bombing with conventional weapons.

  The lethal possibilities of atomic warfare in the future are frightening. My own feeling was that in being the first to use it, we had adopted an ethical standard common to the barbarians of the Dark Ages. I was not taught to make war in that fashion, and wars cannot be won by destroying women and children.

  Ther
e is no record of Admiral Leahy—or anyone else in the U.S. government—having expressed such an opinion to his immediate boss FDR in the last month of the president’s life, nor to his next boss, Harry Truman, about the prior four months of destroying women and children in Japan. Those direct attacks on Japanese civilians had begun under Roosevelt, Stimson, and Leahy when, as their subordinate General LeMay put it, “we scorched and boiled and baked179 to death more people in Tokyo on that night of March 9–10 than went up in vapor at Hiroshima and Nagasaki combined.”

  LeMay himself was convinced that fire bombing had brought the Japanese to the point of surrender and that the atom bomb was in no way necessary. That last opinion was not at all confined to Air Force commanders, though Navy commanders, with reason, put more emphasis on the effects of the submarine blockade. The judgment that the bomb had not been necessary180 to victory—without invasion—was later expressed by Generals Eisenhower, MacArthur, and Arnold, as well as Admirals Leahy, King, Nimitz, and Halsey. (Eisenhower and Halsey also shared Leahy’s view that its use was morally reprehensible.) In other words, seven of the eight officers of five-star rank in the U.S. Armed Forces in 1945 believed the bomb was not necessary to avert invasion (that is, all but General Marshall, chief of staff of the Army, who alone still believed in July that invasion might have been necessary). Likewise, the U.S. Strategic Bombing Survey for the Pacific War concluded in July 1946 (in a report primarily drafted by Paul Nitze):

  Based on a detailed investigation181 of all the facts and supported by the testimony of the surviving Japanese leaders involved, it is the Survey’s opinion that certainly prior to 31 December 1945 and in all probability prior to 1 November 1945, Japan would have surrendered even if the atomic bombs had not been dropped, even if Russia had not entered the war, and even if no invasion had been planned or contemplated.