Throughout the period when we were planning our atomic bombing operations against Japan, American strategy was based upon the assumption that an invasion of the Japanese homeland was essential to ending the war in the Pacific.
Under the strategic concept approved by the Joint Chiefs in July of 1944, Kyushu was to be invaded on October 1, 1945, with the final assault on Tokyo following in December of that year. This plan was not dependent in any way upon Russian co-operation.
After the Yalta Conference, however, a debate sprang up over whether it would not be better to encircle Japan and defeat her by attrition than to defeat her by direct attack. Both General MacArthur and Admiral Nimitz, when asked for their opinion, voted for a direct assault. Supported by these highly respected judgments, the Joint Staff reiterated its previous preference for a strategy of invasion, reporting in April, 1945, that:
In view of all factors, we should follow the strategy of early invasion and our course should be:
It was estimated that a force of 36 divisions—1,532,000 men in all—would be required for the final assault, and it was recognized that casualties would be heavy.
The Joint Chiefs of Staff approved the concept and on May 25 the directive for the Kyushu invasion was issued to General MacArthur, Admiral Nimitz and General Arnold. The target date for the invasion of Kyushu was now November 1, 1945.
The record of a meeting between President Truman and the Joint Chiefs on June 18, just before the American delegation left for Potsdam, reads:
THE PRESIDENT said he considered the Kyushu plan all right from the military standpoint and, so far as he was concerned, the Joint Chiefs of Staff could go ahead with it; that we can do this operation and then decide as to the final action later.
In such a climate, no one who held a position of responsibility in the Manhattan Project could doubt that we were trying to perfect a weapon that, however repugnant it might be to us as human beings, could nonetheless save untold numbers of American lives.
However, one very important question arose: should the United States delay any contemplated military action in the expectation that an effective atomic bomb would be produced as scheduled? To any experienced soldier it was obvious that, once an advantage had been gained over an enemy as dangerous as Japan, no respite should be given. If the bomb had been scheduled for delivery in early November, a few days after the scheduled date of the Kyushu invasion, I would have advised a delay in the landing operation. I expressed this point of view in conversations with Secretary Stimson and Harvey Bundy, but I also told them and General Marshall that I would consider it a serious mistake to postpone any feasible military operation in the expectation that the bomb would be ready as a substitute at some later date.
There has been much discussion since the war about the decision to use the atomic bomb against Japan. Decisions of this nature must always be made by only one man, and, in this case, the burden fell upon President Truman. Under the terms of the Quebec Agreement, the concurrence of Prime Minister Churchill was necessary; nevertheless, the initial decision and the primary responsibility were Mr. Truman’s. As far as I was concerned, his decision was one of noninterference—basically, a decision not to upset the existing plans.
When we first began to develop atomic energy, the United States was in no way committed to employ atomic weapons against any other power. With the activation of the Manhattan Project, however, the situation began to change. Our work was extremely costly, both in money and in its interference with the rest of the war effort. As time went on, and as we poured more and more money and effort into the project, the government became increasingly committed to the ultimate use of the bomb, and while it has often been said that we undertook development of this terrible weapon so that Hitler would not get it first, the fact remains that the original decision to make the project an all-out effort was based upon using it to end the war. As Mr. Stim-son succinctly put it, the Manhattan Project existed “to bring the war to a successful end more quickly than otherwise would be the case and thus to save American lives.”
Certainly, there was no question in my mind, or, as far as I was ever aware, in the mind of either President Roosevelt or President Truman or any other responsible person, but that we were developing a weapon to be employed against the enemies of the United States. The first serious mention of the possibility that the atomic bomb might not be used came after V-E Day, when Under Secretary of War Patterson asked me whether the surrender in Europe might not alter our plans for dropping the bomb on Japan.
I said that I could see no reason why the decision taken by President Roosevelt when he approved the tremendous effort involved in the Manhattan Project should be changed for that reason, since the surrender of Germany had in no way lessened Japan’s activities against the United States. A little later some of the scientists began to express doubts about the desirability of using the bomb against Japan. A number of these men had come to the United States to escape racial persecution under the Hitler regime. To them, Hitler was the supreme enemy and, once he had been destroyed, they apparently found themselves unable to generate the same degree of enthusiasm for destroying Japan’s military power.
At this same time a debate arose about how the bomb should be employed. Should we conduct a demonstration of its power for all the world to see, and then deliver an ultimatum to Japan, or should we use it without warning? It was always difficult for me to understand how anyone could ignore the importance of the effect on the Japanese people and their government of the overwhelming surprise of the bomb. To achieve surprise was one of the reasons we had tried so hard to maintain our security.
President Truman knew of these diverse and conflicting opinions. He must have engaged in some real soul-searching before reaching his final decision. In my opinion, his resolve to continue with the original plan will always stand as an act of unsurpassed courage and wisdom—courage because, for the first time in the history of the United States, the President personally determined the course of a major military strategical and tactical operation for which he could be considered directly responsible; and wisdom because history, if any thought is given to the value of American lives, has conclusively proven that his decision was correct.
At about this time, the spring of 1945, another job was dropped into our laps at the MED. The first inkling I had of this added responsibility came in the course of a conversation with General Marshall. We had been discussing the progress of the work and, having mentioned our anticipated readiness date, I suggested that the time was fast approaching when we should begin to make plans for the bombing operation itself, even though we still had no assurance that the bomb would be effective. I asked him to designate some officer in the Operations Planning Division (OPD) of the General Staff with whom I could get in touch so that planning could be started. After a moment’s hesitation, General Marshall replied: “I don’t like to bring too many people into this matter. Is there any reason why you can’t take this over and do it yourself?” My “No, sir, I will” concluded the conversation, which constituted the only directive that I ever received or needed.
General Marshall’s position on this matter came as a complete surprise to me. I could easily understand, in fact I favored, restricting the knowledge of our work to the smallest possible number of people. I realized too that he might be questioning whether it was wise to bring into the operational planning officers who might not be able to understand the technical problems involved. But I had never imagined that he would want to keep the execution phases of our project entirely apart from the OPD.
I immediately informed Arnold of this development, and together we went over the general problems facing us. Our most pressing job was to select the bomb targets. This would be my responsibility.
For many months before this, target criteria had been discussed over and over again with and among the members of the Military Policy Committee. They were finally established only after thorough discussions with Oppenheimer and his senior advisers at Los Alamos, and particularly with Dr. John von Neumann. Before approving them, I went over them with my deputy, General Farrell, and then with Brigadier General Lauris Norstad,1 to be sure that nothing had been overlooked.
I had set as the governing factor that the targets chosen should be places the bombing of which would most adversely affect the will of the Japanese people to continue the war. Beyond that, they should be military in nature, consisting either of important headquarters or troop concentrations, or centers of production of military equipment and supplies. To enable us to assess accurately the effects of the bomb, the targets should not have been previously damaged by air raids. It was also desirable that the first target be of such size that the damage would be confined within it, so that we could more definitely determine the power of the bomb.
In April, Farrell, Parsons and Major J. H. Derry of my office met in Washington and later at Los Alamos with three senior men2 from the Operations Analysis Group of Arnold’s office, so that the Air Force representatives could become familiar with technical aspects of the bomb. Norstad directed his people to find out all they could about the performance and capabilities of the specially modified B-29; to establish the cruising data for the plane; and to determine the best way to fly the loaded plane at various altitudes.
The next step was to set up a special committee to recommend specific targets. Three of the members came from Arnold’s office: Colonel William P. Fisher, Dr. J. C. Stearns and D. M. Dennison; the others were from the MED: Farrell, von Neumann, R. B. Wilson and William G. Penney, a member of the British team at Los Alamos. This group met for the first time on May 2 in Washington. I opened the meeting by pointing out the importance of the committee’s task, the need for the highest degree of secrecy, and the number of targets I thought we should have (four, initially). I emphasized General Marshall’s opinion that the ports on the west coast of Japan should not be ignored as possible targets, since they were vital to the Japanese communications with the Asiatic mainland.
General Norstad told them that the facilities of the 20th Air Force would be made available to any degree necessary, through Stearns and Fisher, to provide related data, operational analyses, maps and any other data or information that might be needed.
Norstad and I then departed, leaving Farrell in direct charge of the work from that point forward. I was kept constantly informed on the committee’s progress by Farrell and the others from the MED, particularly von Neumann, with whom I frequently discussed the many scientific and technical problems with which we were faced.
Certain fundamental guidelines were established at this first meeting, among them the probable maximum range of the loaded B-29; the need for visual bombing; the general weather conditions desired over the targets; the expected blast and damage; and the need to have three targets available for each attack.
One of the most difficult problems was attempting to estimate, or even guess, the probable explosive force of the weapon. An accurate guess was of vital importance because the closer we came to being right, the more effective the bomb would be. The optimum height of burst was entirely governed by the explosive force. If the altitude of burst we used was below or too high above this optimum, the area of effective damage would be reduced; and it was possible, if it was much too high, that all we would produce would be a spectacular pyrotechnical display which would do virtually no damage at all. We calculated that if the bomb was detonated at 40 per cent below optimum altitude or 14 per cent above, there would be a reduction of 25 per cent in the area of the severe damage.
At first, considerations of possible fallout and of direct radiation definitely favored a burst at a maximum altitude. I had always insisted that casualties resulting from direct radiation and fallout be held to a minimum. After the Alamogordo test, when it became apparent that the burst could be many hundreds of feet above the ground, I became less concerned about radioactive fallout from too low a burst.
For the present, however, since we did not know the size of the explosion, our plans had to be based on conservative detonation heights. For the Little Boy, it was estimated that the explosive force would have a TNT equivalent ranging from 5,000 to 15,000 tons. For this weapon, the corresponding desirable height of detonation would vary then from 1,550 feet to 2,400 feet. For the Fat Man, it was thought that the magnitude of the explosion would range from 700 to 5,000 tons. This would require detonating heights between 700 feet and 1,500 feet. We could only hope that we could tighten up these estimates considerably, particularly for the Fat Man, after testing the implosion bomb. This element of doubt meant that we had to have fuses for four different height settings. By then most of us in the project were thoroughly inured to such uncertainties. Indeed, three days later, on May 14, Oppenheimer informed me that he and von Neumann had concluded after a thorough discussion that the probable explosive power of the Fat Man was still uncertain, and that the views of the Target Committee should be amended accordingly. They estimated that the maximum altitude for which we should be prepared to set fuses was about twice the minimum, and even for the Little Boy the minimum and probable altitude was only two-thirds of the maximum.
It was agreed that visual bombing was so important from the standpoint of hitting the target that we should be prepared to await good weather. This was based on an estimate that there was only a 2 per cent chance that we would have to wait over two weeks.
The committee recommended that we have spotter aircraft over each of the three alternative targets so that the final target could be selected in the last hour of flight. In case the delivery plane should reach the target and find visual bombing to be impossible, they thought it should return to its base with the bomb. The drop should be made with radar only if the plane could not otherwise return. Radar and navigational developments should be followed closely so that these conclusions could be altered, if desirable.
We all recognized that the plan to use visual bombing, with the possible long delays entailed, required that the bomb be so designed that it could be held for at least three weeks in a state of readiness that would permit its being dispatched on twelve hours’ notice. This was not considered too great a problem at the time, and later we could ignore it as it became obvious that we would normally have at least forty-eight hours’ notice of possible suitable weather.
It was generally agreed that if a plane in good condition had to return to base with the bomb, it would probably be able to make a normal landing. Frequent practice landings had been made with dummy bombs, some of them filled with high explosives. The committee advised that special training in landing with dummy units should be given to all plane crews who were to carry the bomb.
If the bomb had to be jettisoned, extreme care would have to be exercised. Under no circumstances should it be jettisoned near American-held territory. Prior to actual take-off, definite instructions would have to be furnished the weaponeer to guide him in case of trouble. Careful calculations and the experience gained by the Air Force in England from missions involving bombs as large as two thousand pounds indicated that there was no reason to fear for the safety of the bombing plane if its flight were properly controlled. Discussion of the possible radiological effects did indicate, however, that it would be unwise for any aircraft to be closer than two and a half miles to the burst. To provide protection against blast effects, a distance of five miles was advisable. Also, no plane should be permitted to fly through the radioactive clouds.
At its third meeting, the Target Committee was informed that General Arnold and I had concluded that control over the use of the weapon should reside, for the present, in Washington. This announcement was necessary because some of the Air Force people on the committee had displayed a total lack of comprehension of what was involved. They had assumed that the atomic bomb would be handled like any other new weapon; that when it was ready for combat use it would be turned over to the commander in the field, and though he might be given a list of recommended targets, he would have complete freedom of action in every respect.
I felt, and so did General Arnold, that this was too complicated and all-important a matter to be treated so casually; and, regardless of our own feelings, we doubted whether either Mr. Stimson or General Marshall would ever approve, though until now there had been no discussion of this vital point. I had always assumed that operations in the field would be closely controlled from Washington, probably by General Marshall himself, with Mr. Stimson fully aware of and approving the plans. Naturally I expected that the President also would share in the control, not so much by making original decisions as by approving or disapproving the plans made by the War Department. It was quite evident by now, however, that the operation would not be formally considered and acted upon by either the Joint Chiefs of Staff or the Combined Chiefs. One of the reasons for this was the need to maintain complete security. Equally important, though, was Admiral Leahy’s disbelief in the weapon and its hoped-for effectiveness; this would have made action by the Joint Chiefs quite difficult.
When I had visited him in his office six or seven months before to show him a status report on the project, he told me of his long experience with explosives in the Navy and emphasized his belief that nothing extraordinary would come out of our work. He reminded me that no weapon developed during a war had ever been decisive in that war. Then he went on to say he was sorry that I was involved in the project as it would have been much better for me to have had a different and more usual assignment. His views were undoubtedly colored by his feeling, expressed as early as July, 1944, that Japan’s early surrender was inevitable, and would be brought about by combined Naval and Air Force action. In any case, as he frankly admitted after the war, he never had any confidence in the practicability of the atomic bomb. I want to make it clear, however, that he and the other Joint Chiefs, as well as Field Marshal Wilson of the Combined Chiefs, knew of our general plans before the operation began, even though not informed officially of all the details.
Reinforcing my impression that control would not be passed to the field was General Arnold’s strong desire to retain personal control over the Strategic Air Force, keeping Norstad, its Chief of Staff, in Washington where he could closely supervise his activities. Not until Spaatz was sent over to Guam just prior to the bombing did this situation change appreciably. As far as I know, however, there was never any doubt in the minds of those most concerned that control of the actual bombing would be retained in Washington.
The cities that the Target Committee finally selected, and which I approved without exception, were:
With these selections in hand, I prepared a plan of operations for General Marshall, recommending his approval. This report was in my office when I went to see Secretary Stimson about another matter. In the course of our conversation, he asked me whether I had selected the targets yet. I told him that I had and that my report was ready for submission to General Marshall. I added that I hoped to see the General the next morning.
Mr. Stimson was not satisfied with this reply and said he wanted to see my report. I said that I would rather not show it to him without having first discussed it with General Marshall, since this was a military operational matter. He replied, “This is a question I am settling myself. Marshall is not making that decision.” Then he told me to have the report brought over. I demurred, on the grounds that it would take some time. He said that he had all morning and that I should use his phone to get it over right away.
While we were waiting, he asked me about the targets. When I went over the list for him, he immediately objected to Kyoto and said he would not approve it. When I suggested that he might change his mind after he had read the description of Kyoto and our reasons for considering it to be a desirable target, he replied that he was sure that he would not.
The reason for his objection was that Kyoto was the ancient capital of Japan, a historical city, and one that was of great religious significance to the Japanese. He had visited it when he was Governor General of the Philippines and had been very much impressed by its ancient culture.
I pointed out that it had a population of over a million; that any city of that size in Japan must be involved in a tremendous amount of war work even if there were but few large factories; and that the Japanese economy was to a great extent dependent on small shops, which in time of war turned out tremendous quantities of military items. To reinforce my argument, I read from the description of Kyoto,, included in my report, which had now arrived. I pointed out also that Kyoto included 26,446,000 square feet of plant area that had been identified and 19,496,000 square feet of plant area as yet unidentified. The city’s peacetime industries had all been converted to war purposes and were producing, among other items, machine tools, precision ordnance and aircraft parts, radio fire control and gun direction equipment. The industrial district occupied an area of one by three miles in the total built-up area of two and one-half by four miles.
Mr. Stimson was not satisfied, and without further ado walked over to the door of General Marshall’s office and asked him to come in. Without telling him how he had got the report from me, the Secretary said that he disagreed with my recommendation of Kyoto as a target, and explained why. General Marshall read the target description of each of the four cities, but he did not express too positive an opinion, though he did not disagree with Mr. Stimson. It was my impression that he believed it did not make too much difference either way.
General Marshall never knew how he came to be caught unawares in this matter. He never indicated any displeasure about it to me and I doubt whether he gave it any thought. If he did, he probably guessed what had happened. Personally, I was very ill at ease about it and quite annoyed at the possibility that he might think that I was short-cutting him on what was definitely a subject for his consideration.
After some discussion, during which it was impossible for me discreetly to let General Marshall know how I had been trapped into by-passing him, the Secretary said that he stuck by his decision. In the course of our conversation he gradually developed the view that the decision should be governed by the historical position that the United States would occupy after the war. He felt very strongly that anything that would tend in any way to damage this position would be unfortunate.
On the other hand, I particularly wanted Kyoto as a target because, as I have said, it was large enough in area for us to gain complete knowledge of the effects of an atomic bomb. Hiroshima was not nearly so satisfactory in this respect. I also felt quite strongly, as had all the members of the Target Committee, that Kyoto was one of the most important military targets in Japan. Consequently, I continued on a number of occasions afterward to urge its inclusion, but Mr. Stimson was adamant. Even after he arrived in Potsdam, Harrison sent him a cable saying that I still felt it should be used as a target. The return cable stated that he still disapproved, and the next day he followed it with another which said that he had discussed the matter with President Truman, who concurred in his decision. There was no further talk about Kyoto after that.
Nothing is more illustrative of the relationship between Secretary Stimson and me than this episode. Never once did he express the slightest displeasure or annoyance over my repeated recommendations that Kyoto be returned to the list of targets. Nor did I ever feel that he wanted me to remain silent, once I had learned his views, on a matter of such great importance. I believe the affair was also typical of his attitude toward other senior officers.
Events have certainly borne out the wisdom of Mr. Stimson’s decision. I think, however, he did not foresee that much of the criticism he so scrupulously sought to avoid would come from American citizens; certainly he never mentioned this possibility to me. After the sudden ending of the war I was very glad that I had been overruled and that, through Mr. Stimson’s wisdom, the number of Japanese casualties had been greatly reduced.
When our target cities were first selected, an order was sent to the Army Air Force in Guam not to bomb them without special authority from the War Department. About six weeks or so after Mr. Stimson refused to approve Kyoto, I suddenly realized there was a danger that the Air Force might remove it from the list of proscribed cities. I spoke to Arnold, who promptly saw to it that it remained on the reserved list and that the Air Force Command on Okinawa was also notified. If we had not recommended Kyoto as an atomic target, it would not of course have been reserved and would most likely have been seriously damaged, if not destroyed, before the war ended.
1 Chief of Staff, Army Strategic Air Force.
2 F. R. Collbohm, W. T. Dickinson and D. M. Dennison.