Chapter 22: 22. Army Pearl Harbor Board

The Army Pearl Harbor Board (APHB) was authorized by Congress “to Ascertain and Report the Facts Relating to the Attack Made by Japanese Armed Forces upon the Territory of Hawaii on 7 December 1941, and to Make such Recommendations as It May Deem Proper.”^[1] Lieutenant General George Grunert was appointed president of this “Grunert Board.” Two other Army generals also served—Major General Henry D. Russell and Major General Walter H. Frank. Colonels Charles W. West and Harry A. Toulmin had nonvoting positions as recorder and executive officer respectively, and Major Henry C. Clausen was assistant recorder.

The Grunert Board convened in Washington, D.C. on July 20, 1944. Being an Army board, it was primarily interested in Army’s role, especially that of Army Chief of Staff Marshall and Hawaiian Commander General Walter Short. The Board’s members first reviewed the reports of earlier investigations and studied materials supplied by various government agencies and Congressional committees. It also wrote Secretary of War Stimson and Army Chief of Staff George C. Marshall, listing the subjects the Board hoped to cover when they testified.^[2] The APHB did not have the power of subpoena, but “in no instance [was] its invitation to appear and testify . . . ignored.”^[3] The questioning alternated among the members. The board interviewed 151 witnesses and was in continuous session until October 20, 1944.^[4] Because of the nature of the revelations, much of the testimony taken during the final segment of the proceedings was kept off the record and preserved in a separate TOP SECRET report.

Army Chief of Staff General Marshall: US.-Japan Relations in 1941, Increasingly Tense

Marshall, the Board’s first witness, stated:

[W]e were very fearful of some warlike act by the Japanese, which immediately would have brought about a state of war in the Pacific, for which, at the time, we were not prepared. . . . [T]here were numerous indications . . . all of which indicated a very serious crisis developing in the Pacific in relation to Japan.^[5]

Marshall said he and Chief of Naval Operations Admiral Stark

made it very clear . . . to the Secretary of State, that it was of the utmost importance . . . to delay so long as possible any outbreak in the Pacific. . . . We anticipated, beyond a doubt, a Japanese movement in Indo-China and the Gulf of Siam, and against the Malay Peninsula. We anticipated also an assault on the Philippines. We did not, so far as I can recall, anticipate an attack on Hawaii. . . . [W]e thought, with the addition of more modern planes, that the defenses there would be sufficient to make it extremely hazardous for the Japanese to attempt such an attack.^[6]

In a joint November 27 memorandum Marshall and Stark told the president emphatically:

The most essential thing now, from the United States viewpoint, is to gain time. . . . After consultation with each other, United States, British, and Dutch military authorities in the Far East agreed that joint military counter action against Japan should be undertaken only in case Japan attacks or directly threatens the territory or mandated territory of the United States, the British Commonwealth, or the Netherlands East Indies, or should the Japanese move forces into Thailand west of 100° East or south of 10° North—Portuguese Timor, New Caledonia, or the Loyalty Islands.^[7]

Very soon after Short assumed command of the Hawaiian Department, Marshall advised him (February 7, 1941) of his responsibility for protecting the fleet:

The fullest protection for the Fleet is the rather than a major consideration for us, there can be little question about that. . . . Please keep clearly in mind in all of your negotiations that our mission is to protect the base and the Naval concentration, and that purpose should be made clearly apparent to Admiral Kimmel.^[8]

When asked if “the mission of the Army out there was the protection of the Navy,” Marshall answered, “Yes. That is the reason for the Army’s being there.”^[9]

Cooperation Between Pearl Harbor Commanders Admiral Kimmel and General Short

The Roberts Commission had blamed the Pearl Harbor disaster to some extent on the failure of the two Pearl Harbor commanders, Kimmel and Short, to cooperate. Short acknowledged that there had been some instances when the channels of communication between the Army and Navy seemed to break down. For instance, he did not learn until December 8 about the submarine sunk near Pearl Harbor at about 6:45 a.m. on December 7.^[10] However, generally speaking he thought the Army’s relation with the Navy in Hawaii and his personal relationship with Kimmel had been good. Kimmel’s associates and Hawaiian locals who were questioned agreed.^[11]

General Short Defends His Sabotage Alert

Short had been charged by the Roberts Commission with an error in judgment for having instituted Alert #1 to guard against sabotage and for not having alerted for such an attack as that of December 7. The Army’s July 14, 1941, Standard Operating Procedure,^[12] effective November 5, 1941, had described three alerts. So Marshall was familiar with them. However, he had some definite ideas about implementing them. He did not want the Hawaiian air force used to defend against sabotage and ground attacks, or to provide military police duty. He wrote Short on October 10 that using the air force for anti-sabotage “seems inconsistent with the emphasis we are placing on air strength in Hawaii.”^[13] Marshall told Short to use his Air Force for its normal purposes and not upon antisabotage guard duty.^[14]

War Department’s November 27 “war warning” message #472 had read:

Negotiations with Japan appear to be terminated. . . . If hostilities cannot repeat cannot be avoided the United States desires that Japan commit the first overt act. This policy should not comma repeat not comma be construed as restricting you to a course of action that might jeopardize your defense. Prior to hostile Japanese action you are directed to undertake such reconnaissance and other measures as you deem necessary but these measures should be carried out so as not comma repeat not comma to alarm civil population or disclose intent.^[15]

Short found this confusing; he should “undertake such reconnaissance and other measures as you deem necessary,” but these measures should be carried out “so as not comma repeat not comma to alarm civil population or disclose intent.” Because of this stricture, Short had decided upon Alert #1, designed specifically to guard against sabotage, espionage, and subversive activities, rather than one of the more aggressive Alerts.^[16] Short had then radioed Washington, as requested, that he had “alerted to prevent sabotage.”^[17]

Short explained still further the reasons for his sabotage alert.^[18] Hawaiian Intelligence (G-2) had received a message—#473 from General Sherman Miles, Director of Military Intelligence, G-2—advising that “Subversive activities may be expected.”^[19] Miles explained that President Roosevelt had delegated to the FBI, ONI, and to his military intelligence division responsibility for counter-subversive activities. When Miles found that nothing had been said about subversion in General Marshall’s November 27 war warning, he felt it necessary to warn all G-2 departments. “The policy had already been laid down,” Miles said, “by General Marshall’s telegram;” he “was simply backing up” Marshall’s policy and emphasizing the form of attack for which he, Miles, was responsible.^[20] Upon receiving the War Department’s November 27 “war warning,”^[21] Short thought, “from the caution about not taking any provocative measures against Japan and not alarming the public,” there was still some possibility of avoiding war with Japan.

To take the message of the 16th of October [re Japan’s change in Prime Ministers from Konoye to Tojo] and the 27th of November together, they indicated to me [Short] that they were still hopeful of avoiding hostilities.

Stark didn’t hear anything further from the War Department except the short November 28 message (#482), which went into detail about sabotage,^[22] telling Short to protect his establishments against “subversive propaganda and . . . espionage.”^[23] Short interpreted the several messages from the War Department in Washington as approving his sabotage alert. Other witnesses questioned by the Grunert board also believed Short had been justified in his decision to alert for sabotage. Short had been given no indication that the negotiations in Washington were reaching a breaking point; he had not been told that we were “negotiating with the British and Dutch about coordinated military action in the Pacific area.” If he had known more about what was going on in Washington and about the attitude of Washington officials, it would undoubtedly have made him “more conscious that war was practically unavoidable.”^[24]

Washington Officials See War as Imminent

In their testimony, various Washington officials traced the deterioration of U.S.-Japanese relations back to various points in time. General H.H. Arnold, chief of the Army Air Forces, said it had been apparent as early as January 1941 that relations were “quite strained.”^[25] General Leonard T. Gerow, acting, or assistant, chief of war plans, said the “general buildup . . . between July and November . . . led to the conclusion in November that war with Japan might occur.”^[26] Marshall said it had been “a gradual process”; he had come to the conclusion “some time in the fall of ‘41 that war with Japan was inevitable.”^[27] General Miles of Military Intelligence also saw the situation as precarious from November 27,

when we learned that we had practically given what . . . probably would be considered by them [the Japanese] an ultimatum . . . I considered war as very probable if not inevitable. . . . I thought that very definitely an action by Japan, a pretty radical action, would be taken almost at once; that that necessarily would be an overt and open attack on the United States.

However, Miles pointed out, war was not the only possibility; “there were a good many things Japan could have done, if she did break those negotiations, short of open war with the United States, and we were considering all of those matters.”^[28]

What Did Short Know of the Growing U.S.-Japanese Crisis Buildup?

Not much! Short believed he knew “in an indefinite way” that U.S. policy from sometime in August or September of 1941 was largely one of delaying, playing for time, with the realization that war with Japan was inevitable.^[29] But he had not been told about the September 1941 conference when “General Marshall and others who were in conference with the Secretary of State had decided that war with Japan was inevitable.” Nor had he known that “we were negotiating with the British and Dutch about coordinated military action in the Pacific area.” And no one had told him

an agreement had been reached with all nations, the effect of which was that if the Japanese moved forces into Thailand west of 100 degrees east or south of 20 degrees north we would regard that as an act of war.^[30]

Basically, he knew only what was in the papers.

Short learned from an October 16 Stark-Kimmel message, of the resignation of Japan’s Prime Minister Konoye and the rise to power of the more militant General Hideki Tojo creating

a grave situation. . . . [H]ostilities between Japan and Russia are a strong possibility. Since the US and Britain are held responsible for her present desperate situation there is also a possibility that Japan may attack these two powers.^[31]

All concerned with “the existing grave situation” were “to take due precautions.” Short believed he had done that with his Alert #1: “We had had all the utilities guarded, all the bridges, and . . . I just simply cautioned people that were responsible for that guarding to be unusually careful.” Short’s interpretation was that the Navy Department “felt sure” Japan was going to attack Russia; an attack on the U.S. and G.B. was “only a possibility.”^[32]

After discussion with the Army’s G-2, Army Chief of War Plans Gerow “reached the conclusion at that time that the Navy estimate was more pessimistic than we believed it should be.”^[33]Accordingly the War Department sent Short an October 20 follow-up radiogram in effect toning down the Navy’s warning:

Following War Dept. estimate of Japanese situation for your information STOP Tension between United States and Japan remains strained but no repeat no abrupt change in Japanese foreign policy appears imminent.^[34]

On November 24 the Navy sent another pessimistic message to its field commanders, including Kimmel:

Chances of favorable outcome of negotiations with Japan very doubtful x This situation coupled with statements of Japanese Government and movements their naval and military forces indicate to our opinion that a surprise aggressive movement in any direction including attack on Philippines or Guam is a possibility.

The commanders were advised: “Chief of Staff has seen this dispatch concurs and requests action adees [addressees] to inform senior Army their areas.”^[35] Short did not recall seeing this pessimistic Navy message.^[36]

Marshall defended Washington’s warnings as adequate to have alerted Short to be prepared for the crisis that was coming. “In our own view,” Marshall testified,

an alert of the character, particularly the character of the two that occurred at that time, the Naval alert and then the later Army alert, were sufficient for any commander with a great responsibility.^[37]

Short thought that, if Washington ever really believed that an attack on the United States was imminent, it would have found some means to inform him, as commander in the field, if necessary by scrambler phone. “[O]rdinarily, you could get through in ten or fifteen minutes. It was reasonable to believe,” Short testified, “that if there was going to be a hostile attack, they would have tried to get it to us by more than one means of communication.” Thus he had been forced to conclude that “there was a feeling still at that time that secrecy was more important than the time element.”^[38] But there had been no word from Washington. Under Grunert’s questioning, Marshall admitted that it would have been both “possible and feasible to have sent the substance of this secret information to the Commanding Generals of the Overseas Departments by courier or otherwise.” However, Marshall had been so sensitive to the threat of endangering Washington’s source of intelligence—the MAGIC intercepts—that he had considered it “unwise” to do so.^[39]

Planes and Ships for Hawaii Not a High Priority

The board questioned Admiral Bloch about the Navy’s effort to obtain planes. In 1940 the Navy had gotten money for a 15,000-plane program—a number of PBYs, of which about 108 were allocated to the Fourteenth Naval District (Hawaii) and 150 or so to the U.S. Pacific Fleet. At that time the fleet only had 81. Bloch said he was “quite persistent in . . . trying to get” the planes.

Commander-in-Chief [Kimmel] knew . . . and he supported me. The correspondence went to the Navy Department asking for these planes, and I was told repeatedly they would be given to me but they would not be given to me until sometime that was indefinite in the future.^[40]. . . [I]t wasn’t a question of appropriation. It was question of priorities: The war was in [the] Atlantic; Pacific wasn’t in the war. . . . They say it in the war plan: The war is in the Atlantic; the Pacific is a more or less quasi-defensive [theater] until they get around to it.^[41]

According to Admiral William S. Pye, commander battle force, Pacific Fleet, and commander Task Force 1, the situation was said to be serious “as early really as April 1941.” However, he reminded the board, the Navy Department had detached from the Pacific Fleet in June

one battleship division, one light cruiser division, and two destroyer divisions, to send into the Atlantic. It hardly seemed to the Commanders in the Pacific that if the situation was as bad as it was said to be, that was the time to be moving a large portion of our Fleet into the Atlantic,

especially as “the British Fleet, itself, was many times superior to the available German ships.”^[42]

Not only had the Pacific Fleet been gutted, Pye said, but the commanders encountered resistance from Washington whenever they asked for men and materiel.

During this same period, it became most difficult for the Commander-in-Chief of the Pacific Fleet to obtain patrol planes or even to obtain carrier planes, and, up to December 7, not even all of the carriers were equipped with their normal number of planes. . . . [B]y acts rather than [by] words . . .

Washington failed to indicate urgency. This led to “the almost uniform opinion that while war probably was in the offing,” it was not expected to come without warning. At least the necessary steps to prepare for a surprise attack were not being taken. The impression given was that if war came, it would be “upon the initiative of the United States.”^[43]

Navy Alert to Submarine Attack Threat

Grunert questioned Pye about reconnaissance in Hawaii, the areas in which the fleet operated and patrolled with the available aircraft.

Grunert: Would it have been reasonable to assume . . . that the enemy could not well approach with aircraft carriers to make an attack on the mainland?

Pye: . . . [I]t should be recalled that we were not in a state of war . . . . [T]he patrol was primarily to determine the possible presence of submarines. . . . If attacks had been made by submarines, and the submarine not sighted or sunk or captured, there would have been no way for us to prove definitely that it was not an internal explosion in the ship rather than a torpedo. In addition to that there was always the possibility that German crews might man Japanese submarines or might, in the last analysis, even bring their submarines to the Hawaiian Islands in order to try to force us into war. . . . [T]he implication [of the November 27 “war warning”] was that there was great danger of a submarine attack.

Grunert: Then it would appear from what testimony we have had to date that the Army was sabotage-minded and the Navy may have been submarine-minded.

Pye: I think there is no question but what the Navy was submarine-minded.^[44]

Pearl Harbor Attack Surprised Washington Officials as well as Hawaiian Commanders

The principal task of the U.S. embassy in Japan, particularly of its military and naval officers, was to obtain information concerning probable action on the part of the Japanese Army and Navy.^[45] Yet in the months before the attack the embassy officials in Japan had found this to be increasingly difficult. Ambassador Grew cabled from Tokyo on November 17, 1941: “The Embassy’s field of naval or military observation is restricted almost literally to what could be seen with the naked eye, and this is negligible.”^[46]

According to General Miles, Chief of Army’s G-2, Military Intelligence, many of our sources of intelligence had dried up. To have avoided being surprised on the morning of December 7, would have called for knowing about Japan’s naval bases, staging areas, and within rather fairly narrow limits, the expected time of the attack and the direction of approach. He testified:

I did not think any Intelligence officer ever thought that he could be sure of picking up a convoy or attack force or task force in Japan before it sailed and know where it was going. . . . It would have been almost a military intelligence miracle.^[47]

Pye did “not believe the people in Washington expected the attack any more than the people in Honolulu.”^[48] He “thought the attitude of the officers of the Fleet was just about the same as the attitude of the War and Navy Departments.” Pye, who met Secretary Knox right upon his arrival in the Hawaiian Islands about December 10, said

the first thing Knox said to me was, “No one in Washington expected such an attack—[not] even Kelly Turner.” Admiral Kelly Turner was in the War Plans Division, was the most aggressive-minded of all.^[49]

Marshall testified that he had sent Major General Arnold, commanding general, Army Air Force, and deputy chief of staff, to California specifically to expedite the departure of the B-17 bombers to the Philippines.^[50] Arnold in turn testified that in view of the strained relations with Japan, we were doing “what we could to prepare for any eventuality that might occur, without causing an overt act against the Japanese.”^[51] We had not been “so much worried about the immediate attack on Hawaii. It was always a possibility; but we all thought there certainly would be an attack against Midway and Wake.”^[52] The B-17s left the West Coast for Hawaii without ammunition

because at that time it was a question of gasoline or ammunition for that long 2400-mile hop. . . . So they did not take the ammunition, and they got there right in the middle of the Pearl Harbor attack.^[53]

“Obviously,” Arnold said, “we made an error, an error in judgment. Somebody had to weigh the fact against their certainty of arriving there by providing sufficient gasoline against the probability of their using their machine guns and not getting there” because they were carrying ammunition.

The fact that bombers had been dispatched to Honolulu, unarmed, en route to the Philippines on the night of December 6–7, told Short “that the War Department felt that there was no danger of an air attack on Honolulu, or between Honolulu and San Francisco.” The extra weight in ammunition was considered “a greater hazard . . . than it was to take a chance of meeting the Japs without any ability to return their fire.”^[54] “[V]ery definitely,” Short said, “their estimate was exactly the same as [his], they were not expecting an air attack on Honolulu.”^[55]

Philippines Considered the Most Likely Target

Relations between the United States and Japan were deteriorating in October and November 1941. Washington was expecting a Japanese strike somewhere in the western Pacific or southeast Asia. It seemed logical that she would attack the Philippines to keep the United States from intercepting Japanese ships and planes bound for southeast Asia. Thus, the War Department had been trying desperately to build up U.S. defenses there. According to Marshall, “we were pouring through Hawaii, on the way to the Philippines, convoys [with men and materiel for the Philippines], rushing everybody. Everything was being pushed to the last extreme.”^[56] “[F]rom the information that we were receiving,” Marshall felt “that they [the Japanese] were now getting in a highly nervous state because of the arrival of supplies in the Philippines.” One MAGIC message had asked the Japanese Consul General in Manila “to check up immediately on the presence of Flying Fortresses in the Philippines.” The Japanese consul in Manila was also reporting “the tremendous unloading procedures being carried out at night and the movement of things at night from the docks, and everybody barred from the vicinity.” Marshall concluded that the Japanese were in “a critical posture as to what they must do to prevent us from building up further in the Philippines.”^[57]

“Our own belief,” Marshall said,

was that, once we got the planes out there, and particularly these convoys that were then on the Pacific, which had, compared to what the Philippines already had, a wealth of material . . . the Japanese would be in an extremely delicate strategical position in trying to carry out any enterprise to the south of the Philippines.^[58]

Marshall realized that the shipments being rushed out to the Philippines must be alarming Japan. “Nobody could look at that [buildup],” he told the board,

without realizing that something very critical was in the wind. Our great problem was how to do these things . . . the shipments, and collecting the means and getting them out, particularly to the Philippines, which passed entirely through Hawaii—without giving such notice to the Japanese that it would have an unfortunate effect in our stalling off this affair.^[59]

The Joint Board of the Army and Navy conference on November 3 had urged postponing hostilities as long as possible. The November 5 Marshall-Stark memorandum to FDR had recommended that we not issue an ultimatum that might provoke Japan to attack.^[60] Yet, on November 26, Hull had handed the Japanese ambassadors the U.S. “ultimatum” he knew the Japanese government would not accept.

Washington’s eyes appeared to be glued on the Philippines. Throughout this time, relatively little thought was given to Hawaii. Further confirmation of Washington’s neglect of Pearl Harbor and its concentration on the Philippines came when Marshall appeared before the board on September 29, 1944. He was asked specifically about the “One p.m. Message” of December 7 and his radiogram to Hawaii which had left Washington shortly after noon that day but was delayed in transit and failed to reach Pearl Harbor until after the attack. Grunert asked Marshall’s “[r]easons for not using the telephone to inform General Short of the information contained in the Chief of Staff’s radiogram of 7 December 1941.”^[61] Marshall said that if he had used the scrambler phone to relay that message he “would certainly have called MacArthur [in the Philippines] first, and then I would have called Panama Canal second.” He had thought “we were open in a more vulnerable way in the Panama Canal, than we were in Hawaii.”^[62]

The messages sent Short in Hawaii had been terse and rather cryptic, advising him that

hostilities between Japan and Russia are a strong possibility. Since the U.S. and Britain are held responsible by Japan for her present desperate situation there is also a possibility that Japan may attack these two powers.^[63]

Short had been led to believe, by the urgency of the shipments passing through Hawaii to the Philippines, that Washington must have had definite reasons for believing that the Philippine Islands were the U.S. territory most seriously threatened by Japanese attack.

Kimmel Tells the APHB About Important Intelligence Not Sent Pearl Harbor Commanders

On Friday, August 25, several days after Kimmel testified before the Navy Court of Inquiry, he was called to the witness stand by the APHB which was going on concurrently. He was asked the usual questions about his relationship with Short and other matters pertaining to conditions before the attack.^[64] Kimmel then discussed intelligence. He said he “got information from the Navy Department . . . so far as the efficiency of the Japanese Air Force [was] concerned.” But his sources had then been limited for he could not “send people to the Mandated Islands to discover what the Japanese were doing” there. His

orders were not to go anywhere near them. . . . We wanted to go into the Gilberts to make some surveys down there . . . and the answer was that we should not evince any interest in the Gilberts, because the Japs might find out that we were interested.^[65]

In any event, Kimmel said, “A movement such as that would have had to be approved by the Navy Department.” Kimmel was “convinced that no reconnaissance of the Mandates would have been permitted by the Navy Department at that time.”^[66] He had a statement to make about “the information which was supplied to the two responsible commanders in Hawaii.” He and Short had “thoroughly considered all such information” and had taken “the action which we deemed appropriate. There was no disagreement between the Army and Navy and none between me and my personal advisers.”^[67]

However, Kimmel said,

Since Pearl Harbor information has come to my knowledge that vital information in the hands of the War and Navy Departments was not supplied to responsible officers in Hawaii; in particular, that the War and Navy Departments knew that Japan had set a deadline of 25 November, later extended to 29 November for the signing of an agreement, after which they would take hostile steps against the United States; that on 26 November an ultimatum was delivered to Japan by the United States. This was done notwithstanding a joint recommendation to the president by General Marshall and Admiral Stark that no ultimatum of any kind should be made to Japan.

Kimmel said he “had been advised of this recommendation and had received no qualification of that information.” Moreover, he said he had “had no knowledge of the delivery of the ultimatum to Japan on 26 November, 1941.”^[68]

Further, Kimmel said, he was

certain that several days prior to 7 December, 1941, there was information in the War Department and the Navy Department that Japan would attack the United States and, very probably, that the attack would be directed against the fleet at Pearl Harbor, among other places; that there was information in the War and Navy Departments on 6 December, 1941, that the hour of attack was momentarily imminent, and that early on 7 December, 1941, the precise time of the attack was known. It was known at least three or probably four hours before the attack.^[69]

All this information, Kimmel said, “was denied” to him and to Short, yet he felt they were entitled to it. He had believed that if the War and Navy departments had had such information, they would surely have furnished it to them.

Had we not been denied this, many things would have been different. Had we been furnished this information as little as two or three hours before the attack, which was easily feasible and possible,

Kimmel said, “much could have been done.”^[70]

When Kimmel finished his statement, Grunert said, “Some of the things to which you have referred may become the subject of further investigation before the Board is through” and he asked if Kimmel would provide his source. Kimmel agreed to “cooperate to the best of my ability, in conformity with the restrictions which [had] been imposed upon me.”^[71]

The APHB members could not then pursue Kimmel’s leads as they were flying to San Francisco and Pearl Harbor to question other witnesses. Kimmel’s testimony raised new questions. Grunert wrote Marshall another letter saying he wanted to ask Marshall about information “brought to the attention of the Board, which it did not have when you testified” before.^[72]

APHB Flies to San Francisco and Hawaii

En route to and from Hawaii,^[73] the Board stopped in San Francisco to investigate charges of shoddy construction carried out for the Army by the Hawaiian Constructors.^[74] In Hawaii, the board asked Army officers and local businessmen about Short’s preparations for the islands’ defense. Generally speaking, they approved of Short’s defense preparations. And except for a few Japanese connected with the consulate, American businessmen did not question the loyalty of most ethnic Japanese.^[75]

The APHB members left Hawaii for Washington on September 13^[76] and resumed their hearings on September 26.

Only when they were back in Washington were the Grunert Board members able to follow up on Kimmel’s revelations. And the board did not actually obtain copies of the documents on which they were based until October 6, when the board was winding up its hearings. In view of the sensitivity of this material, and to safeguard the confidentiality of witnesses, their remarks were not included in the regular printed hearings but placed in a separate TOP SECRET supplement.^[77]

Ambassador Grew Describes the Tokyo Situation

Joseph C. Grew had been U.S. ambassador to Japan from June 14, 1932, until December 7. Once the two nations were at war, he was placed under house arrest and held until an exchange of diplomats could be arranged. Grew was questioned about U.S.-Japanese relations in general. “[T]he trend of our relations during . . . the years 1940 and 1941 was almost steadily down-hill.” Grew thought that in the embassy they had done everything possible

to arrest that trend. . . . But we were up against what I would call a “tidal wave” of military extremism in Japan. . . . [N]ot being a defeatist by nature, I was unwilling to admit that war was inevitable, up to the last minute.^[78]

Grew had warned Washington that economic embargoes should not be imposed

until we were prepared to go all the way through with whatever might result from those embargoes. . . . [O]ur relations with that country were bound to go steadily down-hill and it might, and probably would, end in war.^[79]

And he reminded the board, there were “not only the embargoes, but also the freezing order and the denunciation of our treaty and commerce with Japan.”^[80]

Grew thought the attack must have been a surprise also to the civil authorities in Japan. It was “perfectly possible,” he said, “that the Cabinet was not informed of the plans for attacking Pearl Harbor.”^[81] He had had a conversation with Foreign Minister Togo at half past midnight on December 7—about three hours before the attack—and was “convinced from the nature of that conversation that Mr. Togo did not at that moment know that Pearl Harbor was about to break.” Grew added:

That does not for a moment mean that they were not informed of the likelihood that under certain circumstances war might occur. Of course, they knew that, without any shadow of doubt, and Nomura and Kurusu knew that, too. I was referring purely to the attack on Pearl Harbor, itself.^[82]

General Short Raises More Questions

After Short appeared before the APHB, he asked to be furnished the testimony of other witnesses and the board had agreed.^[83] When he appeared again on September 29,^[84] he said he was concerned about the criticism levied against the Hawaiian commanders because the attack had taken them by surprise. They were not the only ones surprised; he was convinced the attack had been a surprise to Washington officials also. On the theory that “actions speak louder than words,” he pointed to Washington’s pre-attack actions. He reminded the board of Washington’s

constant denials of requests for increases in personnel, for money for the improvement of defenses; and things like sending out planes [from California] the night before the attack without ammunition—all kinds of things, that really were stronger in their effect than mere words.

Assuming “that they were acting in good faith,” Short continued, “you have to arrive at the conclusion that they undoubtedly were not contemplating an air attack on Honolulu.”^[85]

The Army had also been considered negligent because its radar was not operational and had not warned of the impending attack that morning. That was not due to the command’s negligence. Rather, Short said, it was due to “a shortage of supplies for the radar, such as vacuum tubes, and so forth.” In an October 1941 memorandum radar equipment had been requested adequate to operate 24 hours daily, but it had been radically cut back to allow only two hours of operation per day. Why? Because, according to the War Department, “the United States was not threatened with attack.”^[86]

Short raised three important questions:

(1) Had the APHB learned anything about the coded messages several witnesses had mentioned? For instance, what was the basis for Justice Roberts’s question concerning

A Japanese code message . . . intercepted and . . . broken down by the Department in Washington . . . which gave certain key words which would be flashed over the radio directing the attack on Pearl Harbor?^[87]

(2) What new intelligence had prompted Marshall to send the message concerning the Japanese delivery time to the field commanders at noon on December 7?^[88]

(3) What were the grounds for Kimmel’s statement to the board about information available in Washington during the crucial days before the attack, information not furnished Short and Kimmel?^[89]

Short was anxious to learn about the coded messages. He had written Stimson that very day “asking that a search be made . . . and that, if it [the information] is not to be found in the War Department files, that a demand be made on the Navy for the information. . . .”^[90] Short felt that “all pertinent evidence” should be made available to the board and to him, because the War Department had not permitted him to have a representative attend the APHB hearings and examine witnesses.^[91]However, Short’s path to securing this information wasn’t easy. When he wrote Secretary of War Stimson later (September 29, 1944) asking for permission to see the SECRET documents,^[92] Stimson agreed that Short’s military counsel, Brigadier General Thomas H. Green, should have access to this material (October 2, 1944).^[93] Grunert tried to countermand that permission,^[94] but the War Department refused. Grunert was told to “comply with the instructions of the Secretary of War as issued”^[95] and to allow Green to see the TOP SECRET material. But Short was not allowed access. Green was even asked to sign a letter “swear[ing] that you have been appropriately warned relating to the military security concerning these matters.”^[96]

Marshall Remembers Some Events, Does Not Recall Others

When Marshall appeared before the board again, he was asked about the questions in Grunert’s letter of August 31, 1944.^[97] Marshall was asked about the Japanese-imposed deadlines—November 25 at first, then November 29. “[T]he first date of the 25th of November . . . puzzled us greatly,” Marshall said.

[T]he only thing that we could think of at the moment was . . . that on that day the anti-Comintern pact expired. . . . During all this period the Japanese had been involved in actions in the China theater and towards Indo-China, which indicated . . . that they were either about to embark on a war in the Malaysia area, at least, or were in the process of carrying out very dire infiltration operations. . . . However, we later received information from our secret sources . . . that the date had been extended to the 29th of November. That, in our view, wiped out any thought that the original date of the 25th of November pertained to the anti-Comintern pact. . . . November 29th arrived and passed, and we entered into December without anything happening other than the continuation of these movements, which we could follow fairly well, down the China coast and Indo-China and headed quite plainly towards Thailand and the Gulf of Siam.

[I]n all the past procedures of the Japanese, they had taken very bold measures . . . on the assumption, I presume—that they could get away with them without the United States entering into war. Their feeling, so nearly as we could determine, was one that the United States would not participate in a war and they could take advantage of that by doing things that otherwise would immediately provoke a state of war.^[98]

Throughout November, Marshall and Stark were urging the administration to postpone any confrontation with the Japanese until they could build up their Philippine defenses. The British, preoccupied at home with their struggle against Germany, were overextended and wanted to avoid open conflict with the Japanese. Yet the embargo on oil to Japan, with the cooperation of the British and Dutch, imperiled the Japanese. Also the reopening of the Burma Road by the British made it easier for China to be provisioned, and this was a thorn in the side of the Japanese. Marshall said he believed that the Japanese

were going ahead to get in as strong a position as possible, on the assumption that the reluctance of the United States and the reluctance of the British Government in its dilemma of the moment would permit them to establish themselves.^[99]

He thought

the Japanese were capitalizing on the belief that it would be very difficult to bring our people into a willingness to enter the war. That, incidentally, was somewhat confirmed by the governmental policy on our part of making certain that the overt act should not be attributed to the United States, because of the state of public mind at that time. Of course, no one anticipated that that overt act would be the crippling of the Pacific Fleet.

Marshall believed “That the Japanese were going to take every conceivable advantage and finally would reach the point where they could safely declare war, involve us in war, and get all the other things they were after.”^[100]

By the fall of 1941, Marshall said, he had come to the conclusion that war with Japan was inevitable. “Prudence dictated that warnings be sent by the War Department to those officers responsible for the defense of all our areas within reach of Japanese action.” However, information available in the War Department led him to believe that any Japanese attack would take place in the vicinity of the Malay Peninsula and the Philippines.^[101] He wasn’t particularly concerned about Hawaii, especially as he considered it better supplied and better prepared to defend itself than other U.S. outposts.^[102]

APHB board member General Russell questioned Marshall on various points Kimmel had raised. Marshall admitted to having no recollection of the several advanced warnings received in Washington. He did not recall having learned that November 29, Tokyo time, “was definitely the governing date for offensive military operations of some nature.” He had no recollection of any messages on November 26, December 1, and December 4 giving “specific evidence of Japan’s intention to wage an offensive war against both Britain and the United States.”^[103]

Marshall recalled something about a “Winds Code” setup and the alerting of our code clerks to listen for the crucial words. But, he said,

Colonel Bratton was unable to find that a—our records do not show that a Japanese message using the “Winds” code was intercepted by the F.C.C. or the Army Signal Corps until after Pearl Harbor.

However, he admitted that it did appear “from the record,” that “a Japanese message using the ‘Winds’ code had been intercepted.” This indicated, Marshall said, that “Japanese-Great Britain relations were to be broken.” Marshall didn’t mention a possible break in Japanese-U.S. relations.^[104] Russell noted this was the Army interpretation. He said,

The Navy people say that the executive order [the “Winds Execute”] whenever it came in—and they alleged it came in on the 3rd [sic] of December . . . meant that war was coming with the United States and with Britain, but not with Russia.^[105]

Marshall had understood from Bratton that the intelligence officers in Hawaii were privy to the information about the “Winds Code” message.^[106]

Another important matter which Marshall did not remember related to the first 13 parts of the Japanese reply to our “ultimatum.” According to Kimmel’s statement, they were received during the evening of December 6. Marshall said was he was unaware of this.^[107]

Responding to Grunert’s question as to when on December 7 Marshall had learned “the precise time of the attack,” he reviewed his December 7 morning movements—his early horseback ride; his arrival at his War Department office about 11 a.m.; his meeting with Miles, Gerow, and Bratton; his discovery of the long 14-part Japanese message; and, finally, the “One p.m. Message.” “Something was going to happen at 1:00, it was quite evident to us.” After digesting all this material, Marshall drafted the message that went out to the field commanders at noon.^[108]

According to Marshall, he had come to the conclusion about November 1 that war with Japan was inevitable. Concerning the TOP SECRET information known to top Washington officials, he admitted that to have sent this intelligence to the commanding generals of the overseas departments “by courier or otherwise, thereby avoiding the danger of exposing the codes that I was striving so diligently to protect,” would have been “both practical and feasible.” On the morning of December 7, when confronted with incontestable evidence that Japan was planning some definite action that very day at 1:00 p.m. Washington time, Marshall said it would have been possible to notify the commanders by a more rapid method than the coded radio message actually dispatched at noon that day. But he felt then that that would have been “unwise.”^[109] Moreover, Marshall was convinced that Short had been sent “sufficient information” and that he had “adequate weapons, ammunition, and other means for the discharge of his defensive mission in the protection of the Island of Oahu.”^[110]

Marshall apologized for not being better prepared to answer the board’s questions, but as chief of staff he had been “busy with the war with Japan and Germany” and had not been able to keep up on the Pearl Harbor situation.^[111]

There were still more questions the board wanted to pursue with Marshall, who pleaded pressure of other business—appointments with a Chinese official, the combined Chiefs of Staff, the U.S. Chiefs of Staff, the ambassador going to France. And “confidentially” he was leaving for France himself the next Tuesday.^[112] Grunert said if it “appears necessary,” the Board might ask Marshall for “another hour early next week.”^[113] The next day, Saturday, September 30, the board sent Marshall still more questions.^[114] In the meantime, two important witnesses testified—Army courier Colonel Rufus Bratton and Navy Captain L.F. Safford.

Colonel Bratton Testifies About Japanese Intercepts

At the time of the attack, Colonel Rufus S. Bratton had been Chief, Far Eastern Section G-2 and Army Courier. When testifying before the APHB,^[115] he referred to a memorandum written December 10 detailing the events of December 7,^[116] and to a “Summary of Far Eastern Documents,” based on documents from 1937–1941 and compiled August 1943 by the Far Eastern section, intelligence group, and by War Department’s G-2. It had been compiled for submission to the Army chief of staff and the president. Bratton was “conversant with the Japanese language.”^[117] Although he had not been involved with the actual interception and decoding of the Japanese messages on which that Summary had been based—that had been the responsibility of the Army’s Signal Intelligence Service (S.I.S.) and the Navy’s Office of Naval Intelligence (O.N.I.)—he had seen “all secret messages relating to the Japanese situation received by the War Department.”^[118] In August 1940, U.S, cryptographers had succeeded in deciphering the Japanese diplomatic code, which became known as “purple,” and ever since then we had been reading many, if not most, of the Japanese intercepts transmitted in this code. The intelligence derived from this source was considered so valuable that it had been code-named MAGIC. Much of the information Kimmel referred to, that had not been furnished the Pearl Harbor commanders, had come from MAGIC.

According to Bratton, tight security maintained in the distribution of the Japanese intercepts.

As to the intercepts, and translations of Japanese intercepts, they were handled in a special way. . . . In 1941, certainly in the latter part of it, I was the custodian and the disseminator of this type of intelligence. . . . The translations, made either in the Signal Corps SIS section or in the corresponding section of Naval Communications were sent to me in sextuplet, six copies of each one. Out of the mass of material . . . say 10 to 20 percent was of intelligence value; the remainder dealt with administrative or personal matters . . . and that material I destroyed by burning. The remaining “flimsies” containing military intelligence of value to our Government officials was arranged in cardboard folders which in turn were placed in locked dispatch cases, one for the Secretary of State. one for the Secretary of War, one for the Chief of Staff, one for the AC of S, G-2, and one for the AC of S, WPD.

I delivered these pouches in person to the officers concerned, who had keys to the pouches. . . . I collected all of these pouches on my next visit, or on my next round the following day, and destroyed the contents of them by burning and retaining in my file a complete copy of everything that had been seen by all of these officials.^[119]

The critical messages sent by G-2, in November and early December, 1941: “The translated intercepts, you refer to, are on file in G-2, War Department.”^[120]

“During this period,” Bratton said,

the president, the Secretary of State, the Secretary of War, the Chief, the AC of S G-2, and the Chief of the War Plans Division all saw the same material, they all read the same translations, as fast as I could get them to them.^[121]

These materials

did not go out to the field. . . . We felt considerably hampered in G-2 by two restrictions that were placed upon us. The first I have mentioned as the policy which prevented us from giving out intelligence to G-2s in tactical units or in overseas department, which might have the effect of bringing about operational results. The other restriction was imposed on us by the Navy, who refused to allow us to send any of this intercept intelligence out to any of our people in the field over Army net, using any Army code cipher . . . fear of the Japanese breaking our Army code, and finding out that we were reading their own. It was a security measure.^[122]

Bratton testified that on the morning of December 7, between 8:30 and 9:00 a.m., he had received the short Japanese English-language intercept “relating to the destruction of the code machines and the delivery of the ultimatum.” He realized this was “about the most important message” he had received during this period and immediately phoned Marshall’s quarters.^[123] Bratton was told Marshall “had gone horseback riding.” He “requested his [Marshall’s] orderly to go out and find him at once and ask him to call . . . as soon as practicable, as [he] had an important message to deliver to him.” In spite of the urgency of the message, it was not until “sometime between ten and eleven” that Marshall returned Bratton’s call. Bratton then told Marshall that he “had a message of extreme importance which he should see at once” and offered to bring it out to Marshall’s quarters. Marshall told him to report to him in his office.^[124]

Marshall arrived in his office at “about 11:25.” Then finally, almost three hours after this “message of extreme importance” had been received, Bratton was able to show it to Marshall. Marshall discussed it with Bratton, Miles, and Gerow, who were present. They “thought it probable that the Japanese line of action would be into Thailand but that it might be into any one or more of a number of other areas.”^[125] Marshall then radioed the Army field commanders “by the fastest possible safe means, giving the Philippines first priority,”^[126] advising them of the 1 p.m. deadline, and telling them to “be on the alert accordingly.”^[127] This was the message that reached Pearl Harbor in the afternoon, hours after the attack had ended.

The APHB wanted to know whether or not the “Winds Code” setup had ever been “executed,” i.e., implemented. Bratton had known that the FCC had been looking for such a message and he remembered talking about weather messages with Colonel Sadtler, Navy Lieutenant Kramer and Navy Commander McCollum. But his memory was vague. He did remember talking with Sadtler on the morning of December 5, who said something about a message that had come through indicating a break between Japan and Great Britain. Bratton did not remember seeing before the attack an implementation, a “Winds Code” Execute, with reference to a Japanese-United States break.^[128] However, he did remember, vaguely, seeing a “Winds Execute” referring to a Japanese-U.S. break in relations after the attack.^[129]

APHB members Russell and Grunert then asked Bratton how it would affect the Japanese if they learned then—1944—that we had intercepted a “Winds Execute” message in 1941. He did not believe those code words were being used by the Japanese today. He was then asked if the Japanese knew we had intercepted these messages and had broken that code before the war, would it give them any information as to whether or not we had broken the code they are using today.

Bratton: Oh, yes, sir, it would, because these code phrases are a code within a cipher. . . . The whole message about this “Winds” signal was in a very secret cipher. . . .

Grunert: And they are continung to use that cipher? . . . Therefore, the danger that any leak of this thing might affect the war effort exists now as it has in the past?

Bratton: Yes, sir.^[130]

On this point, Navy Captain Safford flatly disagreed, as the APHB soon learned.

Captain L.F. Safford Describes the Japanese “Winds Code” Intercepts

As head of the communications security division, naval operations, in 1941, Safford had been much involved with naval intelligence information.^[131] He remembered many details from 1941. However, when testifying before the APHB on October 2, he consulted a record of the intercepts prepared more recently (November–December 1943 and January–March 1944) from original sources borrowed from OP20G, i.e., “the communication intelligence section, or communication division, of Naval Operations.”^[132] Russell asked Safford about his statement before Hart’s investigation that we had received on November 26 “specific evidence of Japan’s intentions to wage an offensive war against both Britain and the United States.” Safford replied that this message, S.I.S. No. 25392,

said that Japan would announce her intentions in regard to war or possibly breaking off diplomatic relations with Russia, England including the Netherlands East Indies, and the United States by means of a word sent five times in the middle and at the end of their information broadcast.

Safford continued, “On November 28th, 1941, we read another message . . . giving a ‘Winds’ code to be used in their voice broadcasts.”^[133] We had verification of this “Winds Code” setup from other sources—Hart (Manila), Singapore, Batavia (NEI), and our intercept station in the state of Washington.^[134]

Russell: [T]ell us about the follow-up on this code . . . whether or not on or about December 4th you did receive information which indicated that the Japanese Empire had employed this code and the intercepted messages indicated final decisions affecting the United States, Russia, Britain; one or more of these powers.

Safford: Yes, sir, we did. That was received in the morning of Thursday, December 4, 1941. It was received about 8:00. . . by teletype. I saw it when I first came to the office . . . the writing at the bottom in lead pencil in Kramer’s handwriting, “War with England, war with America, peace with Russia.” The message as received was not the way we expected it, because they had mixed up their voice procedure with the Morse code message.^[135]

Distribution of this “Winds Execute” was made, not only in accordance with the special arrangements set up by Noyes of naval communications, but also in the usual fashion, through the war and navy departments.

And also I know that in the Navy Department that copy was distributed around noon, in connection with the daily routine distribution of translations, and that went to the Chief of Naval Operations [Stark], Assistant Chief of Naval Operations [Ingersoll], Director of Naval Communications [Noyes], Director of Naval Intelligence [Wilkinson], and the Director of War Plans Division [Turner], also went to the State Department and to the White House.^[136]

Safford was positive as to the date when the “Winds Execute” came in because its receipt had prompted him to send four messages that very day, between 3:00 p.m. and 3:19 p.m., to the naval attachés at Tokyo, Peiping, Bangkok, and Shanghai, directing them to destroy “all secret and confidential files except those essential for current purposes and all other papers which in the hands of an enemy would be a disadvantage to the United States.”^[137]

Safford believed that all the Army S.I.S. messages he had been describing were in the custody of the Army’s G-2, general staff, and that the same messages, filed by their Navy numbers, were at 20G, the Navy’s communication annex—except for the implementation of the “Winds Code.” “Unfortunately, we cannot find any written record of the [‘Winds Execute’] message,” in spite of having “looked now for more than six months.”^[138] And there was no way to trace it because “all the station logs unfortunately had been destroyed sometime during ‘43,” which was “[m]ore or less” Standard Operating Procedure when a government office moved or expanded.^[139]

More important in Grunert’s view than the “Winds Execute” itself was whether the war effort in 1944 would suffer if the Japanese learned that we had intercepted and deciphered this message in 1941. Russell described the dilemma:

[T]he Board is debating the effect on the war effort of a public disclosure of the contents of the “Winds” message. Assuming that the Japanese Empire knew that the American Government was in possession of those facts which are contained in that “Winds” message, would it . . . cause them to make changes which would make it more difficult for us to obtain Japanese information now?

When questioned, Bratton had said it would. Safford disagreed.

No, sir, not the “Winds” message or this other so-called hidden-word or stop-code message. The setup for those two was sent in what they call a low-grade cipher held by all their Consuls. Everybody was solving that. The Dutch solved it, the British solved it in Singapore, and we solved it ourselves—both of them—and they must know that we have been reading those messages, and I believe that that particular system is not in use any more, anyhow. It is not the high-class machine which is a literal gold mine at the present time. The other stuff it would be very bad to let public.^[140]

By “this other stuff,” Safford meant “the declaration of war,” i.e., the 14-part reply to our “ultimatum” that the Japanese sent December 6–7, and their December 7 “One p.m. Message.” That, Safford said, “is in their [high-class, “purple”] machine which they think no one can read, and they are still talking their fool heads off in it, particularly from Germany.”^[141]

Even though the “Winds Execute” had been sent in the “low-grade” cipher, Safford said it was extremely significant. By announcing the imminence of a break in relations, or of an outbreak of war, with the United States and Britain, Japan was explaining the reason for her November 25 deadline, later changed to November 29. And the deadline showed that the break in relations it portended was not just talk. Thus the “Winds Execute” “made the deadline message mean a lot more, and the deadline message made that [the Winds Code Execute] mean a lot more.”^[142]

Safford told the APHB that we knew from Japanese intercepts picked up December 1 and 3, 1941 that Japanese embassies and consulates in London, Hong Kong, Singapore, Manila, Batavia, and Washington had been told to destroy their codes, ciphers, and code machines. “[T]his destruction of codes immediately threw the ‘Winds’ message into prominence,” Safford said.

Before, we couldn’t understand why they had this [“Winds Code”] setup arranged. It seemed a foolish thing to do . . . but they had this in mind, I think: Well, all right, one step short of war. They are destroying their codes to play safe, but they are still reserving the decision as to peace or war to come in the “Winds” message, which was the reason that, from the first on, we thought the “Winds” message was so highly important, and yet that information did not get out to either Commander-in-Chief Pacific Fleet or Commander-in-Chief Asiatic Fleet until 48 hours after we had the news.^[143]

Safford had recognized the significance of the “Winds Code Execute” at the time. However, his responsibilities were limited to communications security. “The evaluation [of a message] was out of my hands, and that is a function of Naval Intelligence.”^[144] Thus, Safford had focused his December 4 messages to the naval attachés necessarily on the issue of security.

Grunert then turned to another subject. Safford had “indicated . . . that at sometime in the not too distant past it was not intended to give the Navy Court of Inquiry and the Army Board certain secret information.” Had special instructions been issued to that effect?

That was “rather a long story,” Safford said. He explained that Kimmel had asked to see the Hart report. On the basis of information revealed there, Kimmel had requested permission for his counsel, Captain Lavender, “to inspect all the files out at 20G, communications intelligence files, to see what information had been in existence in the Navy Department.” Lavender had been permitted to see the files and had then asked for copies of about 60 messages. The department had assembled the intercepts, turned them over to the director of naval communications, and notified S.I.S. S.I.S. had protested. The assistant secretary of the Navy, Ralph A. Bard, had also disapproved of their release. But when Navy Secretary Forrestal, then in London, returned to Washington, he reversed that decision and directed that the intercepts be made available to the NCI. But the APHB had not obtained copies. Safford suggested they put in a request to the secretary of the Navy.^[145]

Marshall Defends Government’s Policy of Secrecy; Doesn’t Recall Important Documents

On Monday, October 2, 1944, Marshall returned once more to the APHB to answer questions posed in the board’s September 30 letter. He explained that it was considered essential at all cost to prevent the enemy from learning that Japanese intercepts were our source of secret information. War and Navy department policy concerning “secret, ultrasecret information” directed that “No action is to be taken on information herein reported, regardless of temporary advantage, if such action might have the effect of revealing the existence of the source to the enemy.” According to Marshall,

there have been cases where convoys have been permitted to go into the most serious situations rather than diverting them from the assemblage of the so-called wolf packs because of the fear that that would convey to the Germans that we had some means of knowing just how this was managed.

Apparently, Marshall continued, the Japanese thought we were obtaining knowledge of these convoy movements from spies and observation posts. So long as they did, we felt free to go ahead.

[B]ut if there is any danger of our giving away our sources, then we would have to hold off somewhat on seizing each opportunity, for fear we would lose tremendous long-term advantages.^[146]

Marshall believed he had been “kept fully informed by the State Department on the development of the relations between the Japanese Empire and the American Government.” Even so, he couldn’t remember some important events. He didn’t recall the department’s November 26, 1941, memorandum, or “ultimatum,” rejecting Japan’s proposal for a modus vivendi. Nor did he remember Secretary of State Hull’s remark to Secretary of War Stimson the next morning to the effect that Hull had broken off discussions with the Japanese: “I have washed my hands of it and it is now in the hands of you [Stimson] and Knox, the Army and Navy.”^[147] Marshall admitted, however, that he

must have known on the 26th of November that the negotiations were nearing an impasse, because Admiral Stark and I evidently directed the preparation of a draft of the 27th of November warning on that day, the 26th.^[148]

Marshall didn’t remember the War Department’s November 27 warning to Short (#472) advising that “Negotiations with Japan appear to be terminated” and asking Short to “report measures taken.”^[149] Nor did he remember Short’s “sabotage alert” reply (#959). And he had not realized that his failure to respond to Short’s sabotage alert, admittedly inappropriate for defense against attack, meant that it had “obtained during the entire period 27 November–6 December inclusive.” Marshall was forced to admit that Washington’s November 27 warning “did not accomplish the desired results”^[150] of defending Pearl Harbor against attack.

Perhaps most astonishing of all, however, was that Marshall still maintained that he had heard nothing at all prior to the morning of December 7 about the Japanese reply to the U.S. November 26 “ultimatum.” This in spite of the fact that the first 13 parts of that reply had been delivered to the White House and the State Department and were “in the hands of some agency of the War Department” during the evening of December 6. Marshall believed that it was

not . . . until I was before the Navy Court here recently that I knew this had come in, had been made available to the Secretary of State, the larger portion of that message, the night before.^[151]

Russell pointed out to Marshall:

The evidence which is before the Board at this time is to the effect that as early as 8:30, possibly not later than 9:00 a.m., on the morning of December 7th, the message which indicated that the ultimatum would be delivered by the Japanese Ambassadors at 1:00 on that day, and that the code machines were being destroyed, was in the hands of a Colonel. . . . Bratton of G-2.

Bratton’s energies from the time he received that message “were devoted exclusively” to trying to locate Marshall and Miles.^[152] Yet Marshall couldn’t be reached for a couple of hours, not until he finally arrived in his office at about 11:30 a.m. The situation was further complicated by the fact that the Army apparently had no clear plan for handling emergencies when Marshall was not available.^[153] Moreover, this was a message, APHB member Frank commented, “where the time of its delivery by two hours would have made an awful lot of difference.”^[154]

Marshall offered no explanation for his inaccessibility that Sunday morning, except to describe his activities:

I remember very distinctly the message from Colonel Bratton because it came to me as I was coming out of a shower, as my habit was to ride at 8:30 on Sunday morning, and it takes me about fifty minutes to go around the only available loop to ride in. It takes me about eight or ten minutes to get a shower and dress. And when the message came from Colonel Bratton he wanted to come out there, and I said, “No. I am on my way down to the War Department.” And it couldn’t have been more than five or ten minutes at the outside before I had left to come down here. I have a very clear recollection of that because naturally I thought about it at the time. . . . I was not aware of the fact that this message had been available the night before.^[155]

Marshall concluded his testimony with a further comment on the importance of secrecy. “[E]verybody that is concerned with this top secret thing is very cagey about saying anything about it.”^[156] This, he implied, explained the reluctance of the War Department to release TOP SECRET intercepts to those investigating the Pearl Harbor attack.

Army Courier Bratton Reports His Deliveries of Japanese Intercepts

Now that more information about the Japanese intercepts had come out, Bratton returned twice more to testify —October 2 and 6. He said the secret “Summary of Far Eastern Documents,”^[157] copies of the translated Japanese intercepts documenting the “Summary,” together with an Appendix containing many of the crucial Japanese intercepts, were “on file in G-2, War Department,”^[158] being held in tight security and “will be made available to you later.”^[159]

Bratton had been convinced on December 3 that war with Japan was imminent. On that day a December 2 message from Tokyo directing the Japanese embassy in Washington to burn its codes, to stop using their code machine, and to destroy it completely, had been translated.^[160] “After the receipt of this translation,” Bratton said, “any further intercepts that were brought to me would simply contribute toward the climax that I saw coming. This was it.”^[161] He “had a feeling that further warnings or alerts should be sent out to our overseas commands. Gerow felt that sufficient warning had been sent. Miles thought he couldn’t go over Gerow’s decision” because of the War Department policy then in effect that War Department G-2 (Intelligence) should not send out any intelligence to the G-2s of tactical commands or overseas departments “which might produce an operational reaction, without the complete concurrence of the War Plans Division.”^[162]

Bratton “still felt uneasy” and thought “further warnings should be sent out.” He went to the Navy Department to see Commander McCollum, head of the Far Eastern Section in ONI. McCollum felt as Bratton did and was going to write up a warning and “try to get the Chief of Naval Operations to dispatch it.” McCollum told Bratton also that the Navy’s “S.I.S. man in Honolulu, a Commander Rochefort . . . had all the information that we had, and was listening for this Japanese winds-weather broadcast.” McCollum suggested that Bratton instruct Army’s G-2 in Hawaii to talk with Rochefort at once, “as in a short period of time Rochefort could tell Colonel Fielder, our G-2, exactly what was going on and what we knew.” Thus Bratton tried by this indirect route to communicate his fears to Army’s G-2 in Hawaii.^[163]

When delivering the Japanese intercepts, Bratton’s usual practice was to go first to Marshall, Miles, and Gerow, and then to the State Department. He had followed this procedure the evening of December 6, when delivering the first 13 parts of the Japanese reply. Bratton said he “very seldom” delivered the locked pouch of intercepts to Marshall in person. That evening he had left the “locked bag” containing Marshall’s copy with his secretary, Colonel Bedell Smith, advising him “that it was an important document. . . and that the Chief of Staff . . . [s]hould see it right away.”^[164]

Bratton had then made delivery in person to G-2’s Miles, with whom he had discussed the message at some length.^[165] He had left the copy of the message for Gerow with his “executive officer,” Colonel Gailey.^[166] Then, at about 10 or 10:30 Bratton had gone with the 13-part message to the Department of State, where he had delivered the locked pouch “to the watch officer in the State Department, with the request that it be gotten to Mr. Hull immediately.”^[167]

The next morning, after receiving the last installment of the Japanese reply and the 1:00 p.m. deadline message at about 8:30–9:00 a.m., Bratton spent a couple of frantic hours trying to locate Marshall. When he finally reached him by phone, Marshall asked Bratton to wait for him at his office. Marshall arrived at 11:25 a.m.; Bratton was sure of the time because he “kept looking at the clock on the wall and at my watch.” The long Japanese reply was on Marshall’s desk when he came in. Marshall read it and then “a discussion of the entire communication” ensued.^[168]

Bratton urged the APHB to obtain, not only the Japanese reply to the U.S. November 26 “ultimatum,” but also

the 30 or 40 other messages which preceded it; that is, the exchanges between the Ambassador in Washington and the foreign minister in Tokyo. . . . And consider the picture that lay before all of our policy-making and planning officials, from the Secretary of State down through the Secretary of War, to the Chief of the War Plans Division. They all had the same picture; and it was a picture that was being painted over a period of weeks, if not months.^[169]

APHB Obtains Japanese Intercepts – Finally!

On October 6, 1944, the board gained access to the 45 to 50 intercepts requested.^[170]

After it actually had the intercepts in hand, it questioned a few final witnesses and then its proceedings were concluded.

The Army Pearl Harbor Board Report

The APHB’s hearings and those of the NCI had run concurrently, the NCI from July 24 through September 27, 1944, the APHB from August 7 through October 6. The NCI report was dated October 19, that of the APHB, October 20, 1944.^[171]

The APHB issued a detailed report describing the background of the attack, the situation in Washington and in Hawaii, and the responsibilities of the several officials. It had brought out in the course of its hearings several significant points not previously covered in depth:

a. The Army was clearly responsible for the defense of the fleet when it was at its home base in Pearl Harbor;

b. Given the instructions he had received, Short appeared justified in ordering Alert #1 for sabotage;

c. Short’s Washington superior commander, Marshall, was obviously familiar with Short’s system of alerts and should have notified Short if his order for a sabotage alert, issued in response to Washington’s November 27 warning, was not considered adequate;

d. Short had been told very little about the crisis that Washington officialdom knew was looming;

e. The attack on Pearl Harbor apparently took everyone by surprise, not only in Hawaii but also in Washington. Officials both in Washington and Hawaii had expected the first Japanese strike would be in the western Pacific or southeast Asia and, quite likely, the Phillippines.

f. Two witnesses—Kimmel and Safford—revealed that, as a result of decoded Japanese intercepts, Washington officials had had access to considerable intelligence concerning Japanese intentions, which was not furnished the Hawaiian commanders.

The APHB was impressed by the quantity and quality of the intelligence available in Washington. The record shows that from informers and other sources the War Department had complete and detailed information of Japanese intentions. Information of the evident Japanese intention to go to war in the very near future was well known to the Secretary of State, the Secretary of War, the Chief of Staff of the Army, the Secretary of Navy, and the Chief of Naval Operations. It was not a question of fact; it was only a question of time. The next few days would see the end of peace and the beginning of war.

If it be assumed that for any reason the information could not have been given to the Hawaiian Department, then it was a responsibility of the War Department to give orders to Short what to do, and to tell him to go on an all-out alert instead of a sabotage alert.^[172] The board was especially concerned about the warnings sent Short, his sabotage alert response to the November 27 warning, and the failure of Washington to respond.

Having asked for a report of what he was doing, the War Department placed itself in the position of sharing the responsibility if it did not direct Short to take such measures as they considered adequate to meet this serious threat.^[173]

However, the APHB pointed out, Short “had two threats.” Yet “he only took measures as to one.” The message on which he particularly relied as to sabotage came from G-2 on November 28, the report said,

after he had made his decision to go to Alert Number 1. This last message . . . does not in any way change previous messages. Short should have known, as a trained soldier, that a G-2 message is informative and is of lesser authority than a commanding message from the Chief of Staff.^[174]

After the conflicting Navy and Army dispatches of November 27, and the additional November 28 sabotage messages from Army G-2 and from the adjutant general,^[175]

Short had only silence from Washington. He was given no further clarification of this conflict amongst the messages. There is no explanation why Short was not told of the so-called [November 26] ultimatum. It was known to the Japanese because it was handed to them.^[176]

In its report, the APHB discussed the intelligence available in Washington and Hawaii, the “amiable relationship” between Short and Kimmel,^[177] the warnings that had been sent to Hawaii, the Hawaiian commanders’ defense plans, the Army’s radar facilities, Short’s sabotage alert, the shortage of planes in Hawaii for long-range reconnaissance, and so on.

As has been repeated so many times, there was positive evidence in the War Department that it was only a matter of days before war would ensue and the War Department had notice that Hawaii was on only a sabotage alert, inadequate for full warfare. Had a full war message, unadulterated, been dispatched or had direct orders for a full, all-out alert been sent, Hawaii could have been ready to have met the attack with what it had. What resulted was failure at both ends of the line. Responsibility laid both in Washington and in Hawaii.^[178]

Among other things, the APHB report criticized Marshall “for not providing an arrangement by which another could act in so critical a situation when he could not readily be reached.”^[179] No accounting for this was made even though

[t]he evidence indicates that the manner in which authority to act was delegated or not delegated had its influence on this situation. The Chief of Staff had three deputies, Generals Bryden, Arnold, and Moore. None of these three was given the secret information concerning the known Jap intentions. . . . Complete authority to act in General Marshall’s absence does not seem to have been given to any one subordinate. Had there been an officer either with authority or with courage to act on the information that was in the War Department on the evening of December 6, and had he sent a message to Short, Hawaii should have been fully alerted.^[180]

The board report attributed the extent of the Pearl Harbor disaster

primarily to two causes: (1) The failure of the Commanding General [Short] of the Hawaiian Department adequately to alert his command for war, and (2) The failure of the War Department, with knowledge of the type of alert taken by the Commanding General, Hawaiian Department, to direct him to take an adequate alert, and the failure to keep him adequately informed as to the developments of the United States-Japanese negotiations, which in turn might have caused him to change from the inadequate alert to an adequate one.^[181]

The board cited several factors that contributed to the disaster: The failures of (1) The Secretary of State, Cordell Hull, (2) The Chief of Staff of the Army, General George C. Marshall, (3) Chief of War Plans Division, War Department General Staff, Major General Leonard T. Gerow, and (4) Commanding General of the Hawaiian Department, Lieutenant General Walter C. Short.^[182] The APHB report then detailed the extent to which each shared in the responsibility.

The report was submitted to Stimson only a couple of weeks before the November 7 presidential election. The APHB had been critical of Short, who was no longer on active duty. But it had also criticized Secretary of State Hull, Army Chief of Staff Marshall, and General Gerow, all of whom were still actively involved in the administration and the war. The report’s release could prove an embarrassment to the administration, the president, and the war effort.

When Stimson received the report, it was announced in the press that it would not be released until it had been “reviewed for security by appropriate military authorities.”^[183] Then, on December 1, after Roosevelt had won his election to a third term, Stimson announced that

it would be highly prejudicial to the successful prosecution of the war and the safety of American lives to make public during the war the report of the Army Pearl Harbor Board or the record on which it is based.

Thus both secretaries, war and navy, refused “[o]n the ground of national security . . . to make the real story of Pearl Harbor,” as revealed in the NCI and APHB reports “public until the war had ended.”^[184]