Pearl Harbor Betrayed

by Michael Gannon

  Short had, he thought, good reason for expecting local attempts to inflict damage on Army and Navy assets. Thirty-seven percent (about 161,000 persons) of the population of Oahu was of Japanese descent. Of those about 40,000 were aliens. Many were suspected (unfairly as it turned out) of disloyalty. Furthermore, Short had to worry about 239 registered Japanese consular agents, as well as 70 or 80 more Japanese who were thought to be informants to the Japanese consulate. With a sizable number of Japanese living in close proximity to airfields, Short regarded his aircraft as particularly vulnerable. Dispersed in scattered bunkers, their fuel tanks, gas-filled engines, and ammunition could easily be torched by fifth column saboteurs sneaking across the unfenced airport perimeters. (Funds for fencing airfields had finally been allocated by the War Department, but not in time to have the barriers erected by November–December.) Short was mindful that an Army inspector’s report, dated 9 July, had pointed up the vulnerability of Hickam Field, adjacent to Pearl Harbor, to “a few bold, ruthless and intelligent saboteurs.”14

  Rear Admiral Bloch, who had responsibility for defense of Navy assets in the yard proper, already had armed naval personnel patrolling the two tank farms, upper and lower. The upper farm, adjacent to a public highway, had an “unclimable” fence; it was also protected by three elevated sentry stations equipped with searchlights. Marines alternated with the Army in maintaining constant guard over the water supply and electric power lines. It was thought impossible, by Bloch, to achieve absolute security of the Pearl yard without disruption of its work. Though an elaborate system of photograph passes and random surprise searches was in effect, there was no guarantee that a saboteur might not successfully penetrate the yard among the 5,000 employees of civil contractors, the 5,000 civil service employees, and the several thousand naval personnel who entered the yard each day.

  * * *

  Short stated after the war that he did not believe the 27 November war warning message (No. 472) had been authored by Marshall, though his name was affixed to it. He was correct in that, since, as shown above, Marshall was in North Carolina observing maneuvers when the message was drafted by Gerow, Stimson, and Stark. Marshall had worked with Gerow the day before on language for a warning to MacArthur in the Philippines, but it was not he who inserted the caution “not to alarm civilian population.” That, it appears, was Stimson’s doing. On his own, Gerow might have included the words “needed measures for protection against subversive activities should be taken immediately,” but he was persuaded by the deputy chief of staff (Major General William Bryden), Miles, and Colonel Charles W. Bundy, chief of the War Plans Group, to send that particular warning as a separate G-2 dispatch: that would be Miles’s No. 473 on the twenty-seventh.15

  What tipped Short off to the absence of Marshall’s hand was the directive to undertake reconnaissance. Marshall would have known that, by agreement with the Navy on 31 March, aerial reconnaissance of the approaches to Oahu was the responsibility of the Navy. Army aircraft flew regular inshore patrols, but distant (six hundred miles out) air patrols were a Navy function, by signed agreement well known to, and signed off on, by Marshall. Short stated in his testimony before the JCC that, “This message was written basically for General MacArthur in the Philippines and then adopted to the rest of us, and in the Philippines they had no such agreement. The Army was responsible for reconnaissance and they got together with the Navy and agreed upon what sectors that each would cover.”16 (It is not recorded that Short knew this in November 1941.) In response to that part of the directive, Short did order Aircraft Warning Service radar reconnaissance daily from 0400 to 0700—the most dangerous three hours (the two and a half hours preceding dawn and one half hour after) for a dawn air attack, as specified in the Martin-Bellinger estimate. He placed the Army’s Interceptor Command and Information Center on the same schedule. In addition, radar would operate daily, except Sundays, from 0700 to 1100 and from 1200 to 1600, except Saturdays and Sundays, for training and maintenance work. (The Japanese aircraft would make their approach after the predicted 0400–0700 window, and on a Sunday.) At this time, and through December, Short had only six mobile radar stations; three planned fixed stations were not yet operative, their booms and antennas still on a pier in Oakland, California.

  That Short did not think himself obligated to go beyond the No. 1 Alert was owed, he said later, to two known facts about Marshall: first, at the time Short was appointed to the command in Hawaii, Marshall “had definitely indicated his intention to direct personally any genuine prewar alert.”17 Second, the one previous Marshall-authored alert to Hawaii had been so overt in language and intent that no one could mistake its meaning. On 17 June 1940, when there was information in Washington suggesting that Japan might make a military move against the United States, but hardly the amount of intelligence and threat that existed in November 1941, Marshall wired an alert to the then commander of the Hawaiian Department, Major General Charles D. Herron. The wire may stand as a model of what a war alert should look like:

  JUNE 17, 1940.

  NO. 428

  IMMEDIATELY ALERT COMPLETE DEFENSIVE ORGANIZATION TO DEAL WITH POSSIBLE TRANS-PACIFIC RAID, TO GREATEST EXTENT POSSIBLE WITHOUT CREATING PUBLIC HYSTERIA OR PROVOKING UNDUE CURIOSITY OF NEWSPAPERS OR ALIEN AGENTS. SUGGEST MANEUVER BASIS. MAINTAIN ALERT UNTIL FURTHER ORDERS. INSTRUCTIONS FOR SECRET COMMUNICATION DIRECT WITH CHIEF OF STAFF WILL BE FURNISHED YOU SHORTLY. ACKNOWLEDGE.18

  Herron at once placed the department on the only alert he had in his procedures—total. Within the hour he conferred with his opposite number, Bloch and with Vice Admiral Adolphus Andrews, commander of the Hawaiian Detachment. There was no corresponding Navy Department warning. The same day, Herron wired the War Department an acknowledgment as unambiguous as Marshall’s warning:

  ALL ANTI-AIRCRAFT OBSERVATION [posts manned] AND DETACHMENTS IN POSITION WITH LIVE AMMUNITION AND ORDERS TO FIRE ON FOREIGN PLANES OVER RESTRICTED AREAS AND IN DEFENSE OF ANY ESSENTIAL INSTALLATIONS. SOME LOCAL INTEREST IN AMMUNITION ISSUES BUT NO EXCITEMENT. NAVY INSHORE AND OFFSHORE AIR PATROLS IN OPERATION.19

  The all-out Marshall-Herron alert remained in force until 20 June, when Marshall ordered it tapered down; it ended for all practical purposes on 16 July. Herron assured Marshall that the alert did not “dull the keen edge, or exhaust morale.”20 When Herron was relieved by Short in February 1941, Herron briefed his successor on the 1940 alert, and, according to Short, “acquainted me with the relation which had existed between himself and General Marshall during the all-out alert which began June 17, 1940. In that alert, General Marshall had directed the alert and had closely supervised its continuance.”21 Short argued before the JCC, in January 1945, that he had taken the Marshall-Herron alert as a “precedent” of what he might expect to receive if Marshall ever wanted him to go on all-out alert. He expected that the Chief of Staff, who had personally supervised the thirty-day-long 1940 alert, “would certainly have the time and interest not only to read and understand my succinct report … but to send further word in the event that he disagreed in any way with the measure I had taken in obedience to his November 27 directive.”22 But no further communication was received from Marshall, corrective or otherwise.

  Like the November 27 directive, the Marshall-Herron alert was a “do-don’t” message: do take the following measures; don’t excite the local population. But the do parts of the 1940 warning were explicit and unambiguous, starting with the verb alert. By comparison with the plain Army language of Marshall in June 1940, the November 1941 message, No. 472, was awkward in expression, abstract in tone, and unsure in intent. It did not seem to come from the same hand (which it did not). And it contained no warning whatever of a possible sea-air attack—“trans-Pacific raid”—on Hawaii. In fact, Short observed, among the military intelligence estimates prepared by G-2 at the War Department and sent to his department, “in no estimate did G-2 ever indicate the probability of an attack on Hawaii.”23 And no Magic intercepts had been sent to him. There were ot
her problems, Short contended: if the War Department had wanted him to upgrade his alert to No. 2, which read, “defense against sabotage and uprisings and, in addition, defense against an air attack or against an attack by surface and subsurface vessels,”24 it would not be possible to obey the injunction to “limit dissemination [of the warning] to minimum essential officers.” If No. 2 were declared, he would have to inform officers all up and down the line—air, antiaircraft, coast artillery, and infantry. As it was, under Alert No. 1 Short informed only his staff, General Martin of the Army Air Forces, Major General Henry T. Burgin, commander of the coast artillery, and the two infantry division commanders. Furthermore, if Alert No. 2 or No. 3 had been ordered, there would have been no way to keep the warning secret, not to mention refrain from alarming the locals:

  All of the coast artillery, all of the anti-aircraft artillery, and all of the air would have immediately taken up their duties as described in that alert. Part of the coast artillery was right in the middle of the town. Fort de Russy was within two or three blocks of the Royal Hawaiian Hotel. The public couldn’t help seeing that they were manning their seacoast guns. Placing live ammunition. Some of the guns were practically in the middle of the park.25

  Use of Alerts No. 2 or No. 3 would also have interfered with his training mission, Short argued. In large part, his command was made up of newly commissioned officers and recent enlistees and inductees. They required constant training in order to achieve battle readiness. If they were taken out of training to man alert posts, they would not be properly prepared to carry out their specialized tasks. “The War Department message had not indicated in any way,” Short testified to the JCC, “that our training mission was modified, suspended, or abolished, or that all troops were to go immediately into tactical status.”26 The same problem would face the Army Air Forces, which had the mission of training bomber combat crews and of ferrying B-17s to the Philippines. General Martin was then making maximum use of his few bombers and more numerous pursuit (fighter) planes for combat training. “If war were momentarily expected in the Hawaiian coastal frontier,” Short said, “these considerations would give way. But every indication was that the War Department expected the war to break out, if at all, only in the far Pacific and not at Hawaii.”27 No warning of a possible Japanese attack on Hawaii had been received at Fort Shafter since the Marshall-Herron alert of June 1940. In the Army Pearl Harbor Board investigation, Marshall confirmed Short’s reading of those facts:

  We anticipated, beyond a doubt, a Japanese movement in Indochina 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; the reason being that we thought, with the addition of more modern planes, that the defense there would be sufficient to make it extremely hazardous for the Japanese to attempt such an attack.28

  Yet, as Short was to learn shortly after exactly that attack on Hawaii materialized on 7 December, Generals Marshall and Gerow, as well as Colonel Bundy, were all the while confident that Short had at the least ordered Alert No. 2 against air raids. Marshall particularly was amazed and acrid that Short had failed to go on all-out alert, when the commanders in the Philippines, the Canal Zone, and at San Francisco, who had received basically the same message, did go on full alert. Short’s basic defense was, Why, when I sent my acknowledgment stating the alert I had ordered, was I not told that it was insufficient? As he stated in the JCC, “The War Department had nine days in which to tell me that my action was not what they wanted. I accepted their silence as a full agreement with the action taken.”29 (On the twenty-ninth Short sent the War Department a detailed report on the precautions he had taken against subversion and sabotage. That report, specifically acknowledging the sabotage warning, No. 473, was not answered either.)

  We can track the course of Short’s first acknowledgment (No. 959) through the corridors of the Munitions Building, from the time, 5:57 A.M. on the twenty-eighth, when it arrived in the WD code room. Specifically marked as a reply to WD’s No. 472, the dispatch was clipped together with a dispatch from General MacArthur (who was responding to the parallel warning sent him the day before), which had arrived in the code room about an hour before. MacArthur had declared an all-out alert that included extensive air reconnaissance. In the way the two dispatches were clipped, the MacArthur message was on top. Their first distribution was in Marshall’s office, where the chief wrote on the top document (MacArthur’s), “To Secretary of War GCM.” He did not initial either the top or the under document at the space marked “Chief of Staff.” The routing slip shows that the two messages left Marshall’s office at noon on the twenty-eighth and made their way to Stimson’s office, where the secretary initialed each of them, “Noted—H.L.S.” The messages were then walked to the War Plans Division, where General Gerow initialed both, “L.T.G.” While there, both messages were read by Colonel Bundy, head of the Plans Group, and by Colonel Charles K. Gailey, executive officer of the division. Gailey initialed the routing slip that had been attached in Marshall’s office, and the messages passed thenceforth into the War Department files. Not one of the officers (or the secretary) through whose hands Short’s reply passed found any reason to correct it by transmission of a more explicit warning. In the JCC hearings, Marshall was asked,

  Would this be true from an Army viewpoint, that when an overseas commander is ordered to take measures as he deems necessary and to report measures taken to you, is he correct in assuming that if his report is not the kind of action that you had in mind that you would thereafter inform him specifically of the difference?

  General Marshall: I would assume so.30

  In the Army Board investigation of 1944, General Gerow accepted the blame for not having recognized the insufficiency of the reply from Fort Shafter. He agreed that, as War Department rules read, “The merit of a report is not measured by its length. A concise presentation of important points usually is all that is required.” And to the question asked him, “Would General Short’s reply comply with that regulation?” he answered, “Yes, sir,” although he suggested that, if the message had read, “alerted against sabotage only,” he might have caught its import more clearly.31 Marshall refused to let Gerow accept the blame for not picking up on the inadequacy of Short’s report. “He [Gerow] had a direct responsibility and I had the full responsibility.”32 Marshall went on to express the breakdown in this language: “It did not register on Colonel Bundy, it did not register on General Gerow, it did not register on me and it carries Mr. Stimson’s initials also.”33 Marshall’s biographer Forrest C. Pogue has commented that Marshall was attempting to do so many things—“hold too many threads of operations in his hand”—that he simply missed noticing the kind of signal Short was sending and missed as well his own opportunity to intervene. Inability to take each order in turn and to give it exactly the attention it deserved was to Pogue “a reasonable defense from an overworked Chief of Staff, but it did not exonerate him.”34 Marshall testified that the officer expressly charged with checking on compliance with orders sent was Bundy. About an hour and a half after the Pearl Harbor attack, Marshall had a conversation with Bundy. The officer stated to the chief that he had gotten the impression from the words “liaison with Navy” that Short and Bloch had activated the Joint (Coastal Frontier) Defense Plan.35 Which they had not.

  Confusion now hath made his masterpiece!

  Macbeth, Act II, Sc. 3

  Bundy was killed in an airplane crash while en route to Hawaii shortly after 7 December.

  * * *

  On the morning of the twenty-seventh, just prior to receiving his war warning from Washington, Short met for three hours with Kimmel at the latter’s headquarters. With Kimmel were Admirals Bloch, Halsey, Brown, and Bellinger, and certain staff members, including “Soc” McMorris. Short had brought along Major General Martin and his chief of staff, Lieutenant Colonel James. A. Mollison. At conferences such as this one Short had learned that, accordi
ng to the best naval intelligence, the Japanese carriers were either in home ports or were proceeding south. That knowledge would give him some confidence later in the day in declaring the alert against sabotage only. He had to have been reassured, too, when, in the course of this particular conference, Colonel Mollison having observed that the Japanese had the capability of making an air attack on Oahu, Kimmel turned to his fleet war plans officer and, according to Short, asked, “What do you think about the prospects of a Japanese air attack?” and McMorris answered, “None.”36 (Ten days later, Short, Kimmel, and, of course, McMorris would rue that fateful response.) Even the principal item on that morning’s agenda lent itself to easing disquiet: the War and Navy Departments had authorized the reduction of Short’s P-40 fighter strength on Oahu by about half, the balance to be sent to Wake and Midway Islands. The fact that Marshall and Stark felt comfortable in making that recommendation persuaded those around the conference table that Washington “did not consider hostile action on Pearl Harbor imminent or probable.”37