The signal crackled through Akagi’s earphones on 2 December.
There had been no diplomatic breakthough in Washington. The operation was go.
Shortly after receipt of the order, Nagumo’s Striking Force crossed the 180th meridian into the Western Hemisphere.
* * *
Among the critical predicates of success in the Hawaii Operation, joining the ability to refuel ships in the rough North Pacific seas and avoidance of premature discovery by the enemy, was the availability of continuous intelligence about the presence of principal U.S. fleet warships in Pearl Harbor. That information was supplied on an almost daily basis through the Japanese consulate in Honolulu by a trained naval intelligence agent named Yoshikawa Takeo. Arriving by ship at Honolulu on 27 March 1941 under the cover of a junior diplomat named “Tadashi Morimura,” the twenty-nine-year-old, English-speaking Yoshikawa took up ostensible duties as chancellor of the consulate. He was rarely in his office, however, with the full blessing of the also newly arrived consul general, Kita Nagao, and his vice consul, Okuda Otojiro, who both knew of and fully supported his mission, which was to reconnoiter all U.S. Army and Navy bases and airfields on Oahu, with special emphasis on Pearl Harbor, where he was to observe and report regularly on ship movements.
Yoshikawa was not the first to undertake this work. The consular treasurer, Seki Kohichi, who had briefly attended Japan’s naval academy at Eta Jima, made regular observations at Pearl Harbor, seven miles distant from the consulate, in 1940 and in the first months of 1941. But Yoshikawa, a graduate of Eta Jima, was the first agent in Honolulu with a professional espionage background. On paper he had a thorough knowledge of the U.S. fleet, and, soon, with the help of a Navy-wise taxi driver named John Yoshige Mikami and a young dual citizenship–holding Nisei with a 1937 Ford named Richard Masayuki Kotoshirodo, Yoshikawa was making regular observations of ship movements in and out of Pearl. The ships were easily observable from various highway and street sites, from Aiea Heights, from Kamehameha Highway, between Aiea and Makalapa, from a pier at Pearl City, northwest of the harbor, and from a telescope-equipped teahouse on Alewa Heights that overlooked both the harbor and Hickam Field. Driving with Mikami along the Kokokahi Road on the east coast of Oahu, he also studied the Kaneohe Naval Air Station, home of the patrol bomber (PBY) wing. Twice he and Mikami drove by the Army’s Schofield Barracks, and once, in autumn, he took a geisha friend on a tourist airplane flight over southern Oahu that provided excellent views of Pearl and Hickam as well as of the airfields at Wheeler and Ewa. Everything he saw—ships, airstrips, aircraft, hangars—he reported regularly and in detail to Tokyo, utilizing the consulate code room and radio facilities.49 It was his information, for example, that produced the ship movements signal that reached Akagi on 2 December.
On 24 September, Foreign Minister Togo, acting on the request of the Naval Staff’s Third Bureau (Intelligence), directed the Honolulu consulate to generate a second series of reports, these to focus not on ship movements but on berths and anchorages of major warships in port. The “strictly secret” Message No. 83 read:
1. The waters [of Pearl Harbor] are to be divided roughly into five sub areas. (We have no objection to your abbreviating as much as you like.)
Area A. Waters between Ford Island and the arsenal [navy yard].
Area B. Waters adjacent to the island south and west of Ford Island. (This area is on the opposite side of the island from Area A.)
Area C. East Loch.
Area D. Middle Loch.
Area E. West Loch and the communicating water routes.
2. With regard to the warships and aircraft carriers, we would like to have you report on those at anchor (these are not so important), tied up at wharves, buoys, and docks. (Designate types and classes briefly. If possible we should like to have you make mention of the fact when there are two or more vessels alongside at the same wharf.)50
After consulting with Yoshikawa, Consul General Kita responded to No. 83 on the twenty-ninth. His message proposed an even more precise reporting system utilizing code designators:
1. Repair dock in the Navy Yard (the repair base referred to in my message to Washington #48): KS.
2. Navy dock in the Navy Yard (Ten Ten Pier): KT.
3. Moorings in the vicinity of Ford Island: FV.
4. Alongside at Ford Island: FG (east and west sides will be designated A and B respectively).
On 18 November another area, N, was added to the grid.51
The effect of these five designators was to place what has been called an “invisible grid” over Pearl Harbor. Togo’s and Kita’s messages, which were sent in a low-grade consular cipher known as J-19 (but also under the umbrella of Magic), were intercepted by U.S. Navy operators in Station Hypo at Pearl Harbor, named after the “H” of Heeia on the east side of Oahu where the Pacific Fleet’s radio intercept towers were located. But Commander Rochefort, chief of the Combat Intelligence Unit of the Fourteenth Naval District, was under orders from Main Navy to send all consular intercepts, unread, by Clipper pouch to Washington, for decryption and evaluation there. This meant, said Kimmel’s intelligence officer Layton, that “the Pacific Fleet, the principal instrument of our military power in the Pacific, was not equipped to monitor the enemy beyond its harbor wall.”52 In Washington, where consular traffic was given a low priority, Togo’s message of 24 September was not decrypted (by the Army) until 9 October; and Kita’s scheme was not read (by the Navy) until the following day. Colonel Rufus C. Bratton, chief of the Far Eastern Section in Army G-2, routed Togo’s message to Stimson, Marshall, and Gerow, chief of War Plans, but none found it of special interest, much less alarming. On the Navy side, Lt. Comdr. Alwin D. Kramer, of ONI’s Translation Section, Communications Division (OP-20-GZ), routed Togo’s and Kita’s messages through his own hierarchy of Knox, Stark, and Turner, eliciting the same indifference.
There were several reasons why the messages received scant attention. Some midlevel Navy officers thought that the grid system represented an attempt to reduce either the volume or the expense of radio traffic; or that it was a plan for sabotage or for a submarine attack. Others were more impressed by the fact that “ships in harbor” messages had been requested of Japanese embassies and consulates elsewhere: during 1941 there would be six such messages from Seattle, Washington, eighteen from Panama, fifty-five from Manila, and sixty-eight from Honolulu. That the greatest number originated from Honolulu was explained for some by the presence there of the greatest concentration anywhere of U.S. naval strength. No one seems to have thought on the basis of those messages that Pearl Harbor stood out as especially pinpointed for hostile action—except that Honolulu alone was asked to divide harbor waters into precise sectors and to provide such detail as ships moored alongside each other—though one or more unnamed Navy officers did propose to Kramer that the grid “might be a plan for an air attack.”53 In any event, said Kramer, he was under the “impression” that the grid messages were sent to Hart and the Asiatic Fleet with either “an action addressee or information addressee to Admiral Kimmel.”54 The Navy file copy stamp shows it went to Hart but not to Kimmel. Nor was Kimmel sent any of Yoshikawa’s grid reports that followed.
Since Hart was equipped with the Purple decryption machine and regularly read the Magic messages that his cryptanalysts (at Cast Station, located at Cavite Navy Yard, outside Manila) generated, it is ironic in retrospect that Main Navy found it necessary to send him duplicates of what he already had, while withholding the same messages from Kimmel, whose fleet was the grid’s objective. And, grid in hand, Manila relayed no secondary warning to Pearl. Of course, there is no certainty that a Purple-equipped Pearl Harbor would have performed any better than Main Navy or Manila in separating out the relevant warning signals from the background “noise” of competing, irrelevant signals, except that what came to be known as the Pearl Harbor “bomb plot” messages presumably would have made more of an impression on Kimmel’s staff than they apparently made on Main Navy or M
anila; and it might be noted that the intelligence track record of Rochefort and Layton in predicting Midway as Nagumo’s target on the 4 June following proved to be far more discerning and accurate than the record of their counterparts in Washington who insisted otherwise, both as regards target and date.
When, several years later, Layton was shown the bomb plot messages, he pronounced himself “astonished and outraged.” The failure of OpNav to send them to Pearl he attributed to “blind stupidity at the least and gross neglect at best.”55 When Short saw them in 1944, he called the messages “a bombing plan for Pearl Harbor.”56 And Kimmel, who was the most directly affected, and who fought hard and successfully to have all Magic decrypts from 1941 made available to the NCI in 1944, forcefully characterized the Navy Department’s dereliction in failing to send him the messages as “affirmative misrepresentation,” in that he had been promised all intelligence relating to the fleet, and his estimates and actions had been based on what he had received. Now the Magic decrypts disclosed that he had been misled:
These Japanese instructions and reports [withheld from me] pointed to an attack by Japan upon the ships in Pearl Harbor. The information sought and obtained, with such painstaking detail had no other conceivable usefulness from a military standpoint.… Its effective value was lost completely when the ships left their reported berthings in Pearl Harbor.… No one had a greater right than I to know that Japan had carved up Pearl Harbor.… Knowledge of these intercepted Japanese dispatches would have radically changed the estimate of the situation made by me and my staff [and] afforded an opportunity to ambush the striking force as it ventured to Hawaii.57
The senior command personnel in the War and Navy Departments were asked about the bomb plot messages in the JCC hearings. Marshall answered that he had no recollection of seeing them.58 Gerow stated that he had seen the messages but was unable to explain why their contents had not been communicated to Kimmel and Short. The reason was that Marshall had ordered that not even the “gist” of Magic decrypts should be distributed outside of Washington to the theaters. But when asked, “Were you aware that information and enemy intelligence was [sic] being withheld from G-2 in Hawaii?” Marshall answered, “I was not aware of that, sir.”59 Miles also saw the message of 24 September but averred that it was not his responsibility to evaluate it because it pertained to naval matters, best left to ONI.60 On the Navy side, Turner testified that he never saw the messages, but if they should have been sent to Kimmel it was Wilkinson’s job to do so.61
Wilkinson remembered seeing the messages, and spoke at length about them. “The Japanese for many years had the reputation,” he testified, “and the facts bore out that reputation, of being meticulous seekers for every scrap of information.… I thought that [the grid] was an evidence of their nicety of intelligence.” He thought at the time that the information sought was for the purpose of projecting how long a time it would take individual ships to sortie from harbor. The inside vessels at double-banked berths, for example, “might take some time to go out.” He said that he discussed the 24 September message with McCollum and “possibly” with Turner or Ingersoll. In a later appearance before the JCC he allowed that a day-by-day mapping of ship locations according to five sectors would mean more to CINCPAC than to anyone in Washington who was “less immediately concerned.” But, as ONI director, he was forbidden to send Kimmel any decrypt information of an operational nature.62
From the department’s top echelon, Ingersoll could not recall seeing the bomb plot messages, but upon learning about them he thought that Kimmel should have been informed.63 His chief, Stark, thought the same, in marked contrast to Marshall’s strict Army proscription against sending even the gist of Magic information to the theaters. Though he had not seen the 24 September message at the time it was decrypted and translated, and, while he thought that those who did see it might have considered it “as just another example of [the Japanese’s] great attention to detail,”
in the light of hindsight it stands out very clearly, with what we can read into it now [January 1946], as indicating the possibility or at least the groundwork for a Japanese air raid on Pearl Harbor.… This particular dispatch would have been of particular interest, if they [ONI] had so considered it, to Admiral Kimmel. They could have simply sent it out as it was. If they had thought it vital, they could have also brought it to what we call the front office; that is, to Ingersoll or myself, or come through Turner, but I have no recollection of this dispatch having been discussed, certainly not with regard to what in the light of hindsight we would now read into it.64
An Army Chief of Staff orders that no operational intelligence drawn from Magic be sent to his menaced commander in Hawaii, then later states that he was unaware that enemy intelligence was denied him.… An Army intelligence chief, representing the service specifically charged with defending the fleet at Pearl, punts on the grounds that fleet ships, after all, belong to the Navy.… A Navy war plans chief states that any transmission of operational intelligence of this kind should have been sent out by ONI, something he himself never permitted to happen.… A director of naval intelligence discerns in bomb plot messages no more than Japanese curiosity and “nicety” of detail about the time required for ships to sortie from harbor.… And a CNO, as uninformed at the time on this espionage as was the Army Chief of Staff, states four years later that ONI should have sent the information to Kimmel—in direct violation of restraints that his own OpNav office had placed on ONI.… Surely, if ever there was a “fog of pre-war,” it hung low over Washington in the fall of ’41.
The Tokyo-Honolulu traffic reached a suggestive crescendo with three messages outside the usual ship movements and bomb plot categories. The first from Tokyo, dated 2 December, and sent in the J-19 code, read in part:
Wire me in each case whether or not there are any observation balloons above Pearl Harbor or if there are any indications that they will be sent up. Also advise whether or not the warships are provided with anti-torpedo nets.65
The balloons the message had in mind were what were called in the West “barrage” balloons. Tethered to ground moorings, they trailed beneath them steel cables that could cripple low-flying aircraft. Such balloons had been used by the British in defending cities and harbors against the German Luftwaffe. Mention of antitorpedo nets particularly evokes images of a Taranto-type air attack on a fleet at anchor. It stretches credulity to think that Washington would not pick up on these straightforward clues, but we shall never know what the reaction there was, since the message, intercepted by the Army’s communication centers at Fort Shafter, was sent unread by surface mail to Washington, where it arrived on the twenty-third.
The Honolulu consulate duly investigated the subjects cited and replied in two separate messages sent on the same day, 6 December (Hawaiian time). In the first, No. 253, the consulate reported that, though the Army had begun training barrage balloon troops in North Carolina, and had ordered 400 or 500 balloons, at Pearl Harbor they had not as yet set up mooring equipment:
At the present time there are no signs of barrage balloon equipment. In addition, it is difficult to imagine that they have actually any. However … because they must control the air over the water and land runways of the airports in the vicinity of Pearl Harbor, Hickam, Ford and Ewa, there are limits to the balloon defense of Pearl Harbor. I imagine that in all probability there is considerable opportunity left to take advantage for a surprise attack against these places [emphasis added].
The message concluded with: “In my opinion the battleships do not have torpedo nets.”66
This extraordinary document might well indeed have altered history if its contents had been read in a timely way by the sometimes obtuse Washington evaluators, for whom the Pearl Harbor grid portended nothing, but who, we may assume, could hardly have missed the overt meaning of “surprise attack.” (How Yoshikawa and consul general Kita learned or surmised that this was their country’s intent has not been explained.) And one may state the same for t
he second message of 6 December, No. 254, which reported: “It appears that no air reconnaissance is being conducted by the fleet air arm.”67 Unfortunately for Admiral Kimmel and General Short, who may not have been advised of the contents of these dispatches even if decrypted and evaluated in a timely manner, both messages, sent in a very low-grade PA-K2 cipher—the consulate’s J-19 having been burned—fell victim to Washington’s strong preference for Purple messages. Though captured by the Army intercept station at San Francisco at 12:58 P.M. and 6:01 P.M., respectively, and forwarded to the War Department by fast teletype, the two messages were not decrypted until Monday the eighth. Commander Safford said afterward that message No. 253 could have been decrypted in about an hour and a half, and No. 254 in less than an hour.68 But both were delayed in surfacing to view by Washington’s denigration of ciphers employed by mere consulates.69
* * *
There were other provocative messages picked up by intercept stations operated by the two services.70 Particularly ominous where imminence of war in the Pacific (though not necessarily at Pearl Harbor) was concerned was a series of orders sent by Tokyo beginning on 30 November (Washington time) directing embassies to destroy certain codes and ciphers, and, in some cases—the first were London, Singapore, Hong Kong, and Manila—their cipher machines. On 2 December the embassy in Washington was ordered to burn various ciphers, codes, message files, and secret documents. In the same order Washington was also told to “destroy completely” one of its two Purple machines.71 Similar orders were sent to embassies or consulates in Ottawa, Havana, Panama, Los Angeles, Portland, Seattle, Vancouver, and Honolulu. On 5 December the Washington embassy acknowledged the destruction order, but requested from Tokyo “your approval of our desire to delay for a while yet the destruction of one [cipher] machine” for use in the “continuing” negotiations. Tokyo confirmed this understanding of its order on the following day.72
Pearl Harbor Betrayed Page 24
